* Health Effects Laboratory Division and
Division of Respiratory Disease Studies, National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, 1095 Willowdale Road, Morgantown, West Virginia 26505
Received October 4, 2000; accepted December 15, 2000
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: silica; sand; quartz; coal slag; garnet; specular hematite; iron oxide; staurolite; lung; inflammation; fibrosis..
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
One approach to the problem of lung disease in the abrasive blasting industry is the replacement of sand with safer abrasive blasting agents. Although sand is the most popular abrasive blasting agent, a number of substitutes for sand are in current use. A 1992 survey reported that coal slag was the most frequently used of these substitutes (Paumanok Publications, Inc., 1992). Unfortunately, the toxicity of most commercially available substitutes is incompletely investigated. In this study, we examined pulmonary alterations in rats 4 weeks after intratracheal instillation of respirable fractions of blasting sand or 5 abrasive blasting substitutes: coal slag, garnet, specular hematite, staurolite, or treated sand (blasting sand coated with a dust suppressant). The toxicity of these abrasive blasting substitutes was compared to the toxicity of blasting sand and to the pulmonary response to the instillation of the vehicle, phosphate-buffered saline.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Test Materials
Coal slag.
"Black Beauty®" coal slag was purchased from Reed Minerals (Highland, IN). The manufacturer reports that in a typical analysis, the most abundant compounds are silicon dioxide, aluminum oxide, ferric oxide, calcium oxide, potassium oxide, magnesium oxide, and titanium dioxide, which respectively account for 47.2%, 21.39%, 19.23%, 6.80%, 1.60%, 1.47%, and 1.01% of the material (Reed Minerals, 1992). The typical free-silica content is stated to be less than 0.1% (Reed Minerals, 1992
).
Garnet.
Garnet was purchased from Emerald Creek Garnet Co. (Fernwood, ID). This abrasive blasting agent is a naturally occurring mixture of almandite garnet (Fe3Al2(SiO4)3) with some magnesium and manganese substitution for iron. Other components include mica and less than 0.1% crystalline silica (Gorrill, 1996). The supplier's typical chemical analysis noted that the major components were silicon dioxide, ferric oxide, aluminum oxide, magnesium oxide, calcium oxide, and manganese oxide, which respectively comprised 36.79%, 32.70%, 25.51%, 3.08%, 1.15%, and 1.01% of the material.
Sand.
Blasting sand (precision fractionated sand 2340) was obtained from Waupaca Materials (Waupaca, WI) and was identified as containing crystalline silica. This blasting sand contained 49% quartz as measured by x-ray diffraction.
Specular hematite.
"Bar Shot 50" specular hematite or iron oxide was purchased from Barnes Environmental, Inc. (Waterdown, Ontario, Canada). The manufacturer's material safety data sheet lists the components of this blasting agent as 9899% Fe2O3 with less than 0.1% crystalline silica (Barnes Environmental, Inc., 1996).
Staurolite.
"Starblast" staurolite was purchased from DuPont Chemical Company (Wilmington, DE). According to the typical analysis sheet, "Starblast" is about 86% staurolite minerals (FeAl5Si2O12OH), 6% titanium minerals, 3% zircon, 2% kyanite and less than 5% quartz (DuPont Chemicals, 1993).
Treated sand.
Magnum Blast 3.0 treated sand (blasting sand coated with a dust suppressant) was obtained from Fairmont Minerals (Wedron, IL). The supplier's material safety data sheet lists the major ingredient in their "prevent coated silica" as quartz, which comprises more than 90% of the ingredients in a typical analysis (Fairmont Minerals, 1999).
