ENVIRONMENTAL

RESEARCH

31, 189-200 (1983)

Nonasbestos

Pulmonary Mineral General Population’

Fibers

in the

ANDREW CHURG~ Departments Vancouver,

of Pathology, British Columbia, San Francisco,

University of British Columbia, and University of California, California 94143

Received May 6, 1982 Total pulmonary nonasbestos mineral content was determined for a series of 20 patients who had no occupational dust exposure. To extract mineral fibers, lung tissue was dissolved in bleach and the treated sediment transferred to the electron microscope grid. Mineral fibers were identified using electron diffraction and energy dispersive x-ray spectroscopy. The mean number of nonasbestos fibers, 106 x 103/g wet lung, was almost identical to the mean number of asbestos fibers, 102 x 103/g. Thirteen different species or groups of nonasbestos minerals were found: apatite accounted for 18% of the total and talc for 16%. All other forms accounted for less than 8% each. Silica was found in every lung, and talc in 19 of 20 lungs. Of the fibers, 86% were shorter than 5 pm, and most of the fibers had aspect ratios less than 15. No correlations were seen between numbers or types of fibers and age, sex, or smoking. It is concluded that (1) substantial numbers of nonasbestos fibers are present in lungs of the general population. Most of these fibers are short and of low aspect ratio; and (2) by phase microscopy one is as likely to observe nonasbestos as asbestos fibers in these preparations, indicating that light microscopic methods are not suitable for this type of analysis. These data provide a baseline for comparison with patients believed to have a mineral fiber-related disease.

INTRODUCTION

Numerous types of mineral particulates are present in the lungs of everyone in the population. A portion of these particulates are fibers, and most of the published data on fibers in lung relates to one special variety, namely asbestos; information is beginning to accumulate concerning the specific numbers and types of asbestos fibers found in various occupationally exposed and environmentally exposed groups (Churg and Warnock, 1980; Churg, 1982; Ashcroft and Heppleston, 1973; Sebastien et al., 1977; McDonald et al., 1982). The dangers of exposure to asbestos and its potential for inducing pulmonary fibrosis and malignancies are well documented (Becklake, 1976). Asbestos is not the only type of mineral found in lung. Berry et al. (1976) and LeBouffant (1974) identified silica, micas, clays, aluminum silicates such as sillimanite, spinels, gypsum, and a variety of other minerals in patients with and without occupational dust exposure. If one uses the generally accepted definition of a fiber as a mineral particle with 1 Supported by Grant PDT-146 from the American Cancer Society. 2 Address to which correspondence should be addressed: Department of Pathology, University of British Columbia, 2211 Wesbrook Mall, Vancouver, B.C., Canada V6T lW5. 189 0013-9351/83 $3.00 Copyright Au rights

@ 1983 by Academic Press, Inc. of reproduction in any form reserved

190

ANDREW

CHURG

roughly parallel sides and an aspect (length to width) ratio of 3: 1 or greater, then it is apparent that many types of mineral fibers other than asbestos are also found in lung. Aside from the fact of their existence, there is little detailed information about these fibers; in particular, it is unclear whether nonasbestos mineral fibers can induce disease (excluding, of course, minerals such as silica which obviously induce disease, but which have only a minor fibrous component). A few reports have been published in which nonasbestos mineral fibers appeared to produce interstitial fibrosis. The fibrous aluminum silicate, mullite, a component of fly ash as well as a commercially used component of ceramic insulating materials has been tentatively associated with fibrosis in two patients described by Seal (1980), and we have also seen such a case (Golden et al., 1982). Bignon et al. (1980) have recently presented a patient in whom fibrosis was probably the result of exposure to the fibrous clay attapulgite, and Casey et al. (1981) have reported a patient exposed to the fibrous zeolite, erionite, in whom both parenchymal and pleural fibrosis were present. Perhaps more important, fibrous and nonfibrous mineral particles appear to have some potential for inducing neoplastic disease. It has been shown that particles which by themselves are inert, for example iron oxide, will increase the yield of pulmonary tumors in experimental animals when the inert particle is administered along with a known carcinogen (Nettesheim and Griesemer, 1978). This observation may be related to the increased lung cancer rates found in areas with high levels of atmospheric pollution. Some types of mineral particles, for example mullite (fly ash), appear to adsorb potential carcinogens and the greatest concentrations of such carcinogens are found on the smallest most respirable particles (Natusch et al., 1974). Fibrous particles are particularly dangerous: experimentally it has been demonstrated that any long thin insoluble fiber of the proper size (diameter less than 1 pm, length upward of 5 to 8 pm) may induce mesotheliomas when instilled into the pleura or peritoneum of experimental animals (Stanton et al., 1977). This observation has been confirmed by the discovery of mesotheliomas and pleural disease in a region of Turkey where the soil and building materials contain erionite (Baris et al., 1979). In this paper we report the pulmonary nonasbestos mineral fiber content in the lungs of 20 patients who appear to have no occupational fiber exposure. These data will serve to provide background information on exposure of the general population to fibrous minerals and eventually permit evaluation of fiber-induced disease in exposed individuals. MATERIALS

