Document MGpo5agxVzqMwaraZb67VL0KM

gjg&if.-g.vgg'tefr Pergamon <4fw. ocrup. Hyg.. Vol. 39. No. 5. pp. 637-653. 1995 Elsevier Science Ltd British Occupational Hygiene Society Printed tn Great Britain 0003-4878,-95 S9.50-t-0.00 0003-4878(94)00091-3 CHRONIC INHALATION STUDIES OF MAN-MADE VITREOUS FIBRES: CHARACTERIZATION OF FIBRES IN THE EXPOSURE AEROSOL AND LUNGS Thomas W. Hesterberg, William C. Miiller, Philippe Thevenaz* and Robert Anderson Mountain Technical Center, Schuller International, Inc,, Littleton, CO 80127, U.S.A.; and `Research and Consulting Company, Geneva, Switzerland N (Received 9 May 1994) Abstract--Inhalation studies were conducted to determine the chronic biological effects in rodents of respirable fractions of different man-made vitreous fibres (MMVFs), including refractory ceramic fibre (RCF), fibrous glass, rock (stone) wool and slag wool. Animals were exposed nose-only, 6 h per jday, 5 days per week, for 18 months (hamsters) or 24 months (rats). Exposure to 10 mg m~3 of ocidolite or chrysotile asbestos induced pulmonary fibrosis, lung tumours and mesothelioma in !ats, thus validating the inhalation model with known human carcinogenic fibres. Exposure of rats to 30 mg m "3 of refractory ceramic fibres (RCF) also resulted in pulmonary fibrosis as well as significant increases in lung tumours and mesothelioma. In hamsters, 30 mg m'3 of RCF induced a 41% incidence of mesotheliomas. Exposure of rats to 30 mg m `3 of fibre glasses (MMVF 10 or 11) or of slag wool (MMVF 22) was associated with an inflammatory response, but no mesotheliomas or significant increase in the lung tumours were observed. Rock wool (stone wool: MMVF 21) at the same exposure level resulted in minimal lung fibrosis, but no mesotheliomas or significant increase in the lung tumours were observed. Fibre numbers (WHO fibres) and dimensions in the aerosols and lungs of exposed animals were comparable in this series ofinhalation studies. Differences in lung fibre burdens and lung clearance rates could not explain the differences observed in the toxicologic effects of the MMVFs. These findings indicate that dose, dimension and durability may not be the only determinants of fibre toxicity. Chemical composition and the surface physico-chemical properties of the fibres may also play,an important role. INTRODUCTION Man-made vitreous fibres (MMVFs) are a class of materials which have found many applications in both residential and industrial settings. MMVFs are fibrous inorganic substances that are made primarily from rock, clay, slag or glass. Sometimes referred to as man-made mineral fibres (MMMFs), the major classes of MMVF are refractory ceramic fibres (RCFs), fibrous glass, rock (stone) wool and slag wool. RCF, the smallest category of MMVF, represents.only about 1-2% of the world production. It is made by melting A1203 and Si02 in about equal amounts, or by melting kaolin clay, and then `spinning' or `blowing' this molten material into fibres. Most RCF is used as high-temperature furnace insulation. World production of RCF in 1990 was about 80 million lbs. Fibrous glass is the largest category of the MMVFs and is used in insulation, air handling, filtration and sound absorption. The thermal, acoustical and fire resistant properties of these products have led to their widespread jflhjn a variety of residential and commercial applications. Production of fibrous glass ^Hporth America in 1989 was approximately 1.8 million tons. Slag wool and rock wool are composed primarily of calcium, magnesium, aluminium and silica. Since 1975, most slag wool has been produced from the waste slag that resulted from the reduction 637 uU,*l.VV:-f>_. ...- ,f. ...........:, 4(1*4."* . : R_- tVl J ; S "**SS&I-i lfe._';;-J cfiyir' *i_;. DEFENDANT'S EXHIBIT ALL-STATE LEGAL SUPPLYCR HWBUI0007992 638 T. W. Heslerberg et al. of iron ore to iron. Rock wool fibres are made from basaltic rocks with additives such as limestone or dolomite. Slag wool and rock wool are used in residential and commercial low- and high-temperature insulation and in acoustical ceiling tiles and wall panels. In North America about 75% of slag wool production is used in acoustical ceiling tile manufacture. ' Animal studies have been conducted to assess the potential biological effects of MMVFs. This research has been reviewed by the International Agency for Research on Cancer (IARC, 1988), the International Programme on Chemical Safety (IPCS, 1988), the World Health Organization (WHO, 1984), and the U.S. Environmental Protection Agency (Vu, 1988). These reviews are consistent in the judgement that chronic inhalation studies of airborne fibres provide the best model for assessing the. potential risk to man (McClellan et al., 1992). This paper summarizes and compares the results from a recent series of chronic inhalation studies that evaluated the toxic effects of each of the major MMVF categories: fibrous glass, RCF, rock wool and slag wool. Its purpose is to examine the characteristics of the fibres in the exposure aerosols and in the lungs of the exposed animals, in order to understand more fully the critical fibre characterstics that are responsible for the differences in toxicides induced by the different MMVFs. Since RCF was the only MMVF type that induced lung tumours and this effect only occurred at the highest dose (30 mg m~3), this paper focuses on comparing