The Carcinogenicity of WHO Fibers of Silicon Carbide: SiC Whiskers Compared to Cleavage Fragments of Granular SiC

Abstract

A steep dose-response relationship for tumors was reported after application of 0.05 to 25 mg of whiskers of SiC in 1991 in a chronic intraperitoneal injection study in rats. In contrast, doses of 250 and 1000 mg of granular SiC did not cause tumors. The whiskers contained 107,000,000 fibers longer than 5 μm, per milligram while the granular sample was assumed to be free of such fibers. However, it has since been realized that even granular SiC may contain fibrous fragments that fulfill the definition of WHO fibers. It was the objective of our study to reanalyze the quantity of WHO fibers that were administered with the granular SiC. In addition, we examined whether the potency per WHO fiber was different for the SiC fragments and the SiC whiskers. Samples of the original granular and fibrous SiC were suspended in water and filtered. One half of each filter was analyzed by scanning electron microscopy (SEM, magnification ×2500), and transmission electron microscopy (TEM, ×10,000) was also performed for the SiC whiskers. The concentration of WHO fibers was 58,000 fibers/mg for the granular sample compared to 48,000,000 (SEM) and 42,000,000 (TEM) fibers/mg for the whiskers. The aspect ratio of the WHO fibers exceeded 10/1 for only 3.3% of the fragments but in each analysis for 96% of the whiskers. In addition, 0% of the fragments compared to 44% and 30% of the whiskers were more than 10 μm long. In total, 15 and 58 × 10⁶ WHO fibers were injected with the granular SiC even though only 0.8% and 0% tumors were recorded. However, 20.1% and 43.3% tumors would have been expected if the carcinogenic potency were the same for the fragments and for the whiskers. We therefore conclude that carcinogenic potency is a function of the shapes of the WHO fibers and is much lower for SiC fragments than for whiskers. Hence carcinogenicity mainly is restricted to a subgroup of WHO fibers longer than about 10 and thinner than about 1 μm. This conclusion is in line with the optimal exposure index, which recently has been proposed by the U.S. Environmental Protection Agency (EPA) for asbestos.