Document x1r7qVa8MzxZELd7VbvjdqamQ

1 and lobucco on risk of ogical evidence on the a population. Environ. Hal cancer. In "Persons tic Press, New York, moking and neoplasia. garette smoking among 'logical protection. Ra- i ENVIRONMENTAL KLSL.ARCH 38, 319-331 <1985* Metsovo Lung: Pleural Calcification and Restrictive Lung Function in Northwestern Greece. Environmental Exposure to Mineral Fiber as Etiology1 S. H. CONSTANTOPOULOS.- J. A. GOUDEVENOS. N. SARATZIS. A. M. Langer.* I. J. Selikoff.* and H. M. Moutsopoulos Department of Medicine. School of Medit ine. University of lottnnintt. loannina. Greece, and *Environmental Sciences Laboratory. Mount Sinai School of Medicine. New York. New York 10029 Received August 20. 1984 Pleural calcifications are described in 122 of 268 (45.5%) inhabitants of four villages (Metsovo. Anilio. Milea. and Votonosi) in a small area of northwestern Greece (total pop ulation about 5000). All affected individuals are of one ethnic group. Vlachi. Calcifications were not noted in any of the 103 persons in the control group made up of 73 non-Vlachi inhabitants from the same and neighboring villages and 30 Vlachi from distant villages. The calcifications were seen in both sexes, equally, and their frequenty increased with age. from 28.6% between 30 and 39 years to 81.0% in individuals over 70 years of age. When plaque development was extensive, a small restrictive pulmonary function defect was noted. Be cause of its prevalence in the Metsovo area we call this clinical pattern Metsovo lung. The identification of tremolite. related amphiboie fibers, and traces of chrysolite fiber in settled dusts and soil specimens and of identical fibers in tissue specimens obtained at lung biopsy from 8 people with plaques supports the hypothesis that abestiform minerals are the agents responsible for these disease processes. Further, reports of the occurrence of mesothelioma and benign pleural efTusions in inhabitants in the Metsovo area, along with the striking similarities to disease patterns observed in the Karain area of Turkey, add further weight to the hypothesis that mineral fiberfs) in the environment of the four villages are agent(s) in the etiology of Metsovo lung, c ms Academic pInc. INTRODUCTION Diffuse bilateral pleural calcification frequently occurs in individuals exposed to asbestos (Selikoff, 1965). Although such exposures are most often occupational in nature, there have been reports of endemic pleural calcifications from Finland (Kiviluoto, I960), Bulgaria (Burilkov and Michailova, 1970), Czechoslovakia (Navratil and Trippe, 1972), and southwest Turkey (Yazicioglu, 1976) in clearly nonoccupational settings. These scattered foci of pleural disease appear to be unrelated to the direct mining, milling, fabrication, or commercial use of asbestos. In all of the forementioned occurrences, plaques have been attributed to envi ronmental exposure to chrysotile in Czechoslovakia, to anthophyllite + tremolite + sepiolite in Bulgaria, to anthophyllite in Finland, and to tremolite and chry sotile in Turkey. 1 This study was supported in part from a grant from the National Institute of Environmental Health Sciences. ES 00928. 2 To whom reprint requests should be addressed. 319 0013-9351/85 *3.00 Copyright 1915 by Academic Press, Inc. AD rights of reproduction in any form reserved. 320 CONSTANTOPOULOS ET AL. In South Africa and central Turkey, pleural changes (thickening in South Africa, thickening and plaque development in Turkey) have been associated with malig nant mesothelioma as well (Wagner et al., 1960; Artvinli and Baris. 1979). These changes occur in individuals without history of employment involving asbestos but rather are linked to environmental exposure to crocidolite in South Africa and erionite in Tbrkey. A recent report suggests that asbestos fiber may also play some role in the etiology of the diseases observed in central Turkey (Rohl et al., 1982). In 1969, during the antituberculosis campaign in Greece, it was found that the inhabitants of three neighboring villages of northwestern Greece. Metsovo, Anilio, and Milea, where there are no known asbestos outcroppings, mining, or any heavy industry, showed a high incidence of pleural calcification. This was re evaluated in 1980. Bazas et al. (1981) reported that 28% of a selected subset of this population had signs of pleura) calcification. This latter report, while high lighting these findings, offered no explanation for these occurrences although direct asbestos exposure (especially occupational) was ruled out as a factor in the etiology. We now report clinical, roentgenological, and functional profiles of the disease we call ``Metsovo lung." On the basis of analytical electron microscopic analysis of eight biopsies obtained at bronchoscopy, mineral fibers, some consistent with asbestiform tremolite, were present in the pulmonary tissues of these individuals. It is likely that this exposure is attributable to environmental