fibre characteristics only at this highest dose. A more detailed comparison of the fibre characteristics at the lower doses will be addressed in another paper. The MMVFs used in this series ofstudies were prepared to be respirable by rats and to have dimensions comparable to those of fibres found in the workplace air (Hesterberg and Hart, 1995). Fibres were pre-selected for their size, and the actual size distributions of the aerosols used for fibre exposure were verified. Further, it was essential to document that fibre preparation, handling and aerosolization did not alter the physical-chemical characteristics of the fibre, since these are critical determinants of fibre toxicity. The methodology and results from these individual studies have been described in detail elsewhere (Hesterberg et al., 1993; Bunn et al., 1993; Mast et al., 1995; McConnell et al., 1994, 1995). MATERIALS AND METHODS r Fibres j Five respirable-size test fibres were prepared from MMVF products far these studies: a kaolin-based refractory ceramic fibre (RCF 1), two fibrous glass compositions (MMVF 10 and 11), a rock wool (MMVF 21) and a slag wool (MMVF 22) (see Table 1). In order to simulate fibre size distributions found in workplace air, these test fibres were size-selected from commercial fibre products by a water based separation technique that produces fractions of fine, respirable fibres. Positive control animals were exposed to intermediate length National Institute of Environmental Health Sciences (NIEHS) chrysotile asbestos (Hesterberg et al., 1993) or to size- selected crocidolite asbestos (McConnell et al., 1995). These two asbestos types were included in the studies to validate the animal model. Because the average dimensions of these two asbestos types were very different from those of the MMVFs it is not appropriate to compare them with the MMVFs on a fibre-by-fibre basis. HWBUI0007993 ` ``"E ^;wrv.''. - .?>*V'*' ' '"'W0.Ytfr :* nff&fJT 'v " " ` '* *' '' - . *V^'**--y 'i**tjirt4^y- $&*$&$&'&i ^bs: Chronic inhalation studies of MMVFs Table 1, Respirable test fibres that were size-selected from man-made mineral fibres (MMVF) used in a recent series of rodent inhalation studies conducted at Research and Consulting Company (RCC), Geneva, Switzerland RCF 1 MMVF 10 MMVF 11 MMVF 21 MMVF 22 Refractory ceramic fibre Fibreglass Fibreglass Rock wool (stone wool) Slag wool 639 Experimental design Tikis series of studies was initiated in mid-1988 at Research and Consulting Company in Geneva, Switzerland, The time-lines of these studies, with their various phases of exposure, recovery and analysis, are shown in Fig. 1. Six-week-old male Fischfer 344/N rats and Syrian golden hamsters were obtained from Charles River Laboratories. Hamsters were exposed to RCF or chrysotile asbestos for 18 months. In similar studies, rats were exposed for 24 months to RCF, fibrous glass, rock wool, slag >ol or crocidolite asbestos. A schematic showing the exposure and recovery phases as ,,eU as the disposition of animals during the rat inhalation studies is shown in Fig, 2. There were 140 animals in each of the MMVF-exposed and negative control (exposed to filtered air only) groups. Groups of three or six randomly selected animals from each exposure group were killed at 3, 6, 12, 18 and (rats only) 24 months to follow the progression of histopathological changes and to determine lung fibre burdens. An additional six `recovery' animals were removed from each exposure group at 3, 6, 12 and (rats only) 18 months and held without further treatment until the end of the exposure period, when they were killed to assess progression or regression of lung lesions and lung retention and clearance of fibres after cessation ofexposure. Following the exposure period, the remaining animals were held for lifetime observation, i.e. until \ MMVF Inhalation studies 1992 1993 ill Inhalation HI Recovery J23 AntlyiU i SSI ''**=? **' v.? t xi* '. -* - ' sM" I V 3. Time lines for the series ofstudies Ifgun ir. mki-1988 to assess the chronic inhalation effects ofthe three .jor categories of MMVF; fibre glass, refractory ceramic fibre (RCF) and mineral wool. Rats were exposal iit 24 months and held without further e-/- .sure (Recovery) until they reached about 20% survival when all remaining animals were killed. Hamsters were exposed (RCF only) for 18 months and also held without furthe. tsposure until 20% survival. HWBUI0007994 .rC 640 T. W. Hestcrberg et al. Chronic Inhalation udy Begin explore I Ttaeto*. Atiilk Hud exposure (24 months) f jLaag hiitopathology ind fibre buried | Fig. 2. Schematic of protocol for exposure and recovery groups in a typical chronic inhalation study in rats. Groups of 140 rats per dose group were exposed nose-only, 6 h per day, 5 days per week for 24 months. Periodically (3, 6, 12, 18 and 24 months) rats were killed and their lungs examined microscopically for pathological effects; the accessory lobe was frozen for fibre lung burden determination. Also at these time points (except 24 months), six mis per group were removal from exposure and held until 24 months for assessment of reversal of lung histological changes and for determination of fibre disappearance from the lung. All remaining animals were killed at about 20% survival, approximately 30 months. about 20% survival. This observation period was chosen because it has been