Elemental Analysis
Each test agent was subjected to elemental analysis for aluminum, arsenic, barium, beryllium, cadmium, calcium, chromium, cobalt, copper, iron, lead, lithium, magnesium, manganese, molybdenum, nickel, phosphorus, platinum, quartz, selenium, silver, sodium, tellurium, thallium, titanium, vanadium, yttrium, zinc, and zirconium. The percent quartz in the samples was determined by x-ray diffraction (NIOSH Method NMAM 7500); arsenic, beryllium, cadmium and lead were analyzed by graphite furnace (NIOSH Methods NMAM 7901, 7102, 7048 and 7105, respectively); and the remaining elements were analyzed by atomic emission spectroscopy (NIOSH Method NMAM 7300) as previously described (NIOSH, 1994). Analyses were performed by Data Chem Laboratories, Inc. (Salt Lake City, UT). Results of this analysis are given in Table 1
.
|
Clearly, the bulk samples, as obtained by the manufacturers, were too large for direct use in pulmonary experiments. The absence of a significant small-particle component, made direct sieving or sedimentation inappropriate. Further, given the observation of inclusions (which would certainly be released in the blasting process), size reduction was required. Samples were initially ball milled and the respirable fraction was enriched by liquid sedimentation of the milled particles.
Figure 1 shows scanning electron microscope images of the bulk garnet sample (Figs. 1A and 1B
) and the size-classified garnet sample (Figs. 1C and 1D
). Although there is roughly a 3 log order size difference between the bulk and size-classified sample, similarities in cleavage characteristics can be observed.
|
|
|
|
The garnet samples have been used here to illustrate the comparison of features between the bulk, air, and laboratory prepared samples. Similar relationships were seen with each of the other abrasives.
Intratracheal Instillations
For intratracheal administration, rats were briefly anesthetized with methohexital (Brevital, 37.5 mg/kg). The test agents were suspended at a concentration of 33.3 or 8.33 mg/ml in sterile Dulbecco's phosphate-buffered saline (Sigma Chemical Company, St. Louis, MO) and sonicated for 12 min. Rats were exposed to 10 or 2.5 mg blasting agent/rat by instillation of 0.3 ml of the concentrated or diluted suspension, respectively. The earliest timepoint for significant fibrosis in silicosis is at 4 weeks after instillation (Driscoll et al., 1990b), the exposure duration selected for our study. The 10 mg exposure was based upon previous studies which showed that 50 mg/kg of crystalline silica (Min-U-Sil) in 180200 mg rats (
10 mg/rat) was the minimum dose producing histopathologic evidence of pulmonary fibrosis in rats 4 weeks after instillation (Driscoll et al., 1990b
). The 2.5 mg exposure was designed to provide dose-response data. A 0.3 ml volume of the particle suspension or vehicle (phosphate-buffered saline) was administered via the trachea using a 20-gauge 4-inch ball-tipped animal feeding needle (Perfectum, New Hyde Park, NY). The test agents were administered once. Six rats were exposed to each concentration of each blasting agent except for specular hematite. Due to the limited amount of available specular hematite, 4 rats were each exposed to 0.3 ml of the 33.3 mg/ml suspension of specular hematite (10 mg/rat).
Collection of Biological Samples
Rats were deeply anesthetized using an overdose of intraperitoneal pentobarbital 4 weeks after exposure. After assuring deep anesthesia, the abdominal aorta was transected. The left lung lobe was then briefly ligated.
With the left lobe ligated, a tracheal cannula was inserted and the right lung inflated with 3 ml of chilled Ca2+/Mg2+-free phosphate-buffered saline (PBS; BioWhittaker, Walkersville, MD). This procedure was repeated using 4 ml volumes of PBS until a 40 ml total volume of bronchoalveolar fluid (BALF) was obtained. The BALF was centrifuged at 500 x g at 4°C for 10 min. The first lavage sample was centrifuged separately from the subsequent lavage samples with the supernatant collected for later analyses. After removal of supernatant, lavage cells from each rat were combined and resuspended in HEPES-buffered solution (145 mM NaCl, 5 mM KCl, 10 mM HEPES, 1 mM CaCl2, 5.5 mM dextrose, pH 7.4).
Immediately after completion of bronchoalveolar lavage, the right lung lobe was ligated and removed below the ligature for hydroxyproline analysis. The temporary ligature on the left lung lobe was then removed, and the left lung lobe was perfused via the trachea with 3 ml Carnoys solution.
Cellular and Biochemical Assays of BALF
The BALF cell counts and differentials were performed on a Coulter Multisizer II cell counter (Coulter Electronics, Hialeah, FL) with a cell-sizing attachment to distinguish between rat lavage cell types as previously described (Castranova et al., 1979). The polymorphonuclear leukocyte (PMN) designation excludes alveolar macrophages but includes some lymphocytes.
