Comparison of Light and Electron
Microscopy for Defining Occupational
Asbestos Exposure in Transbronchial Lung
Biopsies
Ronald F. Dodson, Ph.D., F.C.C.P;* George A. Hurst, M.D., F.C.C.P.,*
Marion G. Williams, Jr, B. S.;* Carolyn Corn, B. S. ;*
and S. Donald Greenberg, M. D., F. C.P t
Since asbestos burden in the lung can vary among areas,
the usefulness of small tissue samples for identifying past
occupational exposure is examined. Simulated transbronchial biopsy samples and open lung biopsy samples were
collected from autopsy material from 12 former amosite
asbestos workers and ten persons from the general population. Tissue evaluation included (1) paraffin embedment
and light microscopy screening for fibrosis and ferruginous
bodies, and (2) tissue digestion, which was analyzed by the
combination of (A) light microscopy screening for ferruginous bodies and (B) electron microscopy (EM) screening
for uncoated fibers. Using standard pathology techniques
to classify the small samples was generally unsuccessful,
the samples being too small or their size compounding
other random sampling problems. The most reliable method
of establishing which transbronchial biopsy tissue samples
were from the occupationally exposed group occurred when
light and EM analyses were used to evaluate digested
tissue. The combined data from the EM analysis of two
samples per subject indicated controls had two or fewer
observed asbestos fibers, while the amosite asbestos workers had six or more fibers. This distinction was valid even
in those who, 21 years before sampling, had worked for
only a few weeks in the asbestos plant.
Tissue sampling is important for increasing diagnostic accuracy when the underlying cause of interstitial lung disease is unknown or the staging of the
disease is required for therapy selection and/or prognosis.' In such cases, an open lung biopsy may be
done, but technical advances associated with fiberoptic
bronchoscopy now permit another initial sampling
option via a transbronchial biopsy
While the procedure has fewer associated risks than
an open lung biopsy, the size of the transbronchial
samples presents special concern because of the possibility of obtaining nonrepresentative samples of the
diseased tissue and because such small samples often
do not provide sufficient tissue for adequate evaluation.
The known variations in ferruginous body (FB) and
uncoated fiber burdens from one lobe to another in
patients exposed to asbestos further illustrate this
problem.2-4 Multiple sampling is one approach used in
an attempt to overcome these difficulties.
While the need for tissue samples in the validation
of past occupational exposure to dusts such as asbestos
is not as critical when accurate occupational histories
are available, a significant number of contradictory or
incomplete histories makes such a procedure useful
in clinical evaluations in these cases.5 This is even
more important when clinical progression is sus-
*The University of Texas Health Center at Tyler, Texas.
tBaylor College of Medicine, Houston.
Manuscript received October 12; revision accepted February 5.
Reprint requests: Dr. Dodson, University of Texas Health Center; US
Hwy 271 at State 155, Tyler 75708
366
pected.
The methods available for analyzing small lung
samples include the classic pathology evaluation by
light microscopy used for grading fibrotic involvement
and identifying the presence of FBs in tissue sections
and the less widely used, but potentially more specific,
electron microscopy (EM) analysis.4'6 Lung tissue
digests, in particular,7 8 allow quantification of the FB
as well as fiber content. Electron microscopic techniques, while more complex and time consuming, can
also provide accurate core analyses of the FBs and,
therefore, are used to distinguish asbestos bodies from
other FBs.9-11
The use of EM also allows the detection of uncoated
asbestos fibers which potentially are an even more
sensitive indication of the level of past exposure. This
would be particularly true in individuals who do not
readily form FBs, even when large numbers of uncoated asbestos fibers are present in the lungs.4
We used a simulated biopsy procedure designed to
evaluate the usefulness of transbronchial biopsy specimens in establishing occupational asbestos exposure.
The data were compared with samples similar to those
obtained from an open lung biopsy Light and electron
microscopic techniques were used to evaluate both
groups of samples.
Light and Electron Microscopy in Occupational Asbestos Exposure (Dodson et al)
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21581/ on 06/15/2017
and two samples added to the above pooled wet/dry weight.
Forceps biopsy tissue used for the digestion studies was placed
directly into particle free containers. The four pooled samples were
placed in preweighed containers, weighed wet, oven dried for 72
hours and weighed dry The resulting weights were divided by 4 to
obtain wet and dry weights per site.