shown in previous studies with asbestos that the induction ofmesothelioma does not occur until late in the animal's lifetime (Wagner et al., 1974). For calculating per cent tumour incidence, only animals that were exposed to fibres for at least 1 year were considered at risk for induction of neoplasms as this was the earliest time point a neoplastic finding was observed in this series of studies. Fibre aerosol exposure Animals were exposed by nose-only inhalation to the different MMVF types according to the method of Sachsse et al. (1976). Advanced techniques of fibre preparation and sizing, non-destructive aerosolization and fibre measurement, as described previously (Hesterberg et al., 1991; Bernstein et aL, 1994), were used. Aerosols of the test fibres were produced using the Research and Consulting Company, Geneva (RCC) fibre aerosol generation system (Bernstein et al., 1994), which produces large numbers of unbroken fibres with lower levels of non-fibrous dust than in the aerosols used in previous studies. Temperature, relative humidity and oxygen concentration were monitored continuously. The animals were confined separately in tubes which were positioned radially around the flow-past, nose-only exposure chamber (Cannon et al., 1983). This unique system provides positive pressure laminar flow to each animal individually, and unlike conventional nose-only exposure systems each animal is supplied fresh aerosol and the air exhaled by one animal does not reach any other animal in the chamber. Rodents were exposed in these chambers for 6 h per day, 5 days per week, for either 24 months (rats) or IS months (hamsters): the maximum dose was 30 mg m"J of each type of man-made fibre. Negative control animals were exposed similarly to filtered air. Positive controls were exposed to 10 mg m"3 either of crocidolite or of chrysotile HWBUI0007995 Chronic inhalation studies of MMVFs 641 asbestos. Fibre size distributions were determined on a quarterly basis using scanning electron microscopy. The numbers of fibres per cm3 were determined using the World Health Organization (WHO) criteria (WHO, 1985). Details of the fibre aerosol measurement techniques have been described elsewhere (Hesterberg et al., 1993). Aerosol monitoring and characterization Aerosol samples were collected on Millipore membrane filters at the animal exposure port. To determine the number of fibres cm"3 and eliminate isokinetic sampling bias sampling was isoaxial, directly from the output of one of the laminar flow feed tubes. Filters were then placed between glass slides and clarified for counting. Fibrlp numbers were determined using a Bausch and Lomb Balpan Phase Contrast micr jscope at a magnification of x 400. WHO Monograph 4 counting rules were appl ed for counting the number of WHO fibres cm"3 (WHO, 1985). During pre-exposure trials and once every 3 months thereafter, samples of the MMVF aerosols were captured on filters for determination of fibre length and diameter. Until analysis these samples were stored in glass bottles containing approximately 100 ml distilled water and 8 mg of sodium azide. To retain all fibres the "*cr surface was washed into the bottle, then filters were ashed and the ash was added jk to the fibre suspension in the bottle. The suspensions were diluted to 250 ml with distilled water and homogenized by sonification. Aliquots were filtered onto membrances for examination either by electron microscopy (for measurement of fibre diameters) or by phase-contrast optical microscopy (for measuremet of fibre lengths). Dimensions were determined according to the methods for measuring airborne man made mineral fibres outlined in WHO Monograph 4 (WHO, 1985). Diameters were measured at x 5000 either on a JEOL T 300 SEM or on a JEOL 840 SEM equipped with a Videoplan Image Analysis System. Lung pathology The lungs and attached mediastinal lymph nodes were removed in tow and weighed. The right accessory lobe was tied off, removed, weighed, frozen and stored at - 20C for lung burden analysis (see next section). The remaining lung lobes were then inflated to a pressure of 30 cm H20 with Kamovsky's fixative for 2 h. The lungs were examined under a dissecting microscope prior to fixation. Uniform sections of the left lung and of the right diaphragmatic lobe were embedded in paraffin, cut at a thickness of 4 mm and replicate sections were routinely stained with hematoxylin and eosin (H&E) and with Masson-Goldner's trichome stain for collagen staining. In addition, sections were made from all grossly visible lesions from that and other portions of the lung. Proliferative lesions' of the "pulmonary'parenchyma-were designated as bronchoalveolar hyperplasia (BAH), pulmonary adenoma or adenocarcinoma. Other types of lesions, including those in the pleura were noted where appropriate. Lung burden analysis Immediately after necropsy, the infracardiac (accessory) lobe of each animal's lung s frozen and later analysed for lung fibre burden using methods described by sterberg et al. (1993). Briefly, lung tissue was thawed, rapidly dehydrated with acetone and ashed using a low-temperature process. Fibres recovered were dispersed in distilled water, filtered onto membranes and examined using scanning electron J&HSHL 642 T. W. Hesterberg tt a!. microscopy. The number, dimensions and other characteristics of the lung fibres were