ESVIRONMESTAL

HESEARCH

Asbestos A. P.

WEHNER,

17, 367- 389 ( 1978)

Cement

Dust Inhalation

G. E. DAGLE, W. C. CANNON,

by Hamsters’ AND R. L. BUSCHBOM”

Departments of Biology and *Energy Systems, Battelle. Pacific Northujest Laboratories, Richland, Washington 99352 Received March 13, 1978 Two groups of 96 male Syrian golden hamsters were exposed to respirable asbestos cement aerosol at concentrations of -1 and -10 &liter, respectively, 3 hours/day, 5 days/week. Average fiber counts ranged from 5 to about 120 fibers/cm-‘. Each group was randomly divided into six subgroups of 16 animals. The first subgroup was sacrificed after 3 months of exposure, the second after 6 months, and the third after 15 months. The fourth subgroup was withdrawn from exposure after 3 months, observed for an additional 3 months, and then sacrificed. The fifth and sixth subgroups were withdrawn after 3 and 6 months of exposure, respectively, and maintained for observation up to the U-month exposure point of the third subgroup at which time all surviving animals were sacrificed. All other experimental procedures were similar to those delineated in a previous publication describing the development of an animal model, techniques, and an exposure system for asbestos cement dust inhalation (A. P. Wehner, G. E. Dagle, and W. C. Cannon, 1978. Environ. Res. 16, 393-407). The asbestos cement exposures had no significant effect on body weight and mortality of the animals. Higher aerosol concentration and longer exposure times increased the number of macrophages and ferruginous bodies found in the lungs of the exposed animals. Recovery periods had no effect on the incidence of macrophages and ferruginous bodies. The incidence of very slight to slight fibrosis in the animals sacrificed after I5 months of exposure shows a significant (P < 0.01) trend when the untreated control group and the 1 and 10 &liter dose level groups are compared, indicating a dose-response relationship. Development of minimal fibrosis continued in animals withdrawn from exposure. No primary carcinomas of the lung and respiratory tract and no mesotheliomas were found.

INTRODUCTION

The literature abounds with reports on the biological effects of asbestos, as reviewed by the U.S. Environmental Protection Agency (1971), the U.S. Department of Health, Education and Welfare (1972), the World Health Organization (1973), and more recently by Becklake ( 1976). The most commonly observed lesions resulting from the inhalation of asbestos dust are asbestosis, pleural plaques, bronchogenic carcinoma, and mesothelioma. Among the numerous commercial applications of asbestos, asbestos cement (AC) ranks first (Speil and Leineweber, 1969; Scansetti et al., 1975). Although manufacturing and cutting of AC sheets and pipes generates varying concentrations of AC dust to which workers might be exposed, literature describing the effects of such exposures in workers is rather sparse (Bohlig, 1970; Solte, 1970; Enterline and Weill, 1973; Scansetti et nl., 1975). To our knowledge, no results of inhalation studies with AC cement dust in an animal model under controlled I This paper is based on work conducted under Contract 2311202074 with Eternit, Kapelle-op-denBos, Belgium. 367 0013-9351/78/0173-0367$02.00/O Copyright All rights

0 1978 by Academic Press, Inc. of reproduction in any form reserved.

368

laboratory determine animals.

WEHNER

ET AL.

conditions have been published to date. It was, therefore, important to the biological effects of AC dust when inhaled by experimental EXPERIMENTAL

DESIGN

A summary of the experimental design is shown in Fig. 1. Two groups of 96 male Syrian golden hamsters (Mesocricetus auratus, outbred LaK:LVG strain, Lakeview Hamster Colony, Newfield, N.J.) were exposed to respirable AC aerosol at concentrations of -1 and -10 &liter, respectively, 3 hours per day, 5 days per week. Since the AC contained approximately 10% chrysotile asbestos, the animals were exposed to 0.1 and 1 pg of asbestos per liter of air, respectively. This corresponds to 1 and 10 times the Technische Richtkonzentration (TRK) for chrysotile established by the Deutsche Forschungsgemeinschaft (1977). A third group of 96 hamsters served as untreated control group. From each of the three experimental groups (low-dose level group, high-dose level group, and control group), randomly selected subgroups of 16 hamsters were scheduled for sacrifice or (in case of the two exposed groups) for withdrawal from exposure after 3 and 6 months of exposure, respectively. The number of hamsters that died in a given subgroup prior to the 3- and 6-month serial sacrifices was deducted from the number scheduled for sacrifice (16 hamsters per subgroup per sacrifice) to keep as many animals as possible alive for final sacrifice at the E-month exposure point. For example, if two hamsters in a given subgroup died between the 3- and 6month exposure points, only 14 instead of 16 animals of that particular subgroup