exposure, with the fibers originating from local soils and rock outcroppings. This report may now be added to those which underscore the importance of mineral fiber exposure in a nonoccupational setting in the development of pleural disease. BACKGROUND Geological Setting The villages of Metsovo, Anilio, and Milea are located within the Pindos ophiolite complex of Eocene age (--60 million years ago), in northwest Greece (Fig. 1). The region is mountainous, rugged in terrain, extensively faulted and folded, and represents the southern extremity of the European Alps. Metsovo and Anilio are situated approximately 2 km to the east of a regional fault trace that represents the remnants of a tectonic plate along which suboceanic litho sphere and upper mantle-derived peridotites, dunites, pyroxenites, and other mafic rock types have been thrust into contact with crustal schists and igneous rocks. The mafics and ultramafics have been extensively serpentinized (Institute for Geology and Subsurface Research, 1959; Montigny et al.. 1973). Local drainage flows from the mountainous serpentinized highland (on the north east), southward toward Metsovo and Anilio which lie on schists and granulites west of the serpentine body. Milea, some 15 km to the north of Metsovo, is located within the periodotite-serpentine complex. Many other ultramafic units occur within the area as well, e.g., olivine-gabbros. These are also extensively serpentinized and often occur intercalated within the ophiolitic melange (Mon tigny et al.. 1973). 2HES TSA Fig. I. Location of stud shaded zone represents th Details of the boxed area Major villages are indicate* noted; circles with centers represent "distant" village The area has under files reflect unsorted . mineral components crushed and pulverize particles. These are a Although serpentim asbestos mining or kr bestos mine is locate, outcroppings of asbe exploited commercial show pleural calcifica scribed in the present Regional Rock Types Serpentine rock is * Fibrous in form, chry commerce. Serpcnlini. number of amphibole i naturally asbestiform i into acicular cleavage The amphiboles des aluminum-rich (high-p; 10003116 nc in South Africa, dated with malig-aris, 1979). These nvolving asbestos e in South Africa iber may also play - rkey (Rohl el al., 'as found that the Greece. Metsovo, ~pings, mining, or -:ion. This was re-elected subset of :port, while high-rrences although r>ut as a factor in 'les of the disease roscopic analysis e consistent with these individuals, vposure, with the report may now fiber exposure in ithin the Pindos orthwest Greece vely faulted and i Alps. Metsovo cional fault trace suboceanic lithonites, and other ists and igneous tinized (Institute m). nd (on the north-,s and granulites of Metsovo, is ultramafic units also extensively melange (Mon- METSOVO LUNG 321 XONfTSA Kflovrio 0 votyo uefsovo ^ An-li0 P%ritt*ri AMhOCt0*J B ioSWAa Sit* ` Fig. 1. Location of study area in Greece. (A) The Metsovo area shown on a map of Greece. The shaded zone represents the Pindos serpentine belt. For reference, the city of loannina is marked. Details of the boxed area are shown in (B). Villages in study are shown in relation to each other. Major villages are indicated by circle size. Solid circles represent villages where calcifications were noted; circles with centers represent villages where calcifications have not been found; solid triangles represent "distant" villages where no calcifications were found in Vlachi inhabitants. The area has undergone Pleistocene alpine-type glaciation so that the soil pro files reflect unsorted and unconsolidated debris. The soils in the area also reflect mineral components from the underlying rock types. The forces of glaciation crushed and pulverized bedrock in places producing submicrometer-size mineral particles. These are also components within the soil profile. Although serpentine bodies are hosts for chrysotile concentration, there is no asbestos mining or known outcroppings in the immediate area. The nearest as bestos mine is located some 80 km northeast (near Kozani). There are known outcroppings of asbestos some 30-40 km north of this area, yet they are not exploited commercially. The inhabitants of villages near these deposits do not show pleural calcification in the manner or extent exhibited by the villagers de scribed in the present study. Regional Rock Types as Sources of Mineral Fiber Serpentine rock is composed of a number of minerals, including chrysotile. Fibrous in form, chrysotile represents the most common asbestos fiber in world commerce. Serpentine rock, and the associated ultramafics, may also contain a number of amphibole minerals, many of which may also be fibrous in form (either naturally asbestiform or "fibrous" by processes of comminution (able to reduce into acicular cleavage fragments). The amphiboles described from this area include tremolite, anthophyllite, and aluminum-rich (high-pressure) phases, as well as common