determined. RESULTS Histopathology of rats Chrysolite and croddolite asbestos (10 mg m~3) induced pulmonary interstitial fibrosis in rats as early as 3 months after the,inhalation exposure was initiated. The exposure to crocidolite had to be stopped at 10 months because of excess mortality of the exposed animals. These animals were held without further croddolite exposure until the end of the study and no further excess in mortality was curved, Bach asbestos type induced a single mesothelioma and significant increases in lung tumours in rats (Table 2), thus validating the rat inhalation model for these disease end-points. After 3 months of exposure aU five types of MMVF produced an inflammatory response, including increased macrophage levels in the alveoli, bronchiolization of the alveolar ducts, and microgranuloma formation in the terminal airways. After 6 months of fibre exposure only RCF1 induced minimal pulmonary fibrosis (Table 3). This type of fibre also induced two mesotheliomas and a significant increase m lung tumours. Although rock wool (MMYF 21) produced minimal lung fibrosis after 18 months of exposure, no mesotheliomas were induced and no significant increase in lung tumours was observed. The two fibre glass compositions (MMVFs 10 and ll)and the slag wool (MMVF 22) produced neither pulmonary fibrosis nor mesotheliomas and no significant increase in lung tumours compared to the unexposed control animals. Table 2. Rats were exposed to 10 mg m"3 of NIEHS asbestos, either chrysolite or sire-selected long-fibre crocidolite. Exposure regime was the same as for man-made vitreous fibres, 6 h per day, 5 day per week. Chrysolite exposure continued for 2 years but crocidolite exposure was stopped at 10 months owing to elevated mortality Fibre group WHO fibres cm"s (x 104} Lung fibrosis Lung tumours Mesotheliomas Air control Chrysolite Crocidolite 0 uu 0.16 + 0,1 1-3% 0 + 13 (18.9%) 1 (1.4%) + 15 (14.2%) 1 (0.9%) Table 3. Rat Sung pathology induced by 30 mg m 3 of size-selected man-made vitreous fibres (MMVF): kaolin-based refractory ceramic fibre (RCF I), fibrous glasses (MMVF ID and MMVF 11) or mineral wools (MMVF 21 and MMVF 22) Fibre group Exposure WHO fibres cm*3 (30mgm`3) Lung fibrosis Lung tumours Mesotheliomas Air control RCF 1 MMVF 10 MMVF 11 MMVF 21 MMVF 22 0 187 + 53 232 + 56 246 76 243 + 67 213 62 i-:*% 0 + 16(13.0%)* 2(1.6%) -- 7 (5.9%) 0 -- 3 (2.7%) + 5 (4.4%) 0 0 -- 3 (2.6%) 0 Tumour incidence significantly elevated above air controls (P<0.05). HWBUI0007997 -^pssa-" . Chronic inhalation studies of MMVFs Histopathology of hamsters In hamsters, chrysotile asbestos induced pulmonary fibrosis, but no mesotheliomas or lung tumours (Table 4). In contrast, RCF 1 induced a 41% incidence of mesotheliomas in addition to pulmonary fibrosis. As with chrysotile, RCF 1 did not induce lung tumours in the hamster. The results of this study are reported in detail elsewhere (McConnell et at., 1994), Aerosol fibre characterization The average WHO respirable fibre cm"3 levels that resulted from exposure to 30-mg m'3 of each MMVF type are shown in Table 3. The WHO (World Health Organization) criteria define a respirable fibre as having an aspect ratio of >3:1, a length > 5 /mi, and a diameter < 3 pm. The length distributions of the MMVF types in the jaerosols were comparable, with the possible exception of MMVF 10, which appeared to have more fibres cm "3 in the shorter length categories as shown in Figj 3(a). The diameter distributions of the MMVF types in the aerosols were also comparable, although MMVF 10 appeared to have slightly more thick fibres than the other MMVF types [Fig, 3(b)], It was shown in a previous publication (Hesterberg et 'll.. 1993) that because the size separation procedure used in preparing RCF 1 was .tferent from that used to prepare the other MMVFs, there were from 5- to 10-fold more non-fibrous particulates in the RCF aerosols than in the aerosols of the other MMVFs, Lung fibre burden analyses, rats The dimensions of fibres recovered from the lungs of animals after 13 weeks of exposure to 30 mg m"3 of the different MMVFs are shown in Fig. 4, The length distributions of fibres recovered from lungs [Fig, 4(a)] are much closer to one another than those found in the aerosols [Fig, 3(a)], Many of the fibres longer than 20 ptm were not found in the lung, either because they did not reach the pulmonary region or because they were breaking in the lung. The diameter distributions of all MMVFs in \the lung were also closer to one another [Fig, 4(b)], and many of the large diameter aerosol fibres {>1.5 pm diameter) were not found in the pulmonary region of the lung. Quantification of the number of WHO fibres in the lung at different times during the exposure phase of the study shows that lung fibre levels of the different MMVF types were comparable [Fig. 5(a)], This was also true when the number of fibres in the lung > 10 pm in length were compared at different times during the exposure phase of the study [Fig. 5(b)]. However, the number of lung fibres >20 pm in length were higher at the later time points for RCF 1 and MMVF 21 (rock wool) than for the other two fibre types [Fig. 5(e)].' Table 4. Hamster lung pathology induced by 10 mg m-i size-selected kaolin-based refractory ceramic fibre (RCF 1) or 10 mg m`3 of chrysotile asbestos Fibre group Exposure WHO fibres an** Lung fibrosis