EXPOSURE T O ASBESTOS RECOVERY/OBSERVATION t

0

I 5

SACRIFICE

I 10

I 15

TIME ON EXPERIMENT (MONTHS) I

I

4

3

8

13

.I

18

AGE (MONTHS)

FIG. 1. Experimental design for the low-dose level (- 1 pg/liter) and high-dose level (- 10 &liter) exposure groups. The same number ( 16) of hamsters from an untreated control group was sacrificed at each sacrifice point. Criteria for evaluating results were body weight. mortality, and gross and histopathologic findings.

ASBESTOS

DUST

INHALATION

BY

HAMSTERS

369

were sacrificed at the 6-month exposure point. The hamsters were 12 weeks old at the start of the exposures. Past experience (Wehner ef al., 1974, 1976, and 1977b) indicated that the average life span of Syrian golden hamsters under the described conditions could be expected to be 15 to 16 months. Criteria for evaluating the effects of AC dust inhalation were body weight, mortality, and gross and histopathologic findings. METHODS

The aerosol exposure system, aerosol generation, characterization and concentration, and animal care and postmortem procedures were similar to those previously described (Wehner et al., 1978). The air flows through the exposure chamber were 283 and 3 11 liter/min for the low- and high-level exposures, respectively. These flow rates provided air change times of 11.3 and 10.3 minutes. The AC dust was collected from cutting uncoated AC sheets at the Eternit plant in Kapelle-op-den-Bos, Belgium, and supplied by the company for the inhalation study. The approximate composition of the asbestos cement was as follows: cement, -83%; chrysotile asbestos, -10.5%; water, -6%; cellulose, -0..5%.2 The material was screened in a 210+m sieve to remove large particles which would plug the Wright Dust Feed Mechanism and stored in a vacuum desiccator at 100°C until used. Photomicrographic examination of AC dust samples showed no evidence of physical changes in the aerosolized AC dust as compared to bulk AC dust samples. These observations were supported by the results of Deruyttere (Catholic University, Leuven, Belgium, unpublished data). Aerosol Characterization The AC aerosols were characterized as to total aerosol concentration, concentration of “respirable” aerosol, and aerodynamic equivalent particle size distribution. Total aerosol concentration was measured by collecting samples on preweighed 25mm-diameter glass fiber filters at two locations in the exposure chamber. These positions were opposite one another, 38 cm in from the chamber wall and approximately 15 cm above the animal cages. Total sample mass per week was divided by total sample volume per week to determine the weekly average aerosol concentration. In addition, the balance of aerosol concentrations in the chamber was monitored by comparing the ratio of the average aerosol concentration at one location to that at the other location. Aerosol sampling times were 5 hours per week for the high-level aerosol and 15 hours per week for the low-level aerosol. Concentration of “respirable” aerosol was measured for each exposure level using a Casella 113A Gravimetric Dust Sampler, This device samples at a flow rate of 2.5 liter/mm and selectively transmits particles which conform to the British Medical Research Council’s definition of a “respirable” aerosol. These samplers were operated during the 15 hours of exposure per week, collecting the * Information supplied by Eternit.