hornblende. Both ser- 10003117 r-% 322 CONSTANTOPOULOS ET AL. pentinized and unserpentinized rock types may act as a source of mineral fiber for the environment (Rohl et al., 1977; Puffer el al., 1983). The Populations Affected Of the villages included in this study, Metsovo is the principal one with a population of some 3300 people. Anilio, Milea, and Votonosi are smaller in size with populations of between 1000 and 1500. The principal occupations among the men are shepherding, woodcutting, and carpentry. The women are occupied with housekeeping chores as well as handcrafting, e.g., wool knitting. Although Greek, these inhabitants belong to a different ethnic subgroup (Vlachi) from their closest neighbors. MATERIALS AND METHODS Mineral dust analyses. Soil and settled dust specimens were examined by po larized light microscopy in the standard manner utilizing immersion oils of known indices of refraction. The specimens were further examined by continuous scan X-ray diffraction following the standard protocol for our laboratory (Environ mental Sciences Laboratory, Mount Sinai School of Medicine). Thirty-seven major reflections were characterized in the scan range and those with intensities 3x greater than background were assigned a (/-spacing value. Analysis of these specimens was also carried out with a JEOL 100 CX analytical electron micro scope (AEM) interfaced with a Tracor Northern TN 2000 Energy Dispersive XRay Spectrometry System (EDX). Water-fractionated specimen splits were pi petted onto carbon-coated, Formvar-covered, TEM nickel locator grids. These were allowed to evaporate to dryness, recoated with carbon, and examined by AEM. Morphological, structural, and chemical information was obtained as pho tographic records and selected area electron diffraction patterns and by chemical probing of single particles. Tissue analyses. We studied nine specimens of lung tissue biopsies, obtained at fiberoptic bronchoscopy, from inhabitants of Metsovo who suffered with "Metsovo lung." These specimens, averaging approximately 4 mm3 in volume, were withdrawn from their Formalin containers and placed in a 5% KOH solution, set in a hot water bath maintained at 80C, and allowed to digest. Several cycles of water washing were required for complete cleaning of digestates. Residues were dispersed by sonication for approximately 3 sec and micropipetted onto appropriate TEM substrates for examination by AEM. The analytical technique has been previously described (Langerand McCaughey, 1982). Of the nine, only eight specimens were analyzed, as no residues were recovered from one spec imen. The inorganic residues recovered were scant. Clinical data. We have examined chest roentgenograms of 246 inhabitants of Metsovo, Anilio, and Milea (Table I) and 125 inhabitants from two "control" areas. The first consisted of 95 individuals from the neighboring villages of Votonosi, Anthochori, Megalo Peristeri, and Micro Peristeri; the second group was made up of 30 Vlachi persons from other, more distant villages. Twenty-two inhabitants of Votonosi, who were also Vlachi. were eventually grouped together Pwith Metsovo, Anilu tlhble 1). {The chest X rays e General Hospital (27 voluntarily took pan he clinical data are |of 98 persons from N , complete clinical t [the British MRC qc Aetiology of Chron : tuberculosis; a comp tion data which incl lowing the standards 98. Additionally, w< location and extent ( The spirometric i volume in 1 sec, FE 1971). Mineral Studies Representative sp< Metsovo village wer uous scan X-ray difi ination of the materi Geographical Distrib Nu Village Votonosi Total Total excluding Votonosi 10003118 i jrcc of mineral fiber principal one with a >si are smaller in size ;cupations among the .en are occupied with ing. Although Greek. :hi> from their closest ere examined by po"ersion oils of known i by continuous scan laboratory (Environicine). Thirty-seven hose with intensities je. Analysis of these :ical electron micro- nergy Dispersive Ximen splits were pilocator grids. These n. and examined by a as obtained as pho-ms and by chemical e biopsies, obtained who suffered with y 4 mm3 in volume, a 5% KOH solution, gest. Several cycles iigestates. Residues micropipetted onto analytical technique h. Of the nine, only . red from one spec- f 246 inhabitants of rom two "control" hboring villages of : the second group iilages. Twenty-two *> grouped together METSOVO LL'NG 323 with Metsovo, Anilio, and Milca inhabitants since calcifications were found (see Table I). The chest X rays either were obtained from the medical records of the lonannia General Hospital (273 persons) or were taken during examination of those who voluntarily took part in the survey (98 persons). For the first group of 273 persons, the clinical data are limited to the X-ray films. However, from the second group of 98 persons from Metsovo, Milea, Anilio, and Votonosi, we have the following: a complete clinical evaluation, including a respiratory