Lung tumours Mesotheliomas Air control RCF 1 Chrysotile 0 2!556 3<B0I4G0 + 4- 00 0 42(41%) 00 HWBUI0007998 644 T. W, Hesterberg al. (a) Aerosol fibre lengths (b) Aerosol fibre diameters Fig, 3. (a) Length distributions and (b) diameter distributions of fibres in the exposure aerosols of th five different MMVFs in the chrome rat inhalation studies. Fibres were collected on filters at animal nose- jnly exposure ports, Fibre lengths were determined using phase-contrast optical microscopy, while ibre diameters were determined using scanning electron microscopy. . `Recovery animals' were removed from inhalation treatment at each time point (13, 26, 52 and 78 weeks) and held without further treatment until the end of the 2 years, thus providing groups of animals representing a range of recovery periods of from 13 to 78 vveeks. Figure 6 shows the lung burdens and clearance of WHO fibres in rats exposed to 30 mg m ~3 of fibre glass (MMVF 11). Animals that were killed soon after 78, 52,26 or 13 weeks' exposure (solid bars) tended to maintain a fairly constant lung burden whereas those permitted to recover (striped bars) demonstrated a recovery time-dependent decrease in lung fibre burden. The per cent of WHO fibres, fibres i 1 Chronic inhalation studies of MMVFs Lung fibre lengths 645 (b) Lung fibre diameters Diameter category Fig. 4. (a) Length distributions and (b) diameter distributions offibres from the longs ofrati exposed for 13 weeks to the five dUTerent MMVFs in the chronic inhalation studies. To permit clearance of the upper airways, rats were killed a minimum of 24 h after the exposure was stopped! 0K right tcowwry lobe was frozen and later low-temperature ashed for fibre recovery. Fibre tagthg Were determined using phase contrast optical microscopy, while fibre diameters were determined using scanning electron microscopy. > i 0 pm in length and fibres > 20 /rm in length retained in the lung over time for each MM VF type is shown in Fig. 7. For WHO fibres and fibres > 10 pm in length, the two r,K,re glass compositions (MMVFs 10 and 11) appear to be cleared more slowly from lung than KCF 1. For these two length categories, rock wool and slag wool were reared most rapidly from the lung. When the lung clearance of fibres >20 pm was examined, a different picture emerged [Fig. 7(e)]: (i) the long fibres (>20 urn) of all 644 T, W, Hesierberg et at. Lung burden during exposure (a) Fig. 5 (a). Lung burden during exposure: fibres >10pm (b) Fig. 5 (b). HWBUI0008001  " -HAfaX/i' -. * sragS'&iyt ' Chronic inhalation studies of MMVFs Lung burden during exposure: fibres >20pnt (c) .it'A-: -: :.<", 647 r- ' 'e 5 Lung burdens of (a) WHO fibres, (b) fibres > 10 /rm in length and (c) fibres >20 pm in length per mg lung tissue from the lungs of rats continuously exposed to 30 rag m"3 of the five different MMVFs, Rats re killed at least 24 h after the exposure was stopped, the right accessory lobe was frozen and later lowtemperature ashed for fibre recovery. Fibre retention during exposure and recovery 35 Jfc3cfi >> 'Q m2 io ^ M& m o 30 2S - 20 - 15 10 Exposure Recovery 78/26 52/32 26/78 _ , Exposure/recovery .. (weeks).* 13/91 ' 'i. Lung burdens ofWHO fibres per mg dry lung tissue from the lungs ofrats exposed to 30 mg ru"5 fibre (MMVF 11). Solid bars indicate lung burdens of animals continuously exposed. To permit upper . ,,-y clearance rats were killed at least 24 h after the exposure was stopped. Striped bars indicate lung burdens ofanimals removed from exposure for various recovery periods. The right accessory lobe was frozen and later low-iemperature ashed for fibre recovery. HWBUI0008002 648 T, W, Hesterberg et at Lung clearance; WHO fibres (a) Lung clearance: Fibres >10pm (b) * HWBUI0008003 Chronic inhalation studies of MMVFs Lung clearance: Fibres >20jim (c) 649 ^ig, 7. Disappearance of (a) WHO fibres, (b) fibres > 10 pm in length and (c) fibres > 20 pm in length from ; 1 ungs ofanimals held without exposure for different recovery times. Both length categories ofthe two fibre ...lasses (MMVFs 10 and 11) appeared to clear from the lung more slowly than RCF and the two mineral wools. The percentage of fibres retained in the lungs was calculated by taking the ratio of lung fibres in recovery animals divided by the number of fibres in continuously exposed animal lungs (see Fig. 6). For fibres >20 nm in length, RCF cleared the most slowly, followed by MMVFs 10 and 21. MMVF types appeared to clear more rapidly than the shorter fibres; and (ii) the long fibres of RCF 1 cleared more slowly than MMVFs 10 and 21. The most rapidly cleared long fibres were MMVFs 11 and 22. DISCUSSION x Previous inhalation studies of fibre glass using rodents agree with the findings of the Research and Consulting Company (RCC) studies. Fibre glass has been tested by inhalation in guinea pigs (Gross et al, 1970), hamsters (Lee et al, 1981; Smith et al, 1987) and rats (Gross et al, 1970; Lee et al., 1981; Wagner et al., 1984; McConnell et al, 1984; Mitchell et al., 1986; Muhle et al., 1987; Le Bouffant et al, 1987; Smith et al, 1987). In spite of FG lung burdens in excess of several hundred thousand fibres mg"1 dry lung tissue none of these studies Identified a significant increase either in fibrosis or in neoplasms following glass fibre inhalation. In three of the above studies, the chronic inhalation toxicity of rock and slag wool were alsbexamincd (Wagner et al.', fl984; Le Bouffant et al, 1987; Smith et al, 1987).