370

ET AL

WEHNER

samples on preweighed %-mm-diameter glass fiber filters. Table 1 shows average values and standard deviations of the weekly average concentrations. Figure 2 shows the 4-week average aerosol concentrations as a function of time. Cascade impactor samples of the aerosols were taken periodically throughout the study to determine particle size distributions, using an Anderson (1966) 2000 Inc. Model 20-000 cascade impactor with eight jet stages and a final filter. Petroleum jelly-coated aluminum foil disks were used as impaction plates to avoid particle bounce. These coated disks are weight-stable after 2 days of curing at room temperature. One-hour samples were taken of the high-level aerosol, usually on a weekly or biweekly basis. Seven 15hour samples were taken from the low-level aerosol. Since the cascade impactor was operated each time at a flow rate of 20 liter/mitt, a composite aerosol size distribution was obtained from a stage-by-stage addition of the periodic cascade impactor aerosol samples. The individual sample values for the high-level aerosol were multiplied by a factor proportional to the exposure time represented by them. For example, biweekly samples were multiplied by a factor of 2 to give them twice the weight of a weekly sample. The composite highlevel aerosol particle size distribution is shown in Table 2. In the composite particle size distribution of the low-level aerosol (Table 3) each of the seven 15 hour samples was given equal weight. Figures 3 and 4 show cumulative plots of the two composite aerosol particle size distributions. The aerodynamic equivalent size distributions of the high-level [mass median aerodynamic diameter (MMAD) 2.5 pm; geometric standard deviation (GSD) 2.41 and low-level (MMAD 2.7 pm; GSD 2.4) aerosols are very similar. To obtain an approximate correlation between the gravimetrically determined aerosol concentrations and the approximate number of asbestos fibers per unit volume, fiber counts were made during the lst, 9th, 13th, and 15th exposure months. Only fibers longer than 5 pm and with diameters less than 3 pm were counted. Sets of 25-mm-diameter Millipore HAWP filters (0.45 pm, average pore size) were used to collect small samples of the AC aerosol. These filters were clarified using the acetone vapor fusion technique. The fibers in 60 to 100 fields on each filter sample were counted using a phase contrast microscope with a Pot-ton graticule. The dimensions of the field and graticule were checked each time with a TABLE ASBESTOS

CEMENT

AEROSOL

I CONCENTRATIONS

High-level exposure

Total aerosol concentration, filter sample (pgiliter) Respirable aerosol concentration, Casella 113A @g/liter) Concentration ratio, left sample to right sample

Low-level exposure

Average (67 weeks)

SD

Average (67 weeks)

SD

14.6

3.8

1.7

0.4

9.9

1.2

1.3

0.3

1.01

0.07

1.01

0.11

ASBESTOS DUST INHALATION

13 -

HIGHLEVELEXPOSURES _---------

.

11 10 - _ 0.; 9

-

7

- ‘2

l*

l

.

:

‘* . .

_

.

.

.

.x

.

--

-0

:*

.

0. l

.

+%

,7l

+y+2*D

9.’

. 0’.

::..*.

.

l

. l

5

.

----

.

12-

I@ 8 -

371

BY HAMSTERS

--3

T-2

SD

70

75

LOWIIVELEXPOSURES

2 - ------

~-..~~~~-..~~~~~,1t2SD

1

OL”” 0

5

10

15

20

”

25

30

”

35

40

“““’

45

50

55

60

65

WEEKS

FIG. 2. Weekly average aerosol concentrations as a function of time.

stage micrometer. The average value and standard deviation of the number of fibers per field were determined for each sample. These numbers were multiplied by the ratio of the filter area to the field area and divided by the number of cubic centimeters of aerosol in the sample. Table 4 shows the results of the four sets of fiber counts for each high- and low-dose level aerosol.The averages of the highand low-level aerosol fiber counts reflect the two dose groups. However, the highlevel aerosol fiber counts (which vary from about 120 fibers/cm3 for the first two observations to 47 and 30 fibers/cm3 for the second two observations) taken as a function of time cannot be compared because there does not exist any betweenfilter variation estimate as only one filter was counted at each sampling time. Only the estimate of the within-filter variation (field count variation) is known and that would be expected to be consistently less than the between-filter variation. Not too much importance should be attributed to the variation in the high-level counts because the data are derived from only four spot samples of short duration during the U-month exposure period. They were taken simply to provide an indication as to the magnitude of the fiber counts. Considerable variability in fiber counts must be expected even under standardized test conditions (Walton et al., 1976). Postmortem Procedures