questionnaire (similar to the British MRC questionnaire (Medical Research Council Committee on the Aetiology of Chronic Bronchitis, I960)), with additional questions relating to tuberculosis; a complete physical examination: chest X ray; and pulmonary func tion data which includes spirometry with a 5-liter Stead-Wells spirometer fol lowing the standards of the American Thoracic Society (Ferris, 1978) in 55 of the 98. Additionally, we have graded the pleural calcifications according to their location and extent (Fig. 2). These grades include 0, 1, II, III, IV. The spirometric indices (forced vital capacity, FVC, and forced expiratory volume in 1 sec, FEV,) are expressed as percentages of normal (Morris et at., 1971). RESULTS Mineral Studies Representative specimens of soil and settled dust obtained in houses within the Metsovo village were analyzed by polarized light microscopy (PLM) and contin uous scan X-ray diffraction (XRD). Fine splits were examined by AEM. Exam ination of the materials by PLM showed the presence of quartz, carbonate min- TABLE 1 Geographical Distribution of Populations Studied and Occurrence of Pleural Plaques Village Number of X rays read Cases Sex A Group A (-Plaque) Group B (+Plaque) Plaque (%) Milea Anilio Metsovo Votonosi 23 29 . 194 22 M 12 F 11 M 17 F 12 M 108 F 86 M 12 F 10 5 5 9 7 57 47 7 9 7 56.5 6 8 5 44.8 51 39 46.4 5 27.3 1 Total Total excluding Votonosi 268 246 M 149 F 119 M 137 F (09 146 130 122 45.4 116 47.2 324 CONSTANTOPOULOS ET AL. O. ^ >'A Fig. 2. Grading ofpleural plaques according to location and extent. No plaques = 0; small meniscus on diaphragm or single irregular node = 1; bilateral meniscus or two irregular nodes, in one or both lungs = II; multiple lesions, not exceeding one third of lung fields = III: multiple lesions, exceeding one third of lung fields, tending to coalesce - IV. erals, and feldspars. Plagiociase predominated in the feldspar population. Micas were also observed, as were limonite minerals and clays (with some possible talc). No mineral fibers of any kind were observed, although one fibrous opaline phytolith was noted. The soil from Metsovo, analyzed by XRD, was composed of quartz, feldspar (labradorite, andesine, and oligoclase), calcite, and dolomite as the major minerals, and mica (both biotite and phlogopite), chlorite-vermiculite (possibly some serpentine minerals, as evidence by a split in the 7 reflection region), montmorillonite clay, and, possibly, talc. Occasional chrysolile fibrils were observed by transmission electron microscopy. Trace amounts of amphibole, which appeared to be acicular cleavage fragments, and other silicates were ob served as well. Some rare fibers (only two) appeared to display closely spaced polysynthetic twins coplanar with (100). These fibers displayed selected area electron diffraction patterns consistent with an amphibole structure in that ori entation. Additionally, these fibers displayed the general chemistry of iron-poor tremolite (with measurable concentrations of aluminum). Tissue Burden Studies Although materials obtained at transbronchial biopsy represented only some 1/250 of the mass of tissue usually prepared for similar dust burden studies, fibers were found in five of the eight preparations. Because of the nature of the tissue specimens, e.g., variation in mass and size of tissue and different tissue biopsied, quantitation was not attempted. The fibers3 observed in the pulmonary tissues of patients with "Metsovo lung" 5 The term fiber is used in the morphological sense. s t. \\t 0.000 50 Fig. 3. Diagnostic criteria diameter (fiber is --0.25 pm pattern (pattern obtained on senling values for an amphih composition. Potassium is a 10003120' U ques = 0; small meniscus -!ar nodes, in one or both ultipJe lesions, exceeding jr population. Micas (with some possible i one fibrous opaline <RD, was composed :alcite, and dolomite e). chlorite-vermiclit in the 7 reflection nal chrysotile fibrils nounts of amphibole, er silicates were ob* splay closely spaced clayed selected area structure in that oriemistry of iron-poor oresented only some 'urden studies, fibers . nature of the tissue ?rent tissue biopsied. , ith "Metsovo lung' METSOVO LL\G 325 have been identified according to the three essential diagnostic criteria: (I) mor phological appearance: (2) structure, by selected area electron diffraction; and (3) chemistry, as determined by energy dispersive X-ray spectrometry (Langer et ai, 1973). Complete analysis of a single fiber is shown in Fig. 3. Fibers recovered from these biopsy materials are elongate, some with aspect ratios (length-to-width) exceeding 48:1. Fiber lengths up to 10 p.m were observed (Fig. 4). These features are characteristic of asbestiform materials; in the present instance, these characteristics rule out fragments of calcic pyroxenes, e.g., diopsidic augite. The fibers tend to display curvilinear form parallel to their c axes (length), another feature common to asbestiform minerals; their