`As was seen with fibrous glass, aB three studies demonstrated no tumorigenic response by this route of exposure;- '<.'' .? The results from two previous RCF inhalation studies (Davis it all, 1984; Smith et al, 1987) differ from the more recent RCC studies presented here. Davis et al (1984) reported RCF exposure of rats resulted in an average of 5% pulmoriary fibrosis, -Imonary tumours in eight of 48 rats, and one peritoneal mesothelioma. The lower rosis and tumour response in the Davis study may have resulted from the lower exposure concentration used, 8.4 mg m"3 compared to 30mgm"3 in the present study. In addition, the use of fibres that were not presized, the use of whole-body 650 T. W, Heslerberg et al. exposure, or the fibre generation technique, which may have crushed some of the fibres, could account for the lack of consistency with the present study. Smith et al. (1987) exposed hamsters and rats to RCF at 200 fibres cm ~ 3f 6 b per day, 5 days per week, for 24 months. The rat study showed no significant increase in neoplasms and minimal pulmonary fibrosis in 22% of the exposed animals. In the hamster study, RCF produced only one mesothelioma in 50 animals and no fibrosis was observed. It is difficult to explain why there was little response to RCF in the Smith studies, but it may be related to the different aerosol and exposure technology used or to the low exposure level, 12 mg m~3 compared to 30 mg m~3 in the present study. Previous intracavitary injection studies do nocagrec with the findings of the inhalation studies: intracavitary, injection of nearly, afi compositions of MMVF has resulted in serosal cancer in animals (IARC, 1988; Vu, 1988). However, injection of fibres bypasses the normal defence mechanisms of the lung and can produce abnormal fibre distribution, fibre clumping and overload doses (Hesterberg et al, 1991). Furthermore, when fibres are injected into the pleura or peritoneum of an animal, leaching, degradation, fragmentation or any other transformations are unlikely to be the same as they are after inhalation. The weaknesses of intracavitary injection studies of fibrous materials limit their relevance for human risk assessment (IPCS, 1988; Vu, 1988; NIEHS, 1989). . N The primary purpose of the RCC inhalation studies was to provide information which would be useful in predicting possible human health risks ofexposure to different compositions of MMVFs. These studies also present a unique opportunity to gain a greater understanding of the fibre characteristics that are the critical toxicological determinants, because the experimental design resulted in pulmonary region deposition of large numbers of the different fibre types having comparable dimensions. The major focus of the present paper is to characterize the fibres in the exposure aerosols and in the lungs of animals chronically exposed to the different compositions of MMVF, and to attempt to determine whether fibre dose, dimension or durability could explain the observed differences in lung and pleural pathology. As shown in Table 3, the aerosol exposure concentrations of the different MMVFs could not explain the differences in their toxicity to the lung. Although the exposure to RCF (WHO fibres cm-3) was the lowest of the five fibre types, it was the only MMVF that produced an elevation in lung tumours and mesothelioma. Neither could the dimensions of the RCF fibres in the aerosol explain its greater toxicity to the lung, si ice the length and diameter distributions of RCF were comparable to those of the ot ler MMVF types (Fig. 3). It should, however, be kept in mind that the RCF fibre preparation contained more non-fibrous particulates than the other MMVF types. What effect this mighbftave had on the pathological effects of RCF is not as yet known. It was previously shown that exposure of rats to non fibrous Ti02 and asbestos significantly enhanced the induction of lung tumours and mesotheliomas compared to rats exposed to asbestos alone (Davis et al, 1991). Lung deposition patterns of the five MMVF types could not explain differences in pathogenicity; lung burdens were similar for each of the test fibres. In each case, the fibres that were recovered from the rat lung had even narrower length and diameter ranges than the aerosol fibres (Fig, 4). It is valid to assume that most of the fibres recovered from the lungs were in the pulmonary region or interstitium, because the animals were killed at least 24 h after the exposure was stopped, allowing clearance of HWBUI0008005 imn- Chronic inhalation studies of MMVFs 651 the upper airways by mucociliary mechanims. The numbers of WHO fibres and fibres > 10 pm in length in the lung were similar for each of the different MMVF types [Figs 5(a) and (b)]. However, greater numbers of long fibres (>20 pm long) were found in the lungs of rats exposed to RCF 1 and MMVF 21 (rock wool) than in those exposed to the other fibre types [Fig. 5(c)], Even though lung concentrations of long MMVF 21 fibres were higher than those of long RCF 1 fibres, lung fibrosis occurred much later for MMVF 21 (18 months vs 6 months for RCF 1) and no mesotheliomas or significant increase in lung tumours were observed for MMVF 21. This indicates that the lung pathogenic potential of a fibre may be determined