AND METHODS

Patient selection. The 20 patients included in this report are part of a group initially selected because of a lack of occupational exposure to asbestos (by history), and a light microscopic asbestos body count of fewer than 100 bodies/g wet lung, a level believed to be indicative of a lack of occupational asbestos exposure (Churg and Warnock, 1980). Twenty-one patients were originally included in the detailed report on asbestos fibers in this population (Churg and Warnock, 1980); one was excluded here because she in fact had brief occupational exposure to asbestos with the development of pleural plaques. All of the patients were selected from a general autopsy service and all were 40 years or older at the time of death.

PULMONARY

MINERAL

FIBERS

191

Detailed smoking, residential, and occupational histories were obtained for each patient by interviews with relatives using a standardized questionnaire as described elsewhere (Churg and Warnock, 1980). None of the 20 patients appeared to have a history of occupational exposure to mineral fibers. Specimen preparation. The details of the methods have been published (Churg and Warnock, 1980). Briefly, 3- to 5-g pieces of Formalin-fixed lung are dissolved in bleach and the sediment then treated with 30% H,Oz to oxidize carbonaceous debris. The remaining sediment is collected on a Millipore filter of 0.45pm pore size and the mineral particles finally transferred to coated electron microscope grids by dissolving the filter in acetone. An additional piece of adjacent lung is dried to constant weight to allow conversion of fibers/g dry wt, if desired. For most lungs the ratio of wet to dry weight is approximately 10. The total filter area used in these preparations is a 17-mm disc (289 mm2). The electron microscope grids are 300-mesh nickel with square openings measuring 54 pm on a side. The conversion factor is 97,000 x the number of fibers/square. For each case, samples were prepared from central and subpleural upper and lower lobes in order to study fiber distribution. For each sample, 100 squares of an electron microscope grid were examined, and every fiber seen was counted, measured, examined by electron diffraction, and analyzed for chemical content by energy dispersive x-ray spectroscopy as described by Churg and Warnock (1980). Numbers, types, sizes, aspect ratios, and compositions were tabulated for each case. The total number of fibers analyzed was approximately 2000. For comparison to known standards, a variety of well-defined mineral species were obtained through the courtesy of Dr. Jean Smith of the California Academy of Sciences. For the purposes of this paper, chlorite and vermiculite, which are difficult to separate, were considered as one group, as were muscovite and illite. For silica, pyroxenes, and feldspars, the headings in the tables refer to the broad mineral groups. In most instances, formal analysis of diffraction patterns was not performed (with the exception of a few patterns analyzed to prove the presence of specific minerals such as gypsum), but the presence of a pseudohexagonal pattern was used to confirm the presence of illite and kaolinite and to separate talc from anthophyllite. Histologic examination and clinical data. Histopathologic sections from the autopsy material were reviewed for each case to confirm the overall diagnosis and to define any pulmonary disease which might be present. Clinical details were obtained by review of patient charts. Statistical methods. In our experience numbers of fibers in any given site tend not to be normally distributed, hence the nonparametric Wilcoxon paired rank sum test was used for evaluation of differences in numbers of fibers in different sampling sites. Spearman’s rank correlation test was used to determine correlation coefficients. RESULTS Numbers offibers. Total numbers of asbestos and nonasbestos fibers for each case are shown in Table 1. The average number of nonasbestos fibers for the 20 cases was 106 x 103 + 87 x 103/g wt lung (mean 2 standard deviation). No significant differences were seen between men and women: the 11 men had a mean