For possible automated pulmonary morphometry with the scanning electron microscope in case of appreciable pulmonary fibrosis or emphysema development, the lungs, after weighing, were fixed by infusion of glutaraldehyde through the trachea at 25 cm of hydrostatic pressure in a fixation apparatus. Representative specimens of the remaining tissues were fixed in 10% neutral buffered formalin. The following hematoxylinand eosin-stained paraffin sections were histopathologically examined: left lobe and right diaphragmatic lobe of the lungs from each hamster; liver, heart, stomach, duodenum, jejunum, ileum, colon, cecum, and occasionally mesenteric lymph nodes from hamsters of the high exposure level (10 PgAiter); and control groups only. In addition, Masson trichrome, Wilder’s reticulum, and Perl’s iron stains of lung sections from four hamsters of

” MMAD

2.5 pm; GSD 2.4.

Stage upper limit diameter (wm) Stage mass (mg) Stage percentage Cumulative percentage smaller than upper limit 0.6 15.4 1.7 2.0

0.3

7

PARTICLE

0.4 2.9 0.3

S(Filter)

COMPOSITE

SIZE

TABLE

13.3

1.0 103.4 11.3

6

DISTRIBUTION

2

43.4

2.1 274.3 30.1

5

AEROSOL

66.0

3.1 206.0 22.6

4

78.0

4.8 109.1 12.0

3

CONCENTRATION)”

Andersen stage

(HIGH

92.8

87.2

51.0

5.6

11.6

1

7.0 84.1 9.2

2

100.0

65.9 7.2

0

z E 3 “:

E

$

0.4 2.5 1.7 1.7

0.6 1.4 1.0 2.7

7

PARTICLE

SIZE

TABLE

1.0 13.6 9.5 12.2

6

DISTRIBUTION

2.1 36.3 25.4 37.6

5

AEROSOL

MMAD

3.1 33.3 23.3 61.0

4

2.7

2 7.0 21.3 14.9 90.4

pm; GSD 2.4.

4.8 20.8 14.6 75.5

3

CONCESTRATIOX)”

Andersen stage

3 (Low

’ Composite of samples taken during exposure Weeks 24, 27, 30, 32, 38, 49, 54, and 67.

Stage upper limit (pm) Stage mass (mg) Stage percentage Cumulative percentage

8(Filter)

COMPOSITE

11.6 6.8 4.8 95.2

1

6.9 4.8 100.0

-

0

2 3: q E v)

2 z 5 z 2 ls OJ c:

z

% z 2v1

374

WEHNER

ET AL.

10

:

1 5

g-

-

ij Y 0 & d u

-

LO-

1

:

2

0.5-

MMAD 2.5 pm GSD 2.4

L----i0.1

1

5

M

50

en

95

99

99.9

CUMULATIVE PERCENT SMALLER THAN STATED SIZE FIG.

3. Composite particle size distribution

of the 14.6 &liter

(9.9 &liter

respirable) aerosol.

each of the two dose levels, sacrificed after I5 months of exposure, and of the control group, were examined. RESULTS

Body and Lung Weights The average body weights (ABWs) as a function of treatment and age for the high- and low-dose level subgroups that were exposed to AC dust for 15 months and those for the H-month control group are shown in Fig. 5. The weight curves for the other subgroups look similar and are therefore not shown here. Generally there were no significant differences among ABWs for groups sacrificed at 3 and 6 months, but there were significant differences among ABWs at several weighing points for animals sacrificed at 15 months. These differences usually occurred during the growth period of the animals (weighing points 2 through 8). During this growth period, the 10 or 1 &liter group ABWs, or both, were usually signiticantly higher than the controls. The 10 &liter group ABWs for animals which were exposed for 6 (recovered for 9 months) and 15 months were generally higher than the 1 pg/liter group ABWs during the animals’ growth period, but the 1 pg/liter group ABWs were generally higher than the 10 @liter ABWs for animals exposed for 3 months and recovered for 12 months. These inconsistencies among the various groups indicate that weight differences are not treatment related.