edges tend to be parallel, with an average diameter of about 0.20 p.m (Figs. 3 and 4). Contour banding is occasionally observed across the fiber axis, displaced along both twin (100) and Wadsley defect (010) boundary planes. These features, although not confined to the amphibole asbestos minerals, are common to the asbestiform amphiboies. The presence of mineral fragments as well as irregular fibers was noted but no identification was attempted. Where the fibers were thin enough to permit the transmission of the diffracted beam through the crystal, and coating material was not present on the fiber, the SAED pattern was recorded on a photographic plate (Fig. 4). Unfortunately, this set of conditions occurred infrequently. When present, these patterns were re- 0.800 SB METSOUO LUNG GKO 3 \JF> Fig. 3. Diagnostic criteria used in the identification of fiber in tissues: (A) morphology, including diameter (fiber is -0.23 pm diameter) and internal structure: (B) selected area electron diffraction pattern (pattern obtained on Tiber shown in (A) indicates a lattice repeat of 0.33 and 1.8 nm, repre senting values for an amphibole structure: (C) chemistry of fiber shown in (A) indicating a tremolite composition. Potassium is a residue artifact of the KOH reactant. 10003121 Fig. 4. Fibers extracted from biopsy specimens. (A) Fiber -0.25 urn diameter: length:width ratio is greater than 28:1; Fiber composition is of aluminous tremolite. (B) Fiber -0.20 iim in diameter; length:width ratio is greater than 26:1; composition, aluminous tremolite. (C) Fiber --0.15 urn di ameter; length.width greater than 48:1; composition, silicate mineral (obscured by tissue debris; Xray signal contaminated). rded and the reflc b*), which conform Chemical analyse yses obtained on iard tremolites, ele tra. Na:Si, Mg:S berfound in "Met idard tremolite cmagnesium (14- fuggesting possible impositions not qui ch and silica poor, i Ieavage fragment. In qualitative tern eluded both rare chr phibole minerals occ phibole fibers tended uncommon asbestifo in only five of eight tissue available. Plats micas, were commoi Clinical Studies Roentgenographic showed pleural calci! (27.3%) from Votonc distant villages and in the Metsovo area s Vlachi, like the vast only four non-Vlachi most of their life ther calcification was 29 28.6% in the 30- to . ble 2). Age group 30-39 40-49 50-59 60-69 70 + All ages 10003122 METSOVO LUNG 327 corded and the reflections measured. Values of 0.53 nm (c*) and 1.83-2.06 nm (b*), which conform to an amphibole structure, were obtained (Fig. 3). Chemical analyses of seven "clean" fibers were compared with five sets of analyses obtained on tremolite standards (iron rich and iron poor). For the stan dard tremolites, elemental ratios were calculated from background corrected spectra. Na:Si, Mg:Si, Al:Si, Ca:Si, and Fe:Si ratios have been matched with the fiber found in "Metsovo lungs." Several fibers were in good agreement with standard tremolite composition; three fibers were apparently "low" in calcium and magnesium (14-17% Ca:Si, rather than 25 * 3% Ca:Si)and high in aluminum (suggesting possible Al substitution in both X and Y structural sites). Others had compositions not quite similar to the common hornblende, apparently aluminum rich and silica poor, but not sufficiently high in Ca:Si ratio to suggest a pyroxene cleavage fragment. In qualitative terms, the mineral fibers observed within the tissue debris in cluded both rare chrysotile fibrils and several varieties of amphiboles. The am phibole minerals occurred with much greater frequency than chrysotile; the am phibole fibers tended to be tremolitic and morphologically asbestiform. Additional uncommon asbestiform fibers were considered as well. These fibers were found in only five of eight specimens studied perhaps because of the small amounts of tissue available. Platy particles, consistent in chemistry with serpentine, talc, and micas, were common. Roentgenographic profile. Of the persons from Metsovo, Anilio, and Milea 116 showed pleural calcifications on X ray (47.2%) (Table 1). Also, 6 of the 22 persons (27.3%) from Votonosi showed similar signs. None of the 30 Vlachi persons from distant villages and none of the 73 inhabitants of the three neighboring villages in the Metsovo area showed calcification. All 122 persons with calcifications were Vlachi, like the vast majority of the inhabitants of these villages. We examined only four non-Vlachi from these four villages (mean age 41 years) who had spent most of their life there. None had calcifications. The youngest person with pleural calcification was 29 years old. The incidence of calcifications increased from 28.6% in the 30- to 39-year-old group to 81.0% in individuals older than 70 (Ta ble 2). TABLE 2 Incidence of Calcification by Age Plaques (% positive) Fig. 5. Examples of plaque development: (A) Development of plaques as small meniscus on dia phragm (see arrows, stage U). (B) Development of plaques as large