95/F

43/F 64/F

23

24 27

100

50

64/F

70/M

56/M 67/F 54/F

6YF 55/M 56/M

117

118

120 202 205

209 214 275

Mean

Smoker 130 None

100

20

None

66 Smoker

47/M 81/M

112

111

None 180 60 34 60 60

63/F 63/M 78/M 83/M 68/M SO/M

None

20 25

Smoking (pack-years)

85 92 94 95 96 103

68

Age/ sex

Case No.

CLINICAL

DATA

Housewife Truck driver, probably insulated home Draftsman, supervised work in sawmills and construction sites

Sold building supplies, drove trucks to construction sites Contractor, insulated Dry cleaning business, 20 year Housewife

Teacher, housewife (husband worked on railroad) Beautician, restaurant work 10 years in factory, exact details unknown, installed home insulation Wholesale furniture store (father worked in shipyard, brought clothes home) Social worker Offtce worker Pilot, 1 year as merchant seaman 2 years ship boiler room, offtce worker Clerical, mined gold as hobby 20 years at a cotton mill, steel mill construction Printer, may have insulated home Detective, exposed to dust in rubber factory Clerk Ca

GI bleed MI Renal Cell Ca Pneumonia Ca lung ASCVD

Ca stomach lymphocytic interstital pneumonia Ca colon

Ca pancreas Ca pancreas

Ca colon Cirrhosis MI Ca lung Ca lung Hepatocellular

Ca cervix Pancreatic abcess/sepsis CVA

Ca cervix

Underlying disease

TABLE 1 20 SUBJECTS (11 MEN AND 9 WOMEN)

OF

Occupational/hobbies/other possible asbestos exposure

AND FIBER

102

151 408 215

164 75 63

79

93

220 78

143 66 101 47 40 26

128

94 95

149

Asbestos fibers/g of lung (X103)

106

200 350 220

110 110

48

33

71

33 2

8

2

140 110

$ z

5

140 11 31 82 2.5

42

61 70

230

Nonasbestos fibers/g of lung (X103)

PULMONARY

MINERAL TABLE

TOTAL

FIBERS

BY SEX

AND

SMOKING

193

FIBERS

2 CATEGORIES

(FIBERS/g

WET

Mean (X103)

LUNG)

SD (x 103)

96 11.5 93 140

Men Women Smokers Nonsmokers

106 65 89 85

of 96 x 103 & 106 x 103 and the 9 women had a mean of 115 x 103 ? 65 x 103 nonasbestos fibers/g wet lung. No statistically significant differences were observed between the 15 smokers and 5 nonsmokers who had means of 93 x lo3 ? 89 x lo3 and 140 x lo3 ? 85 x IO3 fibers/g, respectively (Table 2). Total numbers of fibers present in each case were also correlated with age using the Spear-man correlation test to determine if any increase occurred with time: the correlation coefficient of 0.026 was not statistically significant. Numbers of each specific type of mineral fiber encountered in the 20 lungs are shown in Table 3. A total of 13 different mineral groups or species were definitely identified; these constituted 71% of the fibers counted. Of the fibers 2% could not be definitely identified, either because of technical problems and problems of overlapping particles on the electron microscope grids, or, infrequently, because the diffraction and composition data did not appear to correspond to a defined mineral species. TABLE NUMBERS

Mineral or group Apatite Talc Attapulgite Gypsum Silica Rutile Kaolinite Mullite Illite Pyroxene Pyrophyllite Feldspar Vermiculite/ Chlorite NOS” Total

AND

SIZES

3

OF NONASBESTOS

FIBERS

IN

20

PATIENTS

Percentage No. cases?