ASBESTOS

DUST

INHALATION

BY

375

HAMSTERS

10 -

: s4 = Y E 2 2 LOT f 8 5

0.1 -

MMAD 2.7 urn GSD 2.4

I 0.1

I 1

L 5 CUMULATIVE

I 20

I

I

I

I

I

50

80

95

99

99.9

PERCENTSMALLERTHANSTATEO

FIG. 4. Composite particle size distribution

SIZE

of the 1.7 PgIliter ( 1.3 &liter

respirable) aerosol.

Table 5 shows the ABWs and the average wet lung weights of the various hamster subgroups. The ABWs and average wet lung weights for animals sacrificed at the 3-month exposure point are not significantly different (at 5% level) for the three experimental groups. For animals sacrificed at the 6-month TABLE RESULTS OF FIBER COUNTS

4

IN ASBESTOS CEMENT

High-level aerosol (tiberslcm3)* Sample period Start of exposures 9th Exposure month 13th Exposure month 15th Exposure month

AEROSOL

SAMPLES”

Low-level aerosol (fibers/cm3)’

Average

SD

Average

SD

118 122 47 30

65 54 21 14

8.6 13.0 5.0 5.0

5.2 6.0 2.0 2.0

n To facilitate longer sampling times for more representative counts, air flow rates were reduced for the last two samples to avoid overloading of the filters. The low flow rates might have resulted in the loss of some of the larger fibers, thereby contributing to the low counts. * Total aerosol concentration, 14.6 &liter; respirable fraction, 9.9 &liter. c Total aerosol concentration, 1.7 &liter; respirable fraction, 1.3 &liter. One filter per sample period was counted. Standard deviations refer to the 60 to 100 fields counted on each filter. Only fibers longer than 5 pm and with diameters less than 3 Nrn were counted.

376

WEHNER

ET AL.

160

A CONTROLS 85 0

60

120

160

240

3w

360

420

480

OAYSONEXPOSURE

FIG. 5. Mean body weights of subgroups of hamsters exposed to asbestos cement dust for 15 months.

exposure point, the ABWs for the three groups are not significantly different, but the average wet lung weight for the high-dose level group is significantly higher than those of the low-dose level group and of the control group. At the 15month sacrifice point, there is no statistically significant difference between the weight data from the various subgroups and no consistent weight trend is apparent. Mortalit? Table 6 shows the number of hamsters alive at the time of final sacrifice. Eight hamsters were found dead during the first 6 months of exposure (five from the high-dose level group and three from the low-dose level group). The five dead hamsters from the high-dose level group were accidentally soaked with water from a malfunctioning drinking water nozzle. Two hamsters of the low-dose level group had diarrhea, and the cause of death was not determined in the third. Between the 6-month sacrifice point and the 15month final sacrifice, 27 hamsters were found dead. Ten of these hamsters were discarded without examination due to tissue autolysis, 11 had diarrhea, and the cause of death was not determined in the remainder. There is no statistically significant (P < 0.05) difference in the number of animals alive in each group at the time of the final sacrifice (Smonth exposure point). Histopatholog) The only histologic changes that could be related to the AC dust exposures were present in the lungs. Upon gross examination, the lungs were generally unremarkable, but the lungs of the high-dose level animals occasionally had a mottled appearance with slightly rusty-tan colors irregularly replacing the normal pink coloration of the lung. Histopathologic examination showed ferruginous bodies that were associated with alveolar macrophages (Fig. 6), distributed throughout the lung and occasionally aggregated in areas around respiratory bronchioles. The macrophages

ASBESTOS

DUST

INHALATION

TABLE AVERAGE

Exposure group

BY

377

HAMSTERS

5

BODY AND WET LUNG WEIGHTS OF THE HAMSTER SUBGROUPS AT THE 3-, 6-, AND IS-MONTH EXPOSURE POINTS

SACRIFICED

Lung weight as percentage of body weight (5%)

Body weight (8)

Lung weight (la

Sacrificed at the 3-month exposure point 10 &liter 108 ? 15” 1 &liter 103 + 15 Controls 104 k 18