For the open lung biopsy tissue, samples were weighed wet, after
all free fluid had been expressed. Ifthe tissue was used for digestion,
the fluid was collected directly into the digestion vial and processed
with the tissue. Tissue used for detertnining the wet/dry ratio was
weighed wet, oven dried 72 hours and weighed dry. To obtain the
calculated dry weight, the wet weight of digestion tissue was divided
by the wet/dry ratio.
MATERIAL AND METHODS
Autopsy lung tissue for this study was obtained from ten control
subjects who had no known history of lung disease or occupational
exposure to asbestos and from 12 former amosite asbestos workers
who had worked in a local asbestos plant for periods ranging from
two weeks to 15.4 years. Both groups were from the nonurban East
Texas area and, therefore, not exposed to naturally occurring
asbestiform outcroppings. Autopsy lung tissue was collected in
prefiltered 3 percent glutaraldehyde fixative (buffered pH 7.3).
Olympus FB21C transbronchial cupped biopsy forceps were used
to obtain transbronchial lung biopsy samples. Multiple rinses,
sonication, and blanks were used to avoid cross-contamination and
mixing. Areas from lung parenchyma were chosen, which avoided
suspected tumor involvement or major vessels and airways. Sample
size, as well as site, were considered by the pulmonologist who took
the biopsy samples in an effort to obtain specimens similar to those
that might be submitted from a surgical procedure.
One area from the lung of each subject was selected, and 11 sites
sampled as follows. First, three adjacent samples were taken to
simulate an open lung biopsy and used for (1) digestion, (2) paraffin
embedment, or (3) drying to obtain a ratio for calculating dry
weight. Second, two sites located immediately on either side of the
area taken for the open lung biopsy were sampled. On one side,
four adjacent samples were taken to simulate a transbronchial
forceps biopsy One sample was used for digestion, one for paraffin
embedment, and two for the pooled wet/dry weight. On the other
side, four adjacent samples were also taken to simulate a transbronchial forceps biopsy and used for digestion, paraffin embedment,
Pathology Techniques
Material for the pathology evaluation was embedded in paraffin
and submitted as a blind study. The blocks were sectioned, three
slides made, and these were stained as follows: one with hematoxylin
and eosin (H&E) for general details, one with Masson's trichrome
for connective tissue, and one with Perl's Prussian blue stain for
iron. Sections were scanned at 25X, 250X, and 450X. The FBs were
counted and the fibrosis graded. The pathologist in the present
study established a working standard for using transbronchial biopsy
imaterial, which required five alveoli per field to define a lower limit
of meaningful evaluation. Any lung tissue with fewer than five
alveoli was deemed to be of insufficient quantity Fibrosis was
reported when it was seen in at least one of the fields per section.
Fibrosis was scored as none, mild, moderate, or severe. "None"
Table 1-Asbestos Worker Occupation, Interstitial Fibrosis, and Ferruginous Bodies
Fibrosis
FB/Forceps BX*
Patient
No.
1
2
Sex
Age,
Yr
M
72
M
71
Periodt
Worked
0.5 mo
0.6 mo
Occupationt
Loaded box cars
Maintenance
Forceps
BX
None
None
None
Open
Lung
Mild
0
Mild
0
0
4
5
M
68
1.2 mo
Grinder, feeder
None
M
44
58
2.0 mo
2.4 mo
Sacker
Pipe insulation
M
48
3.3 mo
Feeder, builder
0.4
10-20
2.3
0
2
1
<10
9.9
¶
1
0
0
34
<10
1200.0
Mild
0
0
0
0
1
¶
1
1
0
0
0
46
Moderate
Mild
None
Mild
None
7
M
68
2.2 yr
Builder
Mild
Mild
Mild
8
M
57
2.6 yr
Maintenance
None
Mild
None
9
10
M
M
62
57
Moderate
5.0 yr
Builder
8.7 yr
Builder
None
¶T
None
¶1
1
¶1
Mild
11
None
Moderate
¶T
11
M
74
9.9 yr
Sawman
12
M
74
15.4 yr
Builder
Mild
1T
Digest X103§
<10
None
6
Slides
0
2
¶
¶1
M
Digest
0
0
0
¶
3
Slides
FB/Open Lung BX*
Moderate
1
1
1
127
80
257
¶1
Mild
Severe
62
18
40
40
4
1
11
0
1T
I
1
0
1.7
<10
490.0
>20
140.0
>20
520.0
10-20
77.0
10-20
49.0
*Forceps BX = transbronchial biopsy; open lung= open lung biopsy
tData reprinted with permission from reference 3.