by more than its dose and dimensions. Examination of the lung clearance patterns for the different MMVFs revealed some interesting findings. First, it is clear that the long fibres (>20 jam long) disappeared more rapidly than shorter fibres (Fig. 7). Perhaps the long fibres were breaking and contributing to the short fibre populations. Secondly, lung clearance patterns for the five MMVFs could not account for differences in the pathogenicities of the test fibres. Figures 7(a) and (b) show the per cent of WHO fibres and of fibres > 10 pm in length retained in the lung after various recovery periods for the five MMVFs. The two fibre classes (MMVFs 10 and 11) took longer to disappear or to clear from the lung than did 2F, perhaps because the lung macrophage clearance mechanisms were more mhibited in the fibreglass exposures. However, the long fibres (>20pm long) disappeared more slowly from the RCF 1-, MMVF 21- and MMVF 10-exposed rats than from rats exposed to the other fibre types. In general, this agrees with the greater accumulation of long fibres in rats exposed to RCF 1 and MMVF 21 [Fig. 5(c)], Perhaps MMVF 10-exposed rats did not accumulate high levels oflong fibres because the aerosolized fibre was somewhat shorter and thicker than the other fibre types (Fig. 3). The fact that RCF 1 and MMVF 21 disappeared from the lung at similar rates provides further support for the contention that the pathogenicity of a fibre is dependent upon more than simply the dose, dimension, and the durability of the fibres in the lung. \ In a previous paper on this series of studies it was reported that fibre glass and RCF reached similar levels in the lung during continuous exposure and appeared to clear from the lung at comparable rates (Hesterberg ei al, 1994). Fibres recovered from the lungs of animals exposed for only 13 weeks and held without further exposure for 91 weeks (recovery animals) were analysed for chemical change using energy-dispersive spectroscopy (EDS) with scanning electron microscopy. These analyses of recovery lungs showed that much of the alkalis and alkaline earth components had leached from the fibre glass over time. However, only a slight change'in RCF chemiitry was observed. These findings indicate that the leaching offibres and the resultant change in chemical composition, especially the surface chemistry, :may be an important determinant of the biological activity of MMVFs.' The1 importance of chemical composition to the toxic potential of fibres has been recently reviewed (Guthrie and Mossman, 1993). In addition, fibres that are more rapidly leached of their alkalis and alkaline earths may be more readily broken in the lung, which might explain why the more teachable fibre compositions show the most rapid disappearance of long fibres. >re studies are required to determine if a fibre's ability to be leached is a critical . eterminant of its ultimate toxicity to the lung. It is interesting to note that, in the RCC study series, chrysotile asbestos produced (:??r^vjv-..; -; ; & sVtw?#*m '*',- V.'V* , ; :: ;; iSSi HWBUI0008006 I 652 T. W, Hesterberg et al. no lung cancer or mesotheliomas in hamsters, but in rats it produced the entire spectrum of asbestos-associated lung disease observed in humans, including pulmon ary fibrosis, lung tumours and mesothelioma. For this reason, if a single species is chosen for assessing the human health risk offibre inhalation, the rat would be the most appropriate choice. This is in agreement with the consensus opinion of a workshop of fibre toxicology experts (McClellan et al., 1992). However, because of its greater pleural mesothelioma response after RCF inhalation (41%), the hamster has been recommended as a second species for evaluating the human health risk of fibres (Vu and Peat-field, 1993* Vu,-1994).' " ' - ; '' ` '". REFERENCES Bernstein, D. B., Mast, R., Anderson, R., Hesterberg, T, W., Musselman, R., Kamstrup, O, and Hadley, 1. (1994) An experimental approach to the evaluation ofthe biopersistence ofrespirable synthetic fibres and minerals. In Biopersistenee ofRespirable Synthetic Fibres and Minerals (Edited by Bignon, J,, Saracci, R. and Touray, J.-C.). Bunn, W. B,, Bender, J,, Hesterberg, T. W,, Chase, G. R, and Konsen, J. (1993) Recent studies of man-made vitreous fibers. J. occup. Med. 35,101-113. Cannon, W.C., Blanton, E, F.and McDonald, K. E. (1983) The flow-past chamber: An improved nose-only exposure system for rodents. Am. ind. Hyg. Ass. J. 44,923-928. \ Davis, J. M. G., Addison, J., Bolton, R. E., Donaldson, K., Jones, A. D. and Wright, A. (1984) The pathogenic effects of fibrous ceramic alummuin silicate glass administered to rats by inhalation or peritoneal injection. In Biological Effects of Man-made Mineral Fibres, Vol. 2, pp. 303-322. World Health Organization, Copenhagen. ,Davis, J. M. G., Jones, A. D. and Miller, B. G. (1991) Experimental studies in rats on the effects of asbestos inhalation coupled with inhalation of titanium dioxide or quarts. Int, J. ExpL Path. 72 501-525. ,Gross, P., Kaschak, M,, Tolker, E. B,, Babyak, M. A. and De Treville, R. T. P. (1970) The pulmonary reaction to high concentrations of fibrous glass dust. Archs Environ. Hlth 23 67-76, Guthrie, G. D,, Jr and Mossman, B. T. (Editors) (1993) Health Effects of Mineral Dusts. Reviews in Mineralogy, Vol, 28. Mineralogical Society of America, Washington, DC. Hesterberg, T. W. and Hart, G. A. (1995) A comparison ofhuman exposures to fiber glass with those used m a recent rat chronic inhalation study. Regulat. Pharm. Toxicol, (in press). Hesterberg, T, W., McConnell, E. E,, Miiller, W, C., Bernstein, D. M,, Mast, R. and Anderson, R. (1994) Relationship between lung biopersistence and biological effects ofman-made vitreous fibres after chronic inhalation in rodents. In Biopersistence ofRespirable Synthetic Fibres and Minerals (Edited by Bignon, J., Saracci, R. and Touray, J.