Total

12 19 12 6 20 9 12 5 10 5 6 3 6

19,000 17,000 8,400 7,000 6,000 4,100 3,500 3,400 3,200 2,200 730 568 350

20

31,000 -t 58,000 106,000

f 47,000 * 25,000 f 13,000 f 23,000 + 6,400 f 7,808 -c 6,908 f 7,800 f 6,200 f 6,700 + 1,800 f 2,100 * 780

5-9.9 pm

lO+ pm

Percentage of totaP

89 71 100 54 78 98 94 94 81 98 97 98 49

4.9 23 0 33 14 2.0 3.7 6 16 2.0 3.0 2.0 46

6.1 6.0 0 13 8.0 0 2.3 0 3 0 0 0 5

18 16 7.9 6.6 5.7 3.9 3.3 3.2 3.0 2.1 0.7 0.5 0.3

87 86

7.1 11

5.9 3.0

29

l-4.9

pm

a Indicates number of cases (out of 20) in which mineral was identified. b Percentage of total of 106,000 fibers. c NOS = not identified.

194

ANDREW

CHURG

Because of published reports linking both kaolinite and silica particles in lung to cigarette smoking (Brody and Craighead, 1975; Shelbume e? al., 1979), total numbers of pack-years smoked were correlated with kaolinite and silica content for each lung. The corresponding correlation coefficients were 0.317 and -0.194, values which are not statistically significant. When pack-years were correlated with total numbers of fibers, the correlation coefficient was -0.259, a similarly nonsignificant number. Table 3 shows the various mineral species arranged in order of mean numbers present. The most frequently observed fiber overall was apatite at a mean number of 19 x lo3 + 47 x 103 fibers/g wet lung. Apatite constituted 18% of the total nonasbestos fibers. This was closely followed by talc, present at a mean of 17 x 103 ? 25 x 103 fibers/g, or 16% of the total. These two species constituted more than one-third of the total fibers present, and apatite, talc, attapulgite, gypsum, and silica constituted 54% of the total fibers. As shown in Table 3, only silica fibers could be found in every lung and talc was seen in 19 of 20 lungs. Pyroxenes, pyrophyllite, feldspars, and vermiculite/chlorite were seen in small numbers and were present in only a few cases. Using the Spearman rank correlation, no statistically significant correlations could be found when the numbers of the various types of fibers were examined for all possible pairs, nor did any type of fiber correlate with numbers of amphibole or chrysotile asbestos fibers. Fiber sizes and aspect ratios. Table 3 shows the breakdown of each individual mineral and the total collection of fibers by size categories. Overall 86% of the fibers were between 1 and 4.9 pm in length, 11% were between 5 and 9.9 pm, and 3.0% were 10 pm or longer. For most of the individual mineral species, the size distribution was roughly similar; however, for talc, gypsum, and chlorite 29 to 51% of the fibers were 5 pm or longer. Almost all of the attapulgite and mullite occurred as aggregates and the values listed in Table 3 for these species refer to the numbers of aggregates, but the size breakdown refers to the lengths of the longest fibers in the aggregates. The mean aspect ratios for the various fiber types broken down by the same size categories are shown in Table 4. For all fibers between 1 and 4.9 pm in length, the mean aspect ratio was 30 t 58, for fibers 5 to 9.9 pm, 25 + 71, and for fibers 10 pm and longer, 44 ? 180. Aspect ratios greater than 25 were found for some lengths of fibers of apatite, silica, kaolinite, illite, vermiculite, and rutile, and greater than 100 only for long fibers of apatite. Table 5 shows the mean aspect ratios broken down by case: aspect ratios close to or exceeding 100 were seen only in cases 68, 117, and 275. In all three cases this value appeared to reflect the preponderance of apatite fibers. Mean fiber widths are shown for each case in Table 5. Overall, the mean width of fibers was 0.28 -+ 0.41 pm for fibers between 1 and 4.9 ,um in length, 0.88 + 1.1 pm for fibers between 5 and 9.9 pm in length, and 1.7 t 2.3 pm for fibers 10 pm and longer. Distribution. The mean number of fibers in the subpleural upper lobe sample was 30.5 x 103, in the peripheral lower lobe sample 37 x 103, in the central upper lobe sample 19.5 x 103, and in the central lower lobe sample 19.5 x 103 fibers/g