coalesed extensive lesions (stage IV). The plaques appear first on the diaphragmatic pleura, or as an irregular node elsewhere that cannot be differentiated from a parenchymal lesion (Fig. Sa). Plaques increase in size and extent until 40-50 years of age, after which time they cover most of the thorax (Figs. 5b, 6). The overall percentage of roentgenograms with calcifications in the three main villages ranged from approximately 42% in women to 58% in men (Table 1). There was fairly equal sex distribution except for the 10 women of Votonosi (mean age 37.9 years) only 10% of whom had abnormal chest roentgenograms instead of approximately 23% "expected" for their age. With such limited experience de finitive explanations concerning this observation cannot be made at this time. Clinical profile. Clinical information was obtained concerning 98 inhabitants of the four villages where calcifications were noted: 37 had negative X rays (group Fic. 6. Plaque development in the same individual, with time: (A) 1971, (B) 19 7. (C) 1983. A) and 61 showed pi in group A and 8 ir account for any patt (56%) had respirato cough, expectoratic asthma, tuberculosis group B was larger (: in terms of age (A, 4 Spirometric profile and 32 from group B B, because of known Therefore, 18 and 29 FVC and FEV, in gn lively. Group B incl fifteen with stage 111. were 87.8 and 95.0% in stage IV, showing with advancing age a Pleural calcificatio small area in northwe frequency, and merer environmental factor ethnicity living in Me ther Vlachi from dist not Vlachi have beer much to suggest imm suggest the possibility phenomenon. In the past, agents were thought resporn area have been treat chemotherapeutic res pothesis" is suspect, either tuberculosis or There are also repi population. These fin etiology of the pleura: Similar patterns have account for the entire tion, malignant mesot observed in the Karai The general pattern patterns reported fron has been implicated (/ 10003*24 s small meniscus on diaUesed extensive lesions as an irregular node lal lesion (Fig. 5a). :e. after which time ns in the three main men (Table 1). There Votonosi (mean age : nograms instead of lited experience denade at this time, ing 98 inhabitants of zative X rays (group Pr>*- (B) 1977. (C) 1983. METSOVO LUNG 329 A) and 61 showed plaque development (group B): there were only 14 smokers (6 in group A and 8 in group B); there was no apparent occupational factor to account for any pattern of respiratory disease; 20 of the 37 persons of group A (56%) had respiratory symptoms or history of respiratory disease (dyspnea, cough, expectoration, wheezing, history of frequent "colds," pneumonia, asthma, tuberculosis, bronchogenic carcinoma); the incidence of symptoms in group B was larger (52/61 = 85%). However, the two groups are not comparable in terms of age (A, 45.0 years; B, 62.0 years) or sex. Spirometric profile. Spirometry was attempted in 55 persons, 23 from group A and 32 from group B (5 persons were excluded from group A and 3 from group B, because of known pulmonary or cardiac problems that interfered with testing). Therefore, 18 and 29 persons, respectively, were evaluated. The mean values of FVC and FEV, in group A were 92.3 and 96.6% of the predicted values, respec tively. Group B included one with stage I roentgenogram, four with stage II, fifteen with stage III, and ten with stage IV. The mean values of FVC and FEV, were 87.8 and 95.0% in stage II, 81.0 and 88.3% in stage III, and 75.5 and 80.5% in stage IV, showing a slight decrease of lung volume (restrictive lung disease) with advancing age and plaque development. DISCUSSION Pleural calcifications are reported here in inhabitants of four villages from a small area in northwestern Greece. These changes occur in both sexes, with equal frequency, and increase in prevalence with age. This pattern suggests a common environmental factor as an etiological agent. As of now, only persons of Vlachi ethnicity living in Metsovo, Anilio, Milea, and Votonosi have been affected. Nei ther Vlachi from distant villages nor inhabitants of these four villages who are not Vlachi have been affected. We stress ethnicity at the present time not so much to suggest immunologically controlled patterns of response, but rather to suggest the possibility of some social or cultural factor which controls this cluster phenomenon. In the past, agents of an infectious nature (e.g., tuberculosis, histoplasmosis) were thought responsible for this disease pattern. Many individuals within this area have been treated in the past few years for these diseases. No positive chemotherapeutic results have been observed so that the "infectious agent hy pothesis" is suspect. Further, the appearance of the X rays is not consistent with either tuberculosis or histoplasmosis. There are also reports of mesothelioma and benign pleural effusions in this population. These findings caused us to focus on mineral fiber as a factor in the etiology of the