:All FB observed/biopsy sample.
§FB/g dry
INone detected.
TInsufficient tissue.
CHEST / 94 / 2 / AUGUST, 1988
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21581/ on 06/15/2017
367
indicated no appreciable fibrosis and was used if the alveolar walls
were thin, delicate, and showed no change. "Mild" described focal
fibrosis, where one or two alveoli showed minimal fibrosis. "Moderate" indicated that a majority of the alveoli showed some fibrosis.
"Severe" indicated that almost all the alveoli displayed significant
fibrosis and described thickened septa, and distortion of alveolar
spaces.
Open Lung Biopsy Digestion Techniques
Open lung biopsy samples of similar size were taken from the
control group and from the asbestos cohort. Some of the data from
the asbestos cohort were reported in a previous study4 These
samples came from the same sites as the transbronchial forceps
biopsies and received the same treatments and pathology evaluations. The mean wet weight and SD were 0.088 g±.05, while the
mean dry weight and SD were 0.0195 g±.01. The only difference
was that these digestion filters were 47 mm in diameter, not 25 mm.
Transbronchial Forceps Biopsy Digestion Techniques
There were two samples from each subject submitted for EM
and light analysis. The mean wet weight and SD were 0.66
mg + 0.54, while the mean dry weight and SD were 0.30 mg ± 0.28.
These were bleach digested in prefiltered 9.2 percent Wright's
laundry bleach (Wright, Inc) according to the method of Williams
et a17 and collected on a 25-mm polycarbonate filter (Nuclepore),
which had a pore size of 0.2 ,u As a control for contamination,
disposables were used where possible, glassware was coated with
0.25 percent Formvar plastic, all reagents used were prefiltered,
and blanks were run and found to be negative. The whole surface
of the intact filter was scanned by light microscopy at 200X and all
FBs counted. After light microscopy, a rectangle from the above
filter was cut from center to edge of the active filtration area,
mounted on a carbon stub, and coated with gold/palladium for
scanning electron microscopy (SEM) and x-ray energy dispersive
analysis (XEDA). Samples were examined in the slow scan mode at
5,OOOX in an AmRay IOOOA scanning microscope equipped with a
Tracor Northern 1710 energy dispersive x-ray analysis system.
Three nonoverlapping scans of 8.6 mm each were made on each
filter. All fibers having a length/width ratio of 3:1 or greater were
counted if they were within the continuous field or extended beyond
the bottom of the area. All fibers were analyzed for the control
group and up to 100 consecutive fibers were analyzed for the
asbestos workers. The XEDA spectra were compared to spectra
from known asbestos standards.
Fiber and Ferruginous Body Calculations
The following calculations were used, where appropriate, to
determine uncoated fibers per filter, fibers per gram, and FBs per
gram. The uncoated fibers per filter were calculated by multiplying
the fibers counted by the active filtration area divided by the area
scanned at X5,000 in the SEM. This quantity was further divided
by the wet or dry weight of the sample to obtain the number of
fibers per gram. The number of FBs per gram was obtained by
dividing the number of FBs on the filter by the wet or dry tissue
weight.
RESULTS
Light microscopic examination of the transbronchial
forceps biopsy tissue sections was frequently not useful
in distinguishing asbestos-exposed from nonasbestosexposed persons (Tables 1 and 2). Tissue size was a
major problem in applying pathology techniques.
Samples were often too small for processing, and those
embedded and sectioned often resulted in sections
containing insufficient alveoli for adequate evaluation.
This occurred in 41 percent of the H&E and trichrome
Table 2-Control Occupation, Interstitial Fibrosis, and Ferruginous Bodies
Fibrosis
Forceps
Age,
Patient
No.
Sex
Yr
13
M
16
Occupationt
Student
BX*
§
Open
Lung*
None
M
16
§
§
§
14
FB/Forceps BX*
^
Slides
Digestt
§
Student
§
None
None
§
None
0
0
0
0
15
M
22
Handyman
None
None
None
16
M
22
Steelworker
§
None
None
None
None
None
None
0
§
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
17
M
25
Handyman
18
F
39
Arsenal
19
M
45
Miscellaneous jobs
§
§
Mild
§
§
§
20
M
52
Chemical
Mild
None
0
Mild
0
Mild
§
§
§
Mild
§
§
21
F
56
Housewife
None
§
70
F
22
Domestic
§
§
=
*Forceps BX - transbronchial biopsy; open lung open lung biopsy
tAll FB observed/biopsy sample.
