Investigation of the biodurability of wollastonite and xonotlite.

The in vivo durability of wollastonite materials, coated and uncoated, and of xonotlite was tested. Wollastonite is an anhydrous natural silicate and xonotlite is a hydrated synthetic calcium silicate. UICC crocidolite was used as a positive control with high durability. Using a dry-sizing technique, fractions from the stock materials were prepared according to the definition of "thoracic particulate mass" and "respirable particulate mass" of the American Conference of Governmental Industrial Hygienists. Fibers were instilled intratracheally into female Wistar rats, and the evenness of their distribution in the lung was checked by scanning electron microscopy (SEM). After serial sacrifices at 2 and 14 days, 1, 3, and 6 months, and low temperature ashing of the lung, the fibers were analyzed by SEM. The number and size distribution of fibers were investigated. The total number of crocidolite fibers decreased with a half-time of 240 days, but the number of fibers > 5 microns in length was unchanged after 6 months. The elimination kinetics of wollastonite fibers from the lung were relatively fast, with half-times of 15 to 21 days. The coating of wollastonite in Wollastocoat had no effect on this elimination process. For the thoracic fraction of wollastonite, the elimination from the lung was as fast as for the respirable particulate fraction. The elimination kinetics of xonotlite from the lung was very fast. This material consisted of single crystals of acicular morphology with a median length of 1.3 micron and of agglomerates of these crystals. More than 99% of single crystals and about 85 to 89% of the agglomerates were already eliminated 2 days after instillation.


Introduction
The durability of fibers in the lung is one important criterion for assessing their possible carcinogenicity. In a previous biodurability study of three samples of wollastonite with sacrifice dates at 2 and 30 days, half-times of about 15 days were observed (1). These half-times are very short compared to approximately 40 days found for special glass fibers (2) that showed no carcinogenicity in the intraperitoneal test (3). In the present study specific fractions of commercial fiber samples of wollastonite with and without surface modification and of xonotlite, a synthetic mineral of similar composition, were investigated.
Biological durability studies should use the fraction of fibers that could be deposited in lungs of humans. In designing the fiber sizing, the definition of "thoracic particulate mass" and "respirable particulate mass" was taken from published threshold limit values (4).
It has been estimated that about 95% of lung cancer in man originates in the tracheobronchial region (5 in this area are most likely to be relevant for health effects (6), so that the two fractions of fine airborne fibers that would be able to penetrate the lung had to be separated from the coarse material.

Materials and Methods
Wollastonite is a naturally occurring calcium metasilicate (CaSiO3) which, on milling in a rotating plate attrition mill, gives acicular fragments. The wollastonite samples in this study were NYAD G wollastonite, a high aspect ratio product, and NYAD 1250, which has a fine particle size grade with lower aspect ratio. NYAD G Wollastocoat 1001, and NYAD G Wollastocoat 2075 (NYCO Inc., Willsboro, NY) were manufactured from NYAD G by surface modification. The test material for xonotlite (Ca6Si6O17(OH)2) was a commercial product (Promat, Netherlands, trademark Promaxon). The samples for analysis consisted of single crystals of acicular morphology with a median length of approximately 1.3 pm, and of agglomerates of these crystals. UICC crocidolite was used as positive control with high durability. To get a thoracic particulate mass and respirable particulate mass of the test materials, sizing was done in two steps. In the first step the material was aerosolized by a combination of a dust-feeding system and a high pressure, high-velocity dispersion nozzle into a buffer chamber where big clumps of powder were removed. The airborne fibrous material then was routed into a two-stage, heavy grain load impactor, described elsewhere (7). Depending on the operating conditions of this device, the total aerosol was split into two fractions, coarse and fine, at a preset value of the aerodynamic diameter of the fibers. The noninhalable coarse particles were collected on impaction plates impregnated with vacuum oil, leaving the fine particle fraction to be sampled downstream of the separator by a membrane filter. The fine material For each sample a small fraction was suspended in double-distilled water, sonicated and filtered on to a Nuclepore filter (pore size 0.2 or 0.4 pm). A part of the filter was mounted on an aluminum stub and sputtered with approximately 30 nm of gold. These samples were analyzed by a Cambridge Stereoscan 360 scanning electron microscope (SEM), with a magnification to enable the measurement of both the longest and the thinnest fibers with sufficient precision. The length and the diameter were measured of some 400 particles of each sample of sized materials. The length, diameter, and calculated aerodynamic diameter (8) are presented in Table 1.
Two milligrams of fibers, suspended in 0.3 ml of 0.9% NaCl solution, were instilled intratracheally in a single dose into the lung of a female Wistar rat with a body weight of about 200 g. Five animals per group that had received either wollastonite or xonotlite were sacrificed after 2 days, 2 weeks, 1, 3, and 6 months; those that had received crocidolite were sacrificed after 2 days and 6 months. After sacrifice, the lungs were isolated, oven-dried at 1050C and subjected to lowtemperature ashing, which did not influence the size distribution of test materials. This was confirmed by comparing lung ash samples from the lungs of rats two days after intratracheal instillation with the original test materials. A fraction of the lung ash was suspended in filtered water and filtered on a Nuclepore filter (pore size 0.2 or 0.4 pm) within 15 min. These samples were prepared and analyzed by SEM as described for the characterization of the test materials. For each lung ash sample 200 fibers were measured on SEM videoprints and the total number of fibers per lung was calculated. For samples with low fiber content, 50 SEM screens were analyzed. The size distribution of the fibers was also analyzed. From the shape of the fibers, the volume of the particles was estimated assuming cylindrical geometry. Clearance kinetics were calculated from a regression analysis of the logarithm of number or mass of fibers versus time after instillation for individual animals.