pleural calcification and restrictive pulmonary function observed. Similar patterns have been observed in other parts of the world and mineral agents account for the entire spectrum of the diseases observed, i.e., pleural calcifica tion, malignant mesothelioma, and pleural effusion. This pattern is similar to that observed in the Karain area of Hirkey. The genera] pattern described in this part of Greece has striking similarities to patterns reported from central Hirkey, where the fibrous zeolite mineral erionite has been implicated (Artvinli and Baris, 1979). In the past 2 years, three patients 10003125 *$'V *r s- a. >-''* ... >V.'i ^j i^/Wi 330 CONSTANTOPOULOS ET AL. from the Metsovo area were hospitalized with clinical, roentgenological, and pathological findings compatible with mesothelioma. Moreover, a number of in habitants of this same area died with mesothelioma in the past several years in other hospitals. All of these patients were Vlachi. Five cases of pleural mesothelioma, in a population of some 3000 people, in 2 years appears to be a much higher annual incidence than expected. Data for the annual incidence of mesothelioma for the area of Greece are currently unavail able. The incidence can be compared with reports in other parts of the world: one case per million people in Pennsylvania (Lieben and Pistawka, 1967); one case per million in Finland (Nurminnen, 1975); two to three cases per million in Great Britain (Greenberg and Davies, 1967). The prevalence of this disease in this area of Greece is strikingly higher (five cases per 3000 in 2 years). There are further anomalies in this pattern. It is of special interest that none of the 5 patients with diagnosed mesothelioma had pleural calcification. Yet, the incidence of previous, unexplained, pleural effusion was significant. Of the 98 persons whose clinical histories were known, 26 had had episodes of benign pleural effusion. This percentage was higher in individuals with pleural plaques than in the control group. The mineralogical studies of the settled dust and tissues suggest mineral fiber as at least one of the agents in the etiology of these diseases. Some of the fibers identified possess the morphological, structural, and chemical characteristics of. fibers with demonstrated biological activity. The following caveat should be men tioned: the analysis of the transbronchial biopsy reflects mineral dust deposited in and proximate to the airways. These mineral fibers may or may not, either qualitatively and quantitatively, reflect the fiber population at the pleura, the site of tissue injury. Several features of this disease pattern require further explanation. The occur rence of pleural plaques in this ethnic group may be related to some common social habit, practice, or "life-style," confined to this particular group, which exposes them to mineral fiber in concentrations considerably different from others living in the same area. The Vlachi people tend to work in the fields, as shepherds, where they are more likely to encounter dust. Whether this is critical to the development of disease remains to be elucidated. The overall pattern of pleural calcification and pleural disease, and the occurrence of what appears to be asbestiform amphiboles and traces of chrysotile in pulmonary tissues, is consistent with the appearance of these changes. A quantitative case-control study of pul monary tissues is warranted for the future. For the present we propose the term "Metsovo lung" to describe this interesting association. ACKNOWLEDGMENT We lhank Ms. E. Papanikolaou for excellent secretarial assistance and Dr. N. Papadopoulos of (he NtH who kindly provided us (A.M.L. and I.J.S.) with soil and settled dust specimens from Metsovo. Additionally, we thank Professor D. R. Bowes, of The University. Glasgow, and Dr. S. G. Khoury, consultant in Engineering Geology, Athens, for their information regarding the local geology of the Pindos thrust belt and the local geology of the Metsovon Quadrangle. . Aftvinli, M., and Baris, I. 0 of TUrkey: An epidemio: Bazas, T., Bazas. B., Kitas. northwest Greece. Lane Burilkov, T., and Michailova Environ. Res. 3, 443-45 Constantopoulos, S. H.. Aci spiratory involvement ir. Exp. Rheumatol. 1, 233Fenrts, B. G. (1978). Epiderr dunes for pulmonary fun, Greenberg, M.. and Davies, 104. Institute for Geology and Sub 1:50.000. Kiviluoto, R. (1960). Pleural thophyllite asbeslosis. A< Langer, A. M., Ashley, R.. i C. J., Nickolson, W. J., fication of Asbestos in H Langer, A. M., and McCaugl 1103. Lieben, J., and Pistawka, H. 14, 559-563. Medical Research Council C questionnaires on respira' Montigny, R.. Bougault, H., I genesis of the Pindos oph ' Morris, J. F., Koski, A., and adults. Amer, Rev. Rtspii Navratil, M.. and Trippe, F. (1 dust and in the general pc Nurminnen, M. (1975). The ep exposure. Scand. J. Wort Puffer, J. H., Marcsea, G. P.,; New Jersey Area: Sourc Center for Coastal and En Project A-067-N.J., Contr. Rohl, A. N.. Langer, A. M.. . use of quarried serpentine Rohl, A. N., Langer, A. M.. .