Slides
Digestt
0
0
0
0
0
FB/Open Lung BX*
60
57
l
62
tFB/g dry
(lNone detected.
§Jnsufficient tissue.
368
Light and Electron Microscopy in Occupational Asbestos Exposure (Dodson et al)
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21581/ on 06/15/2017
Table 3-Uncoated Fibers in Asbestos Workers' Digests
Open Lung Biopsy Samples
Transbronchial Forceps Biopsy Samples
Period
Worked
Chrysotile
Seen
Amphiboles
Seen
Patient
No.
1
0.5 mo
2
0.6 mo
3
1.2 mo
4
2.0 mo
5
2.4 mo
6
3.3 mo
7
2.2 yr
8
2.6 yr
9
5.0 yr
10
8.7 yr
11
9.9 yr
12
15.4 yr
All Asbestos
Seen
5
28
7
13
60
231
68
199
134
1,086
0
1
7
2
7
3
3
0
2
8
7
2
4
5
8
9
4
1
9
12
16
13
15
61
0
3
4
2
4
15
10
4
11
15
3
1
5
10
5
30
7
14
64
249
71
246
135
1
2
0
2
10
1
0
1
3
2
0
1
9
0
2
0
1
4
18
3
47
1
49
0
2
2
2
2
5
9
4
10
12
1
1
4
1
Nonasbestos
Fibers Seen
1,135
0
Amphiboles/*
g Dry X103
Chrysotile/*
g Dry X103
All Asbestos/
g Dry X103
780
100
880
470
17
487
1,300
t
1,300
1,500
180
1,680
14,000
100
14,100
1,200
36
1,236
51,000
t
51,000
136,000
2,600
138,600
32,000
t
32,000
140,000
t
140,000
39,000
4,400
43,400
127,000
6,900
133,900
*Data reprinted with permission from reference 4.
tNone detected.
slides and in 46 percent of the Prussian blue slides.
Another limitation was that 58 percent of the H&E
and 50 percent of the trichrome stained tissue sections
showed normal morphology and thus could not be
used to differentiate between the occupationally exposed and the nonoccupationally exposed subjects.
This was true even though interstitial fibrosis was
evident in all members of the occupationally exposed
Table 4-Uncoated Fibers in Control Digests
Open Lung Biopsy Samples
Transbronchial Forceps Biopsy Samples
Patient
No.
Amphiboles
Chrysotile
Seen
Seen
13
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
14
15
16
17
18
19
20
21
22
1
0
0
0
0
0
0
0
0
1
1
1
0
0
0
0
All Asbestos
Seen
Nonasbestos
Seen
Amphiboles/
g Dry X103
g Dry X103
All Asbestos/
g Dry X103
0
0
0
0
4
3
2
0
*
114
114
149
149
298
1
16
*
*
*
0
1
0
0
0
0
0
0
1
1
1
0
0
0
0
1
2
6
2
2
3
0
1
3
5
4
7
3
1
1
*
204
204
88
*
88
588
*
588
92
*
92
360
*
360
683
114
797
84
*
84
Chrysotile/
*None detected.
CHEST / 94 / 2 / AUGUST, 1988
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21581/ on 06/15/2017
369
group when the larger open lung biopsy samples were
examined. Only four FBs were seen in the tissue
sections ofthe transbronchial forceps biopsies obtained
from the occupationally exposed individuals, one each
in four workers (Table 1). While this marker would
suggest asbestos as the basis of the interstitial fibrosis,
the limited number and/or lack of FBs in the tissue
sections would not permit such a suggestion to be
made in at least eight of 12 cases.
At least one FB was found in the digestions of the
transbronchial forceps biopsies from all but three of
the asbestos workers, while none was found in the
forceps biopsy digests from the general population
(Tables 1 and 2).