1.06 t 0.07 0.91 k 0.16” 1.05 +- 0.22

0.96 c 0.18 0.90 2 0.13 1.00 I! 0.20

15 I5 16

1.03 + 0.18’

0.80 k 0.10’

16

0.99 t 0.19’

0.76 2 0.12’

I1

0.78 + 0.11

0.62 + 0.10

15

0.87 + 0.11 0.82 k 0.14

0.66 2 0.09 0.67 k 0. IO

15 15

1.77 + 0.23

1.09 k 0.39

II

1.79 + 0.32

1.21 2 0.16

15

2.05 c 0.47

1.35 k 0.23

10

1.55 + 0.27

1.10 c 0.24

12

1.77 2 0.33

1.17 2 0.14

12

1.88 It 0.41 1.75 k 0.31

1.10 + 0.44 1.22 k 0.19

12 12

Sacrificed at the 6-month exposure point 10 j@liter (6month exposure) 130 k 17 10 j&liter (3-month exposure 3-month recovery) 130 k 13 1 pg/liter (6-month exposure) 125 t 11 1 &liter (3-month exposure) 3-month recovery) 130 t 12 Controls 126 zt 15 Sacrificed at the 15month exposure point 10 &liter ( 15 month exposure) 147 2 17 10 pgiliter (&month exposure, Pmonth recovery) 149 ” 19 10 &liter (3-month exposure, If-month recovery) 151 2 22 1 j&liter (15-month exposure) 143 ” 15 1 pg/liter (6month exposure Pmonth recovery) 151 k 17 1 &liter (3-month exposure, 12-month recovery) 157 ? 19 Controls 144 k 13

N

” *SD. ’ Significantly lower than 10 &liter group and controls (P < 0.05, F test). ’ Significantly higher than 1 PgIliter group and controls.

phagocytosing the ferruginous bodies were generally aggregated in alveoli and frequently formed multinucleated cells. The ferruginous bodies measured up to 20 pm long and 2 pm thick, were frequently curved, stained yellow with hematoxylin and eosin, and had a typical beaded appearance characteristic of hemosiderin deposited on asbestos fibers. Numerous macrophages contained two or more

378

WEHNER

TABLE NUMBER

OF HAMSTERS

ALIVE

AT TIME

OF FINAL

ET AL

6 SACRIFICE

( 15-MONTH

EXPOSURE

PERIOD)

Duration of exposure” (months)

10 j@liter

1 pg/liter

Controlsb

15 6’ 3’

11 (6%) 15 (94%) 10 (63%)

12 (75%) 12 (75%) 12 (75%)

12 (75%)

Dose groupsb

DRefers to Dose groups only, not applicable to Controls. * Original number of animals per group = 16. c After the indicated number of exposure months, the hamsters of these subgroups were withdrawn from exposure and maintained for observation.

ferruginous bodies. Portions of the ferruginous bodies frequently appeared to protrude through the plasma membrane. Only rarely were nonphagocytosed ferruginous bodies present. Pulmonary fibrosis related to AC exposure was only apparent in the H-month exposure groups and did not exceed the grade of “slight.” Its incidence in these animals shows a significant (P< 0.01) trend toward higher incidence and severity as a function of dose level when the controls and the 1 and 10 /.&liter dose level groups are compared (Table 7). The fibrosis (Fig. 7) consisted of very small to

FIG.

6. Alveolar macrophage with ferruginous bodies from hamster E 1212. H&E;

x

1500.

ASBESTOS

DUST

INHALATION

TABLE

BY

379

HAMSTERS

7

INCIDENCE OF VERY SLIGHT TO SLIGHT PULMONARY FIBROSIS IN THE THREE EXPERIMENTAL HAMSTER GROUPS

Dose level Q.&liter) 0 (Controls) 1” 10”

Incidence of fibrosis 1114 4116 13115

Incidence (%) 7 25 87

” Exposed for 15 months. The incidence shows a significant (P < 0.01) trend when the controls and the 1 and 10 &liter dose level groups are compared.