PULMONARY

MINERAL TABLE

195

FIBERS

4

ASPECT RATIOS FOR DIFFERENT MINERAL TYPES Mineral or mineral group

l-4.9

Apatite Talc Attapulgite Gypsum Silica Rutile Kaolinite Mullite Illite Pyroxene Pyrophyllite Feldspar Vermiculite/chlorite NO9

pm

pm

5-9.9

IO+ pm

85 7.8 58 8.2 6.6 21 12 19

160 8.1 10 55

870

63 5.1 -

-

11 12 10

10 23 5.7 -

5.8 39 23

15 12 39 92 32

24 63

6.8 37

a NOS = not identified.

ASPECT RATIOS

AND WIDTHS

FOR TOTAL FIBER

TABLE 5 NONASBESTOS SIZE CATEGORY

FIBERS

Aspect ratios Case No. 23 24 27 28 85 92 94 95 96 103

111 112 117 118 120 202 205 209 214 275

l-4.9

pm 17 36 41 98 16 34 26 17 19 14 36 19 17 46 43 13 15 12

pm

5-9.9

BY CASE AND

Widths lO+ pm

l-4.9

/.l

5-9.9

*rn

lO+ pm

15

18

7.6 28 21 7.1 91 6.0 27 38

32 19 65 9.5

0.25 0.44 0.22 0.077 0.25

0.63 0.68 0.45 0.66

1.0

1.2

15

0.16

21 35 23

0.22 0.21 0.26 0.44 0.22 0.21

0.077 1.3 0.77 1.7 0.92 0.48 0.88

1.4 1.3 1.2 0.62 1.7 0.72 1.0 0.047 3.0 -

11

11

24 8.4 8.6 7.0 19 9.4 7.1 9.9

22 17

0.18

1.0

6.8 -

0.10 0.23 0.42 0.41 0.34 0.22 0.062

1.0 0.50 1.1 1.2 0.98 1.3 0.82

1.3 0.84

0.28 *.41

0.88 -cl.1

1.7 22.3

149

223

30 ?58

25 k71

44 k180

15

1.7 3.2 2.1 1.0

510

21 48 15 22 262

15

SEPARATED

1.2

1.5 1.5

196

ANDREW TABLE NUMBERS

OF FIBERS

CHURG 6

IN VARIOUS

Peripheral upper lobe Peripheral lower lobe Central upper lobe Central lower lobe Total peripheral Total central

SAMPLING

SITES

31 37 20 19 68 39*

*P < 0.01.

wet lung (Table 6). Comparison of the actual values for each case using the Wilcoxon test revealed that the differences for the total subpleural fibers (67.5 x 103) compared to the total central sample (39 x 103) approached significance (P -=I 0.07). Correlations with histological findings and diseases. No patients were found with interstitial fibrosis not readily explicable as a result of either treatment or old infectious disease. Numbers of both asbestos and nonasbestos fibers are listed in Table 1 for the three patients with lung cancer and three with gastrointestinal cancer. These values did not appear to be different from those of the group as a whole. DISCUSSION