The zymosan-stimulated chemiluminescence assay was conducted using 5 x 105 alveolar macrophages in a total volume of 0.25 ml HEPES buffer (145 mM NaCl, 5 mM KCl, 10 mM HEPES, 1 mM CaCl2, 5.5 mM dextrose, pH 7.4) as previously described (Porter et al., 1999). Zymosan-stimulated CL was calculated as the counts per minute (c.p.m.) in the assay of zymosan-stimulated cells minus the c.p.m. in the assay of resting cells.
Before storage at 30°C, an aliquot of first BAL fluid was removed for analysis of lactate dehydrogenase (LDH) activity. LDH activity was determined spectrophotometrically using a Cobas Fara II analyzer (Roche Diagnostic Systems, Montclair, NJ) and a previously described technique (Ma et al., 1999).
BALF albumin concentration was assessed colorimetrically at 628 nm based upon bromcresol green binding (albumin BCG diagnostic kit, Sigma Chemical Company, St. Louis, MO) using a Cobas Fara II Analyzer (Roche Diagnostic Systems, Montclair, NJ).
For determination of the hydroxyproline content of the lung, the lavaged right lungs of the rats were liquified by mincing and then hydrolyzed in 6 N HCL for 4872 h at 110°C. Hydroxyproline was measured by a previously published method (Kivirikko, 1967).
Histopathology
Carnoy's-fixed left lungs were embedded in paraffin and stained with Masson's trichrome stain for examination of collagen formation by light microscopy. The slides were scored for the severity and distribution of fibrosis by a board-certified veterinary pathologist blinded to exposure status. Semi-quantitative numeric values were assigned to these scores so that severity was scored as: 0 = none; 1 = minimal; 2 = mild; 3 = moderate; 4 = marked; or 5 = severe. Tissue distribution was converted to numeric values with: 0 = none; 1 = focal; 2 = locally extensive; 3 = multifocal; 4 = multifocal and coalescent; and 5 = diffuse. The total fibrosis score was the sum of the severity and distribution scores.
Statistics
In general, data are presented as means ± SE of 46 separate experiments. Findings from rats exposed to different abrasives were compared to results in the control group or to the blasting sand group, respectively, using the Wilcoxon rank sum test. The Bonferroni adjustment was used to account for multiple comparisons; p-values without the Bonferroni adjustment were also noted in cases where the individual result was statistically significant before adjustment. Nonparametric methods were necessary to account for data which were not normally or log-normally distributed and for heterogeneity of variance among groups, which violates the assumptions of relevant parametric tests such as Dunnett's test. Significance was set at p 0.05.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The persistent PMN response associated with pulmonary exposure to the abrasive blasting agents may be associated with persistent pulmonary cytotoxicity. Blasting sand, garnet, treated sand, and coal slag at the 10 mg exposure dose caused significant elevation in LDH, a cytosolic protein which leaks from cells after membrane damage or destruction (Drent et al., 1996). In addition, exposure to 2.5 mg coal slag also caused elevation in LDH. Furthermore, staurolite (10 mg) resulted in a 91% increase in LDH above control. Since serum levels of LDH are higher than LDH levels in the lung, serum leakage into the lung as a source of the elevated LDH was excluded by determining that the serum protein, albumin, was not significantly elevated by exposure to any of the blasting agents studied. These findings are consistent with the hypothesis that garnet, coal slag, sand, staurolite, and treated sand continue to damage lung cells 4 weeks after intratracheal instillation of these abrasive blasting agents. Coal slag is the most frequently used substitute for sand in abrasive blasting (Paumanok Publications, Inc., 1992
), and it was more inflammatory and caused greater cell damage than blasting sand.
The significant elevation in zymosan-stimulated macrophage chemiluminescence associated with exposure to coal slag (2.5 and 10 mg), staurolite (10 mg) and garnet (10 mg) is consistent with macrophage activation after exposure to these compounds. Exposure to 10 mg of blasting sand or 10 mg of treated sand resulted in a greater than 8- or 6-fold elevation, respectively, in zymosan-stimulated macrophage chemiluminescence (significant when not adjusted for multiple comparisons).