Results
Examination of the distribution of fibers in the lung by SEM 2 days after intratracheal instillation showed fibers in the main bronchi, on the epithelium of the distal segments of bronchioli, and in alveoli; no large agglomerations of fibers were found. The results indicated the evenness of fiber distribution in the lung.
For all wollastonite test materials, the initial mass found in the lung 2 days after instillation was approximately 0.6 to 2 mg calculated from the shape and density of particles. Six months later the relative mass of test material was 0.02 to 0.5% of the initial lung burden. The logarithmic plot of the number or mass of wollastonite fibers versus sacrifice date (Figures 1-3), shows that the elimination of fibers follows first order kinetics, which can be defined by only one parameter, the half-time. For all wollastonite test materials in this study, calculated half-times were between 15 and 24 days ( For crocidolite the number of fibers >5 pm in length was unchanged during 6 months after intratracheal instillation.
Two days after instillation of xonotlite, very few fibers were observed in the lungs. injected test materials were d4 the thoracic fraction, and 0.( alveolar fraction. Since the fibers or agglomerates was v even zero, in animals sacrific and 1 month after instillation five animals were used for regre sis. No fibers or agglomei detected at the 3-month sac] Estimated half-times for the eli the fibers were 4 to 9 days. It v sible to calculate a half-time fr( rapid 2-day clearance.

Discussion
The half-times for the eliminat lastonite and xonotlite sample etectable in short compared both to crocidolite fibers )8% in the and man-made mineral fibers (2,9). These number of results confirm the assumption that lack of ery low, or carcinogenicity after intraperitoneal injeced 14 days tion of 100 mg wollastonite in rats could i, means of be explained by low durability in vivo (10).
:ssion analy-The half-times for the elimination of rates were wollastonite fibers and xonotlite crystals rifice date. are much shorter than the unimpaired mination of retention half-time of insoluble spherical vas not pos-particles such as toner or PVC particles, for om the very which half-times of approximately 80 days were reported (11). This suggests that mechanical clearance mediated by macrophages could be only of minor tion of wolimportance for wollastonite, for which diss were very solution of fibers must be the important clearance process. Since wollastonite is acid labile, dissolution would be fastest in the phagolysosomes of the macrophages, where 0(a) the pH is approximately 4.8 (12). Two days after instillation, the number of wollastonite fibers in the lung was about 100 to 200 x 106. The total number of macrophages in a rat lung is about 15 x 106 (13), giving a mean of approximately 10 wollastonite fibers per macrophage. Because the fraction of fibers >10 pm in length is 10% or less, most of the fibers could be phagocytized completely by macrophages. For all the wollastonite samples in this study, the half-time for elimination of the mass of fibers was very similar to that for total number of fibers, but the decrease of mass from day 2 to day 14 was higher for the thoracic fractions of all the wollastonite 150 200 samples than for the corresponding alveolar fractions ( Figure 3). This effect could be explained by mucociliary clearance of large (t) particles with an aerodynamic diameter (a) >10 pm, which were found only in the thoracic fractions. A fast decrease of the aerodynamic diameter from day 2 to day 14, detected only for the thoracic fractions, tends to confirm this hypothesis. The decrease of number of fiber >5 pm in length is faster than the decrease of total number of fibers for all wollastonite samples up to 1 month (Figures 1, 2). That could be due to breakage of fibers, which would tend to increase the number of fibers <5 pm in length.   why these fibers did not induce lesions in the lung, whereas attapulgite, chrysotile, and Fiberfax fibers did induce lesions 1 month after intratracheal instillation of 1, 5, or 10 mg/lung in rats (14).

Conclusion
The elimination of wollastonite fibers from the lung followed first order kinetics, and was predominantly due to the dissolution of fibers, with very short half-times of 15 to 21 days. The elimination of xonotlite crystals from the lung was even faster. The relatively fast dissolution of the test materials, which are all composed of a "calcium silicate" base, should minimize the health effects related to respired fibers.