-C.). Hesterberg, T. W., Miiller, W.C., McConnell, E. E., Chevalier, J,, Hadley, J., Bernstein, D, M.,Thevenaz, P. and Anderson, R. (1993) Chronic inhalation toxicity of size-separated glass fibers in Fischer 344 rats. Fundam. Appl, Toxicol, 20,464-476. , Hesterberg, T. W,, Vu, V,, Chase, G. R,, McConnell, E. E., Bunn, W. B. and Anderson, R. (1991) E sc of animal models to study man-made fiber carcinogenesis. In Cellular and Molecular Aspects of nber Carcinogenesis (Edited by Brinkly, B,, Lechncr, J. and Harris, C.), pp. 183-205. Cold Spring Hi rbor Laboratory, Cold Spring Harbor, New York. \ IARC (1988) IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Man-made Mi leral Fibres and Radon. International Agency for Research on Cancer, Lyon, France, IPCS (1988) International Moghunmc on Chemical Safety Criteria Document. Man-made Mineral Fibers. Environmental Health Criteria Document 77. World Health Organization, Geneva, Switzerland. \ Le Bouffant, L., Daniel, H,, Henin, J. P., Martin, J. C., Nonnand, C, Thichoux, G. and Trolard, F. (1987) Experimental study on long-term effects of inhaled MMMF on the lung of rats. Ann. occup. Hyg. 31, 765-790. Lee, K. P,, Barras. E.. Griffith, F. D,, Waritz, R. S. and Lapin, C. A. (1981). Comparative pulmonary responses to inhaled inorganic fibers with asbestos and fiber glass. Environ. Res. 24, 167-191. Mast, R. W,, McConnell, E. E., Anderson, R,, Chevalier, J., Kotin, P., Bernstein, D. M., Thevenaz, P,, Glass, L. R,, Miller, W, C, and Hesterberg, T. W. (1995) Chronic toxicity (inhalation) of refractory ceramic fiber in male Fischer 344 rats. Inhal. Toxicol, (in press). ,McClellan, R. O., Miller, F. J,, Hesterberg, T. W,, Warheit, D, B,, Bunn, W. B. et at. (1992) Approaches to evaluating the toxicity and carcinogenicity of man-made fibers. Regulat. Toxicol. Pharmac. 16 321-364, McConnell, E. E.. Kamstrup, O., Musselman, R,, Hesterberg, T. W., Chevalier. J., Miiller, W. C. and HWBUI0008007 Thevenaz, P. (1995) Toxicity study of size-separated rock and slag wool insulation fibers in Fischer 344/N rats. Inhal. Toxicol, (in press). McConnell, E. E., Mast, R. W,, Hesterberg, T. W,, Chevalier, J,, Kotin, P,, Bernstein, D. M,, Thevenaz, P,, Glass, L. R. and Anderson, R, (1994) Chronic inhalation toxicity of a kaolin based refractory ceramic fiber (RCF) in Syrian golden hamsters. Inhal. Toxicol. 6, 571-614. McConnell, E. E,, Wagner, J. C,, Skidmore, J. W. and Moore, J. A. (1984) A comparative study of the fibrogenic and carcinogenic effects of UICC Canadian chrysotile B asbestos and glass microfibre (JM 100). In Biological Effects ofMan-made Mineral Fibres, Vol. 2, pp, 234-252. World Health Organization, Copenhagen. Mitchell, R, 1., Donofrio, D. j. and Moorman, W. J. (1986) Chronic inhalation toxicity offibrous glass in rats and monkeys. J. Am. Coll. Toxicol. 5, 545-575, Muhle, H., Pott, F., Bellmann, B,, Takenaka, S, and Zicm, U. (1987) Inhalation and injection experiments in tats to test the carcinogenicity of MMMF. Ann. occup. Hyg. 31,755-764. NIE^IS (1989) NIEHS workshop on fibre toxicology research needs. National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina, ' ' Sachs se, K., Ullman, L. and Zbinden, K. (1976) Toxikologische Prufungen von Aerosolen in T erexperiment>Oiem, Rdschous 29, No. 38, Suite 1. Smltf, B, M,, Ortiz, L. W,, Archuleta, R. F. and Johnson, N, F. (1987) Long-term health effects in hamsters ai d rats exposed chronically to man-made vitreous fibres. Ann. occup. Hyg. 31, 731-754. Vu, f. (1988) Health hazard assessment of non asbestos fibers. Office of Toxic Substances. U.S. Environmental Protection Agency. Vi. V. (1994) Assessment of the potential health effects of natural and man-made fibers and their testing needs: perspectives of the U.S. Environmental Protection Agency. In Proceedings of Toxic and 'arcinogenic Effects of Solid Particles in the Respiratory Tract. V. and Dearfield, K, L. (1993) Biological effects of fibrous materials in experimental studies and related regulatory aspects. In Fiber Toxicology (Edited by Warheii, D.). Academic Press, New York. "A'-ij-ner, J. C,, Berry, G. B., Hull, R. J., Munday, D. E. and Skidmore, J. W. (1984) Animal experiments with MMM(V)F-Effects of inhalation and intrapleural inoculation in rats. In Biological Effects ofMan-made Mineral Fibres, Vol. 2, pp. 209-234. World Health Organization, Copenhagen. Wagner, J. C,, Berry, G. B , Skidmore, J. W. and Timbrell, V, (1974) The effects ofthe inhalation of asbestos in rats. Br. J. Cancer 29, 252-268. WHO (1984) Biological Effects ofMan-made Mineral Fibers, Vol. 2. World Health Organization, Regional Office, Copenhagen Denmark. WHO (1985) Reference methods for measuring man-made mineral fibers (MMMF). Prepared by WHO/EURO Technical Committee for evaluating MMMF. World Health Organization, Regional Office, Copenhagen, Denmark.