Persistently elevated BAL PMN, increased BAL LDH activity, and enhanced zymosan-stimulated chemiluminescence in macrophages each suggest persistent effects from abrasive blasting agents 4 weeks after instillation. However, the most significant disorder observed in the abrasive blasting industry is pulmonary fibrosis. Biochemical confirmation of pulmonary fibrosis, hydroxyproline elevation, was seen only in rats exposed to coal slag. However, hydroxyproline elevation might not be expected in subchronic disease because the normal lung contains substantial concentrations of fibrous connective tissue which undergo daily remodeling (Marshall et al., 1997). In the presence of active ongoing cytotoxicity, preexisting structural collagen may undergo destruction while new fibrous connective tissue forms. Although unresolved inflammation and tissue destruction are important stimulants for fibrosis in any tissue (Cotran et al., 1999
), all of the microscopic foci of fibrosis associated with exposure to any of the agents investigated in the present study were considered minimal. This is most likely due to the relatively short exposure time in this study. Indeed, we designed our experiment for the earliest time point producing significant histopathologic evidence of fibrosis induced by a similar concentration of crystalline quartz in rats (Driscoll et al., 1990b
). Thus, additional studies will be needed to assess the final result of the unresolved inflammation seen in this subchronic study.
In contrast to the other abrasive blasting agents, specular hematite (10 mg) produced no significant alterations in BAL levels of LDH, numbers of lung PMN, macrophage chemiluminescence, the amount of pulmonary hydroxyproline, or fibrotic score. Since the major component of specular hematite is iron oxide (Barnes Environmental, Inc., 1996), these findings are consistent with the low toxicity of iron oxide in most rat studies (Stokinger, 1984). A recent study in humans also suggests that the initial inflammation associated with intrapulmonary instillation of iron oxide resolves rapidly after exposure (Lay et al., 1999
).
The intratracheal instillation technique used in our study can accentuate the pulmonary toxicity of particles with low toxicity, particularly at higher exposures such as the 10 mg dose used in our study, but generally does not affect the relative toxicity of different particles (Driscoll et al., 2000). We selected this method of exposure because it assured control of the dose to the lung in this screening study and allowed prioritization of those agents with the greatest pulmonary toxicity for follow-up studies. Because we have no evidence of pulmonary toxicity under the conditions of this study for one of the abrasives, specular hematite, we believe additional pulmonary toxicity studies may be warranted on coal slag, garnet, staurolite, and treated sand. While only 4 rats were exposed to the specular hematite, responses on all measures of pulmonary inflammation were numerically similar and statistically indistinguishable from the response seen in the 6 saline control rats. Since particle size and particle count/mass were similar for all the abrasive materials studied, the responses seen in rats exposed to 10 mg of the other abrasive blasting agents may not be the simple result of exposure to a very large number of particles or "particle overload." However, other investigators have seen subtle, persistent pulmonary inflammation after intratracheal instillation of the nuisance dust, titanium dioxide, at concentrations of 50 mg/kg (Driscoll et al., 1990a
).
We have used intratracheal instillation at 2 different doses to screen this group of abrasive blasting agents for pulmonary toxicity. The use of intratracheal instillation for such screening studies has recently been reviewed and is accepted (Driscoll et al., 2000). However, it should be noted that pulmonary fibrosis resulting from the instillation of crystalline silica is a progressive disease and can be induced in 90 days by the intratracheal instillation of as little as 5 mg/kg silica or 1 mg per 200 g rat (Driscoll et al., 1990b
). Since persistent inflammation plays an important role in pulmonary fibrosis, it is certainly possible that some of the agents that cause persistent inflammation in this 4-week study may produce pulmonary fibrosis in a longer study. Because of the ability of coal slag to induce more cell damage and greater PMN influx than a commercial blasting sand and to elevate the lung hydroxyproline concentration in our study, it is possible that coal slag is as fibrogenic as the commercial blasting sand used in our study. Thus, there is a need for additional studies of longer duration comparing the extent of pulmonary fibrosis produced by exposure to low doses of abrasive blasting agents.