small interstitial areas with increased numbers of collagen fibers and occasional fibroblasts. The alveolar septa occasionally appeared to have strands of collagen fibers separating vascular endothelium from alveolar epithelial cells. These areas of fibrosis were associated with aggregates of alveolar macrophages and ferruginous bodies, but were not forming any nodules or granulomas of grossly detectable size. Figure 8 shows a lung section from a control hamster sacrificed at the 15month exposure point. The role of inhaled AC dust as a causative agent of vesicular emphysema in the hamsters remains unclear. In hamsters sacrificed after 6 months of exposure, there was an increased incidence of very slight emphysema in the 10 pg/liter dose

FIG. 7. Slight degree of pulmonary fibrosis in hamster E 1212, exposed to 10 &liter cement dust for 11 months at which time the animal died. H&E; x 600.

asbestos

380

WEHNER

ET AL

FIG. 8. Lung section from a control hamster (E 3304). sacrificed at the 15-month exposure point. H&E: x 375.

group over the controls, but in the hamsters sacrificed after 15 months of exposure there was no significant (P < 0.05) difference in incidence between the exposed and control hamsters. More severe emphysema than seen in control hamsters was present in one hamster exposed to 10 /-&liter AC dust for 3 months and then withdrawn from exposure for 3 months (Fig. 9) and in one hamster exposed to 10 pg/liter AC dust for 15 months. These hamsters also had accumulations of macrophages with ferruginous bodies that could have accentuated a change occasionally seen in control hamsters. A lung section from a control hamster is shown in Fig. 10. There were no other lesions clearly related to AC dust exposure. Bronchiolization of alveolar epithelium occurred in both control and exposed hamsters. Pneumonitis, composed of very small to small interstitial areas with a mixture of inflammatory cells, was present occasionally in control and exposed hamsters. Other changes occasionally observed in the lungs without relationship to AC dust exposure (as they also occurred in controls) included congestion, edema, calciferous bodies in bronchioles, pigment deposition in alveolar macrophages, osseous metaplasia, bronchiectasis, hemorrhage, and vasculitis. The statistical analysis of pulmonary histologic changes comprised analysis of measurements of the following five variables: macrophages, ferruginous bodies, emphysema, fibrosis, and bronchiolization. Data from the animals that died due to faulty water nipples, or where autolysis precluded histologic examination, were not included in the statistical analysis.

ASBESTOS

DUST

INHALATION

BY

HAMSTERS

381

FIG. 9. Lung section of hamster E 1409, showing moderate emphysema. The animal had been exposed to IO &liter asbestos cement dust for 3 months with subsequent withdrawal from exposure for 3 months prior to sacrifice. H&E; x 15.

Nonparametric statistical methods were used to analyze the data because the histologic changes were graded, based on the degree of severity. Tests used to analyze the data were the Jonckheere-Terpstra test, the Kruskal-Wallis test, and the Wilcoxin test. The hypothesis being tested determined the test statistic to be used. The question of whether or not there was a dose effect was the first hypothesis tested. Null hypothesis is that there was no effect due to dose, versus the alternative hypothesis that the effect increased as the dose increased. Results of the Jonckheere-Terpstra test (Table 8) show that there was a significant increase in the number of macrophages and ferruginous bodies as the dose increased from 0 to 1 to 10 &liter for the 3-, 6-, and 15month exposures. Results to determine the effect of dose on emphysema and fibrosis are not as unambiguous because a significant effect for emphysema shows only for the 6month exposure data, and a significant effect for fibrosis was found only for the 15month exposure data. There was no significant effect of dose on bronchiolization. The next question of interest concerns the effect of the length of exposure. Null hypothesis is that there was no effect due to the length of exposure, versus the alternative hypothesis that the effect increased as the length of exposure increased. Results of the Jonckheere-Terpstra test (Table 9) show that there was a significant increase in the number of macrophages and ferruginous bodies as the length of exposure increased for dose levels of 10 and 1 pg/liter and exposure

382

WEHNER

ET AL

FIG. 10. Lung section from control hamster E 3211 sacrificed at the &month exposure point. H&E: x 17.