` disease associated with ex 520. Selikoff. I. J. (1965). The occu N.Y. Acad. Sci. 122, 351Wagner, J. C., Sleggs, C. A.. exposure in the northwest Yazicioglu. S. (1976). Pleural c; Turkey. Chest 70, 43-47. Yazicioglu, S., llcayto, R., B; pleural mesotheliomas and r 10003126 'cnlgenological, and er, a number of in>ast several years in le 5000 people, in 2 neeted. Data for the c currently unavailparts of the world: istawka, 1967); one cases per million in e of this disease in n 2 years), il interest that none deification. Yet, the gnificant. Of the 98 episodes of benign vith pleural plaques iggest mineral fiber . Some of the fibers al characteristics of veat should be menicral dust deposited or may not, either the pleura, the site anation. The occurJ to some common cular group, which iifferent from others ields, as shepherds, is is critical to the II pattern of pleural it appears to be as- ssues, is consistent jntrol study of pule propose the term N. Papadopoulos of the >ecimens from Metsovo.1 . and Dr. S. G. Khoury, the local geology of the METSOVO LUNG 331 REFERENCES Artvinli. M., and Baris. I. (1979). Malignant mesotheliomas in a small village in the Anatolian Region of Turkey: An epidemiologic study. J. Nall. Cancer Inst. 63(1), 17-20. Bazas, T., Bazas, B., Kitas. D.. Gilson, J. C., and McDonald, J. C. (1981). Pleural calcification in northwest Greece. Lancet. Burilkov, T., and Michailova. L. (1970). Asbestos content of the soil and endemic pleural asbeslosis. Environ. Res. 3, 443-451. Constantopoulos, S. H., Acritidis, N. K., Dalavanga. Y. A., and Moutsopoulos, H. M. (1983). Re spiratory involvement in autoimmune rheumatic diseases is frequent and easy to detect. Clin. Exp. Rheumatol. 1, 233-236. Ferris, B. G. (1978). Epidemiology standardization project: 111. Recommended standardized proce dures for pulmonary function tests. Amer. Rev. Respir. Dis. 88, 55 - 88. Greenberg, M., and Davies, T. A. (1967). Mesothelioma register 1967-68. Brit. J. Ind. Med. 31, 91104. Institute for Geology and Subsurface Research (1959). Geological Map of Metsovon Quadrangle. Scale 1:50,000. Kiviluoto. R. (I960). Pleural calcification as a roentgenologic sign of non-occupational endemic anthophyllite asbeslosis. Acta Radiol. Suppl. 194. Langer, A. M., Ashley, R., Barden, V., Berkley, C., Hammond, E. C., Mackler, A. D.. Maggiore, C. J., Nickolson, W. J.. Rohl, A. N., Rubin, I. B., Sastre, A., and SelikofT, 1. J. (1973). Identi fication of Asbestos in Human Tissues. J. Occup. Med. 15(3), 287-295. Langer, A. M., and McCaughey, W. E. (1982). Mesothelioma in a brake repair worker. Lancet 11011103. Lieben. J.. and Pistawka, H. (1967). Mesothelioma and asbestos exposure. Arch. Environ. Health 14, 559-563. Medical Research Council Committee on the Aetiology of Chronic Bronchitis. (I960). Standardized questionnaires on respiratory symptoms. Brit. Med. J. 2, 1665. Montigny. R., Bougault, H., Bottinga, Y.. and Allegre, C. J. (1973). Trace element geochemistry and genesis of the Pindos ophiolite suite. Geochim. Cosmochim. Acta 371, 2135-2147. Morns, 3. F., Koski, A., and Johnson, L. C. (1971). Spirometric standards for healthy non-smoking adults. Amer. Rev. Respir. Dis. 103, 57-68. Navratil, M., and Trippe, F. (1972). Prevalence of pleural calcification in persons exposed to asbestos dust and in the general population in the same district. Environ. Res. 5, 210-216. Nurminnen, M. (1975). The epidemiologic relationship between the pleural mesothelioma and asbestos exposure. Scand. J. Work Environ. Health 1, 238-137. Puffer, J. H., Maresea, G. P., and Germaine, M. (1983). "Asbestos in Water Supplies of the Northern New Jersey Area: Source, Concentration, Mineralogy, and Size Distribution," Report to the Center for Coastal and Environmental Studies, Div. Water Resources, U.S. Dept, of the Interior, Project A-067-N.J., Contract I4-34-000I-2I32. U.S. Dept, of the Interior, Washington, D.C. Rohl, A. N., Langer, A. M., and Sclikoff, I. J. (1977). Environmental asbestos pollution related to use of quarried serpentine rock. Science (Washington, D.C.) 196, 1319-1322. Rohl, A. N., Langer, A. M., Moncure, G. SelikofT, I. J., and Fischbcin, A. (1982). Endemic pleural disease associated with exposure to mixed fibrous dusts. Science (Washington, D.C.) 216, 518520. SelikofT, I. J. (1965). The occurrence of pleural calcification among asbestos insulation workers. Ann. N.Y. Acad. Sci. 122, 351-367. Wagner, J. C,, Sleggs, C. A., and Marchand, P. (1960). Diffuse pleural mesothelioma and asbestos exposure in the northwest Cape Province. Brit. J. Ind. Med. 17, 260-271. Yazicioglu, S. (1976). Pleural calcification associated with exposure to chrysolite asbestos in southeast Tbrkey. Chest 70, 43-47. Yazicioglu, S., Ilcayto, R., Baici, K., Sayli, B. S., and Yorulmaz, B. (1980). Pleural calcification, pleurai mesotheliomas and bronchial cancers caused by tremolite dust. Thorax 35, 564-569.