In this paper we have presented data on pulmonary nonasbestos mineral fibers in a series of patients who appear to have no occupational mineral fiber exposure. Several conclusions can be drawn from this data: (1) On the average there are roughly as many nonasbestos mineral fibers as asbestos fibers present in these lungs. (2) The distribution of nonasbestos fiber sizes is similar to that of the asbestos fibers with more than 80% shorter than 5 pm. (3) The aspect ratios for the nonasbestos minerals are on the whole much lower than for the asbestos minerals. (4) Numbers of fibers do not correlate with amount of smoking and do not correlate with age, suggesting that fibers are not accumulated in the lung over time. (5) Numbers of fibers do not correlate with each other, suggesting that they are inhaled from a variety of sources which are, proximally, atmospheric. These points are discussed below. The mean number of nonasbestos fibers in these 20 lungs, 106 x 103/g, is roughly equal to the number of asbestos fibers, 102 x 103/g. However, the variety of fiber types present is considerably greater. Thirteen different species of nonasbestos fibers were observed, of which five (apatite, gypsum, talc, silica, attapulgite) accounted for more than half the fibers, while chrysotile accounted for more than 80% of the asbestos fibers. The size distribution of the fibers was similar for both asbestos and nonasbestos types, with more than 80% of both categories shorter than 5 pm. In these 20 lungs, more than 20% of gypsum, silica, and talc fibers were longer than 5 pm; these three species together accounted for 55% of the fibers longer than 5 pm. This fact should be borne in mind when attempting to count “asbestos” fibers using a counting chamber and a phase microscope (for example, the

PULMONARY

MINERAL

197

FIBERS

study of Whitwell et al. (1977)), since there is no way to actually identify the type of mineral present using such a system. In the present cases, only about 10 x 103/g asbestos fibers (data derived from Churg and Wamock, 1980) would be visible in a phase microscope (i.e., fibers 5 pm and longer and 0.2 pm or more thick) compared to approximately 6 x 103 fibers of talc, gypsum, and silica and about 11 X 103 nonasbestos fibers total. Thus one is as likely to observe a nonasbestos as an asbestos fiber using the light microscope counting system. The aspect ratios of the nonasbestos fibers are, on the whole, much lower than those of the asbestos fibers, in particular they resemble the amphibole fibers anthophyllite, tremolite, and actinolite which have aspect ratios in the range of 20 to 30. Although one can find fibers of chrysotile or the commercially used amphiboles, amosite and crocidolite, with this low an aspect ratio, overall the mean aspect ratio for these types of asbestos fiber for both occupationally and environmentally exposed cases is closer to 100 (Churg, 1982). Since high aspect ratio appears to be important in the pathogenesis of malignancy (Stanton et al., 1977), it may be that the majority of nonasbestos fibers are of a rather innocuous shape. The particular nonasbestos fibers which do not in fact have very high aspect ratios, mainly apatite, are all very short, and fibers shorter than 5 ,um are probably less pathogenic than those longer than 5 pm (Selikoff and Lee, 1978). These considerations may provide some theoretical reasons why this fiber burden is relatively innocuous; however, as mentioned in the introduction, case reports of interstitial fibrosis secondary to mullite and attapulgite inhalation have been published, and sufficient numbers of even these very short fibers may be dangerous. It has been claimed (Brody and Craighead, 1975) that the platy inclusions which produce the pale gold color of pulmonary macrophages from smokers’ lungs are kaolinite. Shelborne et al. have also shown that cigarette smoke contains silica (1979). Nonetheless, we were not able to show any accumulation of either kaolinite or silica in the smokers as opposed to the nonsmokers, indeed, we were not able to demonstrate any greater number of mineral fibers overall in the smokers’ lungs, despite evidence that clearance in smokers is impaired compared to that of nonsmokers. However, the number of cases examined is quite small, and it is worth noting that three of the five nonsmokers had no kaolinite, while one had 3 1 x lo3 fiber/g, the highest level of any of the patients, a value which clearly biases the mean (Table 7). It is also possible that numbers of fibrous particles are not a TABLE KAOLINITE

Case No. 23 68 85 117 275 Mean for 5 nonsmokers Mean for 15 smokers

AND

SILICA

7 IN FIVE

NONSMOKERS

Total fibers (X 103/g)

Silica (X 103/g)

Kaolinite (x 103/g)