Previous studies of abrasive blasting agents have usually compared a standard crystalline silica sample, e.g. Min-U-Sil, with potential substitutes for blasting sand (Stettler et al., 1988, 1995
). In our study, we examined the toxicity of several commercial abrasive blasting agents, including commercial blasting sand, since human disease in the abrasive blasting industry is associated with exposure to blasting sand rather than pure crystalline silica. Because surface properties and surface contaminants of silica are determinants of pulmonary toxicity, crystalline silica toxicity may not reflect the toxicity of commercial blasting sand (Bolsaitis and Wallace, 1996
; Le Bouffant et al., 1982
). Indeed, data from the present study indicate that the inflammatory and toxic potencies of the blasting sand sample used in our study were substantially lower than that for Min-U-Sil (Blackford et al., 1997
). The quartz content of the blasting sand used in this study was 49% while that for Min-U-Sil is 98%. The range for the quartz concentration in 16 samples of commercially available blasting sands is 39100% (Greskevitch, personal communication, analyses conducted by Data Chem Laboratories, Salt Lake City, UT). Thus the crystalline silica content of blasting sand appears to determine the pulmonary toxicity.
Overall, the pulmonary response 4 weeks after intratracheal instillation of 10 mg respirable blasting sand indicated continued cytotoxicity and very early pulmonary fibrosis consistent with the known role of blasting sand in occupational pulmonary fibrosis. Intratracheal instillation of specular hematite did not cause significant pulmonary fibrosis, cytotoxicity, or persistent inflammation; this is consistent with the general paucity of functional and inflammatory pulmonary changes in workers and research subjects exposed to iron oxide (Lay et al., 1999; Teculescu and Albu, 1973
). The persistent cytotoxicity generally observed after exposure to the other commercial substitutes for blasting sand was accompanied by persistent inflammation, and with coal slag by elevated lung hydroxyproline. This suggests the need for additional inhalation studies of abrasive blasting agents.
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
NOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Blackford, J. A., Jones, W., Dey, R. D., and. Castranova, V. (1997). Comparison of inducible nitric oxide synthase gene expression and lung inflammation following intratracheal instillation of silica, coal, carbonyl iron, or titanium dioxide in rats. J. Toxicol. Environ. Health 51, 203218.[ISI][Medline]
Bolsaitis, P. P., and Wallace, W. E. (1996). The structure of silica surfaces in relation to cytotoxicity. In Silica and Silica-Induced Lung Diseases. (V. Castranova, V. Vallyathan, and W. E. Wallace, Eds.), pp. 7989. CRC Press, Boca Raton.
Castranova, V., Bowman, L., and Miles, P. R. (1979). Transmembrane potential and ionic content of rat alveolar macrophages. J. Cell Physiol. 101, 471479.[ISI][Medline]
Cherniack, R. M., Banks, D. E., Bell, D. Y., Davis, G. S., Hughes, J. M., King, T. E., Schwartz, M. I., Waldron, J. A., Watters, L. C., Cox, J. B., MacIntyre, N. R., Piantadosi, C. A., Ramage, J. A., Barkman, H. W. Jr, deShazo, R. D., Glindmeyer, H., Morgan, J. E., Calhoun, W. J., Christman, J. W., and Diem, J. E. (1990). Bronchoalveolar lavage constituents in healthy individuals, idiopathic pulmonary fibrosis, and selected comparison groups. Am. Rev. Respir. Dis. 141, S169S202.[ISI][Medline]
Cotran, R. S., Kumar, V., and Collins, T. (1999). Pathologic Basis of Disease, 6th Ed. W.B. Saunders, Philadelphia.
Drent, M., Cobben, N. A. M., Henderson, R. F., Wouters, E. F. M., and van Dieijen-Visser, M. (1996). Usefulness of lactate dehydrogenase and its isoenzymes as indicators of lung damage or inflammation. Eur. Respir. J. 9, 17361742.
Driscoll, K. E., Costa D. L., Hatch, G., Henderson, R., Oberdorster, G., Salem, H., and Schlesinger, R. B. (2000). Intratracheal instillation as an exposure technique for the evaluation of respiratory tract toxicity: Uses and limitations. Toxicol. Sci. 55, 2435.