periods of 6 and 15 months. Again the results for emphysema and fibrosis are significant for only one of the tests. The last question involves recovery of an animal (reversibility of changes) after exposure to AC dust. An animal’s condition could improve or get worse. If there is no effect as a function of time in control animals, results of 3 months of exposure followed by 0, 3, and 12 months recovery can be compared using the Kruskal-Wallis test. Results of this test, given in Table 10, indicate that animals exposed for 3 months developed fibrosis during the recovery period while the controls did not. Also, the number of macrophages appears to have increased in the controls as a function of time. The question of recovery can be investigated further by comparing the 3-month exposure plus 1Zmonth recovery group with the U-month controls and comparing the 3-month exposure plus 3-month recovery group with the 6-month controls. Results of the two-sided Wilcoxin test are given in Table 11. Results of the test indicate numbers of macrophages found in the controls and in the exposed animals are closer together (but still significantly different) after a recovery period. This is due to an increase of macrophages in the controls rather than an improvement in the exposed animals. The significant differences in the number of

” Key:

N = normal.

slight

small

3 6 15

Bronchiolization

(or very

12 14 6

3 6 15

Fibrosis

I = very

13 12 16 15 13

6

15

16 14 14 16

3 6 1.5 3

Ferruginous bodies

Emphysema

9 10 3

3 6 15

N

Macrophages

Variable

Length of exposure (months)

amount)

TABLE

0

1 , Z = slight

0 0

0 0 0 0 0

0 0

0

0

0 01

3

ON HISTOLOGIC

amount).

5 2 5

3 3 0 0 4 1 0 0

0 0 0 0 0

0 0 6 0

0 70

13 14 8 11 14 9 0

2

1

Severity”

1 &liter

0 0 0

0 0 0 0 0

0 0 0 0

0 00

3

CHANGES

3 = moderate,

Level of exposure

8

(or small

11

10 13

12 13 16 15 12

1 16

1

5

3 11

N

OF EXPOSURE

0 0

0 0 0 0 0

2 2 0 0 1 4 1 7

0 0

1

0 0

0

0

0 11

2

LEVEL

0

7 94

1

Severity”

0 &liter

EFFECTS OF DOSE

12 14 8

7 12 15 15 2

0 0 0 I5

0 00

N

4 = marked,

3 2 5

5 = extreme.

0 0 0

0 0 0

0 0 7 0 0 2

0

0

0 0 2 0

0 30

3

1

0

8 16 13 0

7 0 0 0 9 2 0 0 6

9 16 12

2

6 00

1

Severity”

10 pg/liter

P < 0.001

P < 0.005

P < o.ooo1 P < 0.0001 P < 0.0001

P < 0.0001 P : i.2

15 6 15 6 15 6 15 6 15 6 15 6 15 6 15 6 15 6 15 6

10 10 1 1 10 10 1 1 10 10 1 1 10 10 1 1 10 10 1 1 5

1 5 0 2 1 3 1

7

7 3 I 13 7 3 6 13 2 0 1 1

1

2

0 0 0 0 0 0 0

2

4 10 6 0 6 10 1 0 0 0 0 0

3 months

13 9 16 12 13 11 15

0 1 1 3 1 1 7 3 12 13 13 15

N

TABLE OF EXPOSURE

Severity”

OF LENGTH

3 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0

3

CHANGES

0 0 1 1 0 0 1 1 12 7 12 12 3 15 10 15 7 14 10 13

N 1 0 8 14 2 0 11 14 3 9 1 3 12 1 3 0 8 2 3 2

1

Severity” 2 13 16 3 0 13 16 1 0 0 0 0 0 0 0 0 0 0 0 0 0

6 months

Length of exposure

ON HISTOLOGIC

9

1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

3

0 9 2 3 6

0 1 12 13 2

5 5

8 11

4

8

1

12

0

1

0

N

0

2

0

7

0

1

6

13

7

12

2

Severity”

15 months

0

0

0

0

0

0

0

2

0

3

3

(’ Key: N = normal, 1 = very slight (or very small amount) , 2 = slight (or small amount), 3 = moderate, 4 = marked, 5 = extreme.

Bronchiolization

Fibrosis

Emphysema

Ferruginous bodies

Macrophages

Variable

Study period (months)

Exposure level (&liter)

EFFECTS

< < < <