230 42 140 71 220 140 93

11 9.9 7.6 1.7 0 6.0 6.0

31 0 0 0 4.0 7.0 2.4

ANDREW

198

CHURG

good measure of numbers of total particles and that differences would be seen between smokers and nonsmokers if all particles were counted. We were also unable to show any accumulation of total nonasbestos fibers with age in either smokers or nonsmokers, nor was there a correlation between number of pack-years smoked and total fibers present, suggesting that, at least at these minimal exposure levels, inhaled fibers are cleared at a constant rate. Although there was no statistically significant difference in numbers of fibers found in the various sampling sites, there was a distinct trend toward accumulation of fibers under the pleura in both animals and humans (Morgan et al., 1977; Sebastien et al., 1977; Churg and Warnock, 1980) and this is probably true of other types of fiber as well. There did not appear to be any tendency to accumulate fibers in either upper or lower lobes. The possible effects of nonasbestos fiber accumulation under the pleura are uncertain; we have speculated previously that accumulation of asbestos fibers in this location may be related to the development of pleural fibrosis, pleural plaques, and mesothelioma (Churg and Warnock, 1980). The source of these numerous and varied mineral fibers is uncertain. Table 8 shows a list of patients with occupations or hobbies which conceivably could result in exposure to mineral particles and the major types of fibers found in those patients’ lungs. No obvious correlations are present, and in fact equally large numbers of every one of the major fiber types listed were found in other patients who had no similar exposure. Possible origins of some of the fibers can be speculated. Mullite is a common component of fly ash, and atmospheric fly ash derived from coal burning (probably power stations) seems the most likely source of exposure for the six patients whose lungs contained this mineral, although patient 95 may also have been exposed in working in a ship’s boiler room. The home environment may sometimes be the source of such fibers. Gypsum is a major component of wallboard, and patient 27 was known to have insulated his home. Cosmetics and powders with talc bases might account for some of the talc found in lungs. It is possible that apatite is actually of endogenous origin, but apatite is also a mineral component of many rocks. The actual original source of most of the minerals identified remains uncertain, but there is good reason to believe that most of the fibers are inhaled as atmo-

POSSIBLE

Case No. 21 68 9s 103 111 112 202

EXPOSURE

Major

SOURCES

fiber

Gypsum Attapulgite MuUite Gypsum Attapulgite Rutile Talc

type

IN SOME

Number (X103/g) 23 100 21 100 52 13 70

TABLE 8 PATIENTS WITH

Possible

LARGE

source

NUMBERS

of exposure

Insulated home Furniture store Ship’s boiler room Cotton mill, steel mill Insulated home Rubber factory Dry cleaning

OF ONE TYPE OF FIBER Mean number of this type of fiber in all patients in study (x 103) 7.0 a.4 3.4 7 8.4 4.1 17

PULMONARY

MINERAL

FIBERS

199

spheric contaminants. Armstrong and Buseck have recovered particles of quartz, clays, mica, feldspars, calcite, gypsum, pyrite, iron oxides, bauxite, zircon, ilmenite, and rutile from urban air (Armstrong and Buseck, 1977). Atmospheric contamination would at least account for the presence of minerals such as silica in every lung examined. The fact that the numbers of the different mineral species in the lung do not correlate with each other implies that each entered the atmosphere from a different source. Although one cannot entirely rule out the possibility that some of these patients had occupational fiber exposure, the magnitude of such exposure, if it occurred, appears small. Thus this group of patients can be used as a baseline to compare numbers and types of fibers found in persons with definite occupational exposure and disease. REFERENCES Armstrong, J. T. Buseck, P. R. (1977). The chemical composition, morphology, and surface properties of individual airborne microparticles. In “Proceedings of the Fourth International Clear Air Congress” (S. Kauga, N. Suzuki, T. Yamada, G. Kimura, K. Inagaki, and K. Onoe, Eds.), pp. 617-620. Japanese Union of Air Pollution Prevention Association. Ashcroft, T., and Heppleston, A. G. (1973). The optical and electron microscopic determination of pulmonary asbestos fibre concentration and its relation to the human pathological reaction. J. Clin.

Pathol.

26, 224-234.

Baris, Y. I., Artvinli, Acad.

M., and Sahin, A. A. (1979). Environmental

mesothelioma in Turkey. Ann. N. Y.

Sci. 330, 423-432.