Driscoll, K. E., Lindenschmidt, R. C., Mauerer, J. K., Higgins, J. M. and Ridder, G. (1990a). Pulmonary response to silica or titanium dioxide: Inflammatory cells, alveolar macrophage-derived cytokines, and histopathology. Am. J. Respir. Cell Mol. Biol. 2, 381390.[ISI][Medline]
Driscoll, K. E., Maurer, J. K., Lindenschmidt, R. C., Romberger, D., Rennard, S. I., and Crosby, L, (1990b). Respiratory tract responses to dust: Relationships between dust burden, lung injury, alveolar macrophage fibronectin release, and the development of pulmonary fibrosis. Toxicol. Appl. Pharmacol. 106, 88101.[ISI][Medline]
DuPont Chemicals. (1993). Material Safety Data Sheet, Staurolite Products. Du Pont Chemicals, Wilmington, DE.
Fairmont Minerals. (1999). Material Safety Data Sheet, Magnum Blast 3.0. Fairmont Minerals, Wedron, IL.
Gorrill, L. E. (1996). Material Safety Data Sheet, Emerald Creek Garnet Abrasive Grains and Powders. Emerald Creek Garnet Co., Fernwood, ID.
Kivirikko, K. I., Laitinene O., and Prockop, D. J. (1967). Modifications of a specific assay for hydroxyproline in urine. Anal. Biochem. 19, 249255.[ISI][Medline]
Lay, J. C., Bennett, W. D. Ghio, A. J., Bromberg, P. A., Costa, D. L., Kim, C. S., Koren, H. S., and Devlin, R. B. (1999). Cellular and biochemical response of the human lung after intrapulmonary instillation of ferric oxide particles. Am. J. Respir. Cell Mol. Biol. 20, 631642.
Le Bouffant, L., Daniel, H., Martin, J. C. and Bruyere, S. (1982). Effect of impurities and associated minerals on quartz toxicity. Ann. Occup. Hyg. 26, 625634.[ISI][Medline]
Ma, J. Y. C., Barger, M. W., Hubbs, A. F., Castranova, V., Weber, S. L., and Ma, J. K. H. (1999). Use of tetandrine to differentiate between mechanisms involved in silica- versus bleomycin-induced fibrosis. J. Toxicol. Environ. Health A 56, 247266.
Marshall, R. P., McAnulty, R. J. and Laurent G. J. (1997). The pathogenesis of pulmonary fibrosis: Is there a fibrosis gene? Int. J. Biochem. Cell Biol. 29, 107120.[ISI][Medline]
NIOSH (1994). NIOSH Manual of Analytical Methods, 4th ed. U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health. DHHS Publication NIOSH 94113.
Paumanok Publications, Inc. (1992). The U.S. Market for Blasting Abrasives19921997 Analysis. Paumanok Publications, Inc., Shoreham, NY.
Porter, D. W., Castranova, V., Robinson, V. A., Hubbs, A. F., Mercer, R. R., Scabilloni, J., Goldsmith, T., Schwegler-Berry, D., Battelli, L., Washko, R., Burkhart, J., Piacitelli, C., Whitmer, M., and Jones, W. (1999). Acute inflammatory reaction in rats after intratracheal instillation of material collected from a nylon flocking plant. J. Toxicol. Environ. Health A 57, 2545.[ISI][Medline]
Reed Minerals. (1992). Material Safety Data Sheet, Black Beauty® Abrasives. Reed Minerals, Highland, IN.
Stettler, L. E., Proctor, J. E., Platek, S. F., Carolan, R. J., Smith, R. J., and Donaldson, H. M. (1988). Fibrogenic and carcinogenic potential of smelter slags used as abrasive blasting substitutes. J. Toxicol. Environ. Health. 25, 3556.[ISI][Medline]
Stettler, L. E., Salomon, R. A., Platek, S. F., Moorman, W. J., Clark, J. C., Krieg, E. F., and Phipps, F. C. (1995). Fibrogenic potentials of coal slags used as abrasive blasting substitutes. J. Toxicol. Environ. Health. 45, 349365.[ISI][Medline]
Stokinger, H. E. (1984). A review of world literature finds iron oxides noncarcinogenic. Am. Indus. Hyg. Assoc. J. 45, 127133.[ISI][Medline]
Teculescu, D., and Albu, A. (1973). Pulmonary function in workers inhaling iron oxide dust. Int. Arch. Arbeitsmed. 31, 163170.[ISI][Medline]