A much more accurate discrimination could be made
between occupational and nonoccupational exposures
by comparing uncoated fiber burdens obtained by EM
of the lung digests. Seventy fibers were found in the
transbronchial forceps biopsy tissue digests from the
control group, ofwhich 65 were nonasbestos, four were
amphiboles, and one was chrysotile. The forceps
biopsy digests from the occupationally exposed group
contained 2,242 fibers, of which 1,886 were amphiboles, 157 were chrysotile, and 199 were nonasbestos
(Tables 3 and 4). An evaluation of the combined data
from the EM analysis offibers observed in two samples
per subject indicated control subjects had two or fewer
asbestos fibers, while the amosite asbestos workers
had six or more asbestos fibers, even in those who, 21
years before sampling, had worked for only a few
weeks.
DISCUSSION
Transbronchial lung biopsy samples may be of
limited value in diagnosing occupational asbestos exposure by conventional light microscopy However, in
this study, such biopsies were found to be diagnostically
useful when analyzed, following digestion procedures,
for the presence of FBs and uncoated fibers by light
and EM, respectively An explanation of the low
chrysotile concentrations observed in the control subjects and workers reflect the nonurban environments
where these subjects lived.
The basic shortcomings in studying such small
biopsy samples by conventional pathology methods
include inadequate lung tissue and random sampling
errors. The problem, encountered in several of these
biopsy samples, was that the amount of tissue was not
sufficient to allow a reasonable evaluation of whether
an underlying disease was present or absent. This was
illustrated when attempting to quantify interstitial
fibrosis as well as the presence of FBs, which are
370
important histopathologic criteria for diagnosis of
asbestosis.6 While interstitial fibrosis was found in each
of the open lung samples from the former asbestos
workers,4 such a discrimination was found to be highly
unreliable in the smaller forceps biopsy samples.
Ferruginous bodies in the small forceps samples
were also inconsistent identifiers of occupational exposure, since FBs were found in only four of the 12
asbestos workers, even though previous digestion
studies4 of larger samples revealed a tissue burden
that was appreciably over occupational exposure levels
(>1,000 FBs/g dry weight). 12
Transbronchial biopsy sections from control lungs,
while being almost totally nonfibrotic and having no
FBs, were not distinguishable from tissue sections
from most of the asbestotic lungs. Thus, light microscopy of the small transbronchial tissue biopsies, without electron microscopy, would have been of very
limited value in differentiating the occupationally from
the nonoccupationally exposed individuals, whereas
the electron microscopy data enabled the subjects to
be placed reliably into these categories.
REFERENCES
1 Fulmer JD. An introduction to the interstitial lung diseases.
Clin Chest Med 1982; 3:457-73
2 Churg A, Wood P Observations on the distribution of asbestos
fibers in human lungs. Environ Res 1983; 31:374-80
3 Dodson RF, Greenberg SD, Williams MG, Corn CJ, O'Sullivan
MF, Hurst GA. Asbestos content in lungs of occupationally and
nonoccupationally exposed individuals. JAMA 1984; 252:68-71
4 Dodson RF, Williams MG, O'Sullivan MF, Corn CJ, Greenberg
SD, Hurst GA. A comparison of the ferruginous body and
uncoated fiber content in the lungs of former asbestos workers.
Am Rev Respir Dis 1985; 132:143-47
5 Kane PB, Goldman SL, Pillai BH, Bergofsky EH. Diagnosis of
asbestosis by transbronchial biopsy Am Rev Respir Dis 1977;
115:689-94
6 Craighead JE. Asbestos-associated diseases. Arch Pathol Lab
Med 1982; 106:544-96
7 Williams MG, Dodson RF, Corn CJ, Hurst GA. A procedure for
the isolation of amosite asbestos and ferruginous bodies from
lung tissue and sputum. J Toxicol Environ Health 1982; 10:62738
8 Dodson RF, Williams MG, Hurst GA. Method for removing the
ferruginous coating from asbestos bodies. J Toxicol Environ
Health 1983; 11:959-66
9 Churg A, Warnock ML, Green N. Analysis of the cores of
ferruginous (asbestos) bodies from the general population. Lab
Invest 1979; 40:31-38
10 Churg AM, Warnock ML. Asbestos and other ferruginous bodies.
Am J Pathol 1981; 102:447-56
11 Dodson RF, O'Sullivan MF, Corn CJ, Williams MG, Hurst GA.
Ferruginous body formation on a nonasbestos mineral. Arch
Pathol Lab Med 1985; 109:849-52
12 Churg A, Warnock ML. Asbestos fibers in the general population. Am Rev Respir Dis 1980; 12:669-78
Light and Electron Microscopy in Occupational Asbestos Exposure (Dodson et al)
Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21581/ on 06/15/2017