Becklake, M. R. (1976). Asbestos related diseases of the lung and other organs: Their epidemiology and implications for clinical practice. Amer. Rev. Respir. Dis. 114, 187-227. Berry, J. P., Henoc, P., Galle, P., and Pariente, R. (1976). Pulmonary mineral dust-A study of ninety patients by electron microscopy, electron microanalysis, and electron microdiffraction. Amer. J. Pathol.

83, 427-438.

Bignon, J, Sebastien, P., Gaudichet, A., and Jaurand, M. C. (1980). Biological effects of attapulgite. In “Biological Effects of Mineral Fibres” (J. C. Wagner, Ed.), Vol. 1. pp. 163- 181. International Agency for Research on Cancer, Lyon. Brody, A. R., and Craighead, J. E. (1975). Cytoplasmic inclusions in pulmonary macrophages of cigarette smokers. Lab. Invesr. 32, 125- 132. Casey, K., Moatamed, F., Shigeoka, J., and Rom, W. (1981). Demonstration of fibrous zeolite in pulmonary tissue. Amer. Rev. Respir. Dis. 123, 98 (Abstract). Churg, A. (1982). Fiber counting and analysis in the diagnosis of asbestos related disease. Human Pathol. 14, 381-392. Churg, A., and Wamock, M. L. (1980). Asbestos fibers in the general population. Amer. Rev. Respir. Dis. 122, 669-678. Golden, E., Warnock, M. L., Hulett, L., and Churg, A. (1982). Fly ash lung. A new pneumoconiosis? Amer. Rev. Respir. Dis. 125, 108- 112. LeBouffant, L. (1974). Investigation and analysis of asbestos fibers and accompanying minerals in biological materials. Environ. Health Perspect. 9, 149- 153. McDonald, A. D., McDonald, J. C., and Pooley, F. D. (1982). Mineral tibre content of lung in mesothelial tumors in North America. In “Inhaled Particles V.” Arch. &cup. Hyg. Morgan, A., Evans, J. C., and Holmes, A. (1977). Deposition and clearance of inhaled fibrous minerals in the rat. Studies using radioactive tracer techniques. In ‘Inhaled Particles IV,” Part 2 (W. H. Walton and B. McGovern, Eds.), pp. 259-274, Pergamon, New York. Natusch, D. F. S., Wallace J. R., and Evans, C. A. (1974). Toxic trace elements: Preferential concentration in respirable particles. Science 183, 202-204. Nettesheim, P., and Griesemer, R. A. (1978). Experimental models for studies of respiratory tract carcinogenesis. In “Pathogenesis and Therapy of Lung Cancer” (C. C. Harris, Ed.), pp. 75 188. Dekker, New York.

200

ANDREW

CHURG

Seal, R. M. E. (1980). Current views on pathological aspects of asbestosis. The unresolved questions and problems. In “Biological Effects of Mineral Fibres, (J. C. Wagner, Ed.), Vol. 1, pp. 217-235. International Agency for Research on Cancer, Lyon. Sebastien, P., Fondimare, A., Bignon, J., Monchaux, G., Desbordes, J., and Bonnaud, G. (1977). Topographic distribution of asbestos tibres in human lung in relation to occupational and nonoccupational exposure. In “Inhaled Particles IV,” Part 2 (W. H. Walton and B. McGovern, Eds.), pp. 435-444. Pergamon, New York. Selikoff, I. J., and Lee, D. H. K. (1978). “Asbestos and Disease.” Academic Press, New York, passim.

Shelburne, J. D., Gutknecht, W. F., Wilder, D. R., Ingram, P., and Hawkins, H. K. (1979). Fed. Proc. 38, 1155 (Abstract). Stanton, M. F., Layard, M., Tegeris, A., Miller, E., May, M., and Kent, E. (1977). Carcinogenicity of fibrous glass: Pleural response in the rat in relation to fiber dimension. J. Nat. Cancer Inst. 58, 587-603. Whitwell, F., Scott, J., and Grimshaw, M. (1977). Relationship between occupations and asbestos libre content of the lung in patients with pleural mesothelioma, lung cancer and other diseases. Thorax

32, 377-386.