Steviol glycoside safety: is the genotoxicity database sufficient?

The safety of steviol glycoside sweeteners has been extensively reviewed in the literature. National and international food safety agencies and approximately 20 expert panels have concluded that steviol glycosides, including the widely used sweeteners stevioside and rebaudioside A, are not genotoxic. However, concern has been expressed in recent publications that steviol glycosides may be mutagenic based on select studies representing a small fraction of the overall database, and it has been suggested that further in vivo genotoxicity studies are required to complete their safety profiles. To address the utility of conducting additional in vivo genotoxicity studies, this review evaluates the specific genotoxicity studies that are the sources of concern, and evaluates the adequacy of the database including more recent genotoxicity data not mentioned in those publications. The current database of in vitro and in vivo studies for steviol glycosides is robust and does not indicate that either stevioside or rebaudioside A are genotoxic. This, combined with a lack of evidence for neoplasm development in rat bioassays, establish the safety of all steviol glycosides with respect to their genotoxic/carcinogenic potential.

Chronic consumption of rebaudioside A, a steviol glycoside, in men and women with type 2 diabetes mellitus.

This trial evaluated the effects of 16 weeks of consumption of 1000mg rebaudioside A (n=60) a steviol glycoside with potential use as a sweetener, compared to placebo (n=62) in men and women (33-75 years of age) with type 2 diabetes mellitus. Mean+/-standard error changes in glycosylated hemoglobin levels did not differ significantly between the rebaudioside A (0.11+/-0.06%) and placebo (0.09+/-0.05%; p=0.355) groups. Changes from baseline for rebaudioside A and placebo, respectively, in fasting glucose (7.5+/-3.7mg/dL and 11.2+/-4.5mg/dL), insulin (1.0+/-0.64microU/mL and 3.3+/-1.5microU/mL), and C-peptide (0.13+/-0.09ng/mL and 0.42+/-0.14ng/mL) did not differ significantly (p>0.05 for all). Assessments of changes in blood pressure, body weight, and fasting lipids indicated no differences by treatment. Rebaudioside A was well-tolerated, and records of hypoglycemic episodes showed no excess vs. placebo. These results suggest that chronic use of 1000mg rebaudioside A does not alter glucose homeostasis or blood pressure in individuals with type 2 diabetes mellitus.

Overview: the history, technical function and safety of rebaudioside A, a naturally occurring steviol glycoside, for use in food and beverages.

Rebaudioside A is a sweet tasting steviol glycoside extracted and purified from Stevia rebaudiana (Bertoni). Steviol glycosides can currently be used as a food ingredient in only a handful of countries. Questions on specifications, safety and special population effects have prevented steviol glycosides from obtaining a legal status permitting their use as a sweetener in most countries. A set of papers reporting results of research studies and reviews has been compiled in this Supplement to definitively answer unresolved questions. Specifically, recently completed studies on the general and reproductive toxicity of rebaudioside A corroborate studies carried out with purified steviol glycosides demonstrating safety at high dietary intake levels. Comparative metabolism studies provide further affirmation of the common metabolic pathway for all steviol glycosides and the common metabolism between rats and humans. Finally, clinical studies provide further evidence that purified rebaudioside A has no effect on either blood pressure or glucose homeostasis. This paper summarizes the information used to conclude that high purity rebaudioside A (rebiana) produced to food-grade specifications and according to Good Manufacturing Practices is safe for human consumption under its intended conditions of use as a general purpose sweetener.

The hemodynamic effects of rebaudioside A in healthy adults with normal and low-normal blood pressure.

Rebaudioside A and stevioside are steviol glycosides extracted from the plant Stevia rebaudiana Bertoni and are used in several countries as food and beverage sweeteners. This randomized, double-blind trial evaluated the hemodynamic effects of 4weeks consumption of 1000mg/day rebaudioside A vs. placebo in 100 individuals with normal and low-normal systolic blood pressure (SBP) and diastolic blood pressure (DBP). Subjects were predominantly female (76%, rebaudioside A and 82%, placebo) with a mean age of approximately 41 (range 18-73) years. At baseline, mean resting, seated SBP/DBP was 110.0/70.3mmHg and 110.7/71.2mmHg for the rebaudioside A and placebo groups, respectively. Compared with placebo, rebaudioside A did not significantly alter resting, seated SBP, DBP, mean arterial pressure (MAP), heart rate (HR) or 24-h ambulatory blood pressures responses. These results indicate that consumption of as much as 1000mg/day of rebaudioside A produced no clinically important changes in blood pressure in healthy adults with normal and low-normal blood pressure.

Subchronic toxicity and teratogenicity of 2-chloro-1,1,1,2-tetrafluoroethane (HCFC-124).

Inhalation studies were conducted to determine the potential toxicity of HCFC-124. Groups of rats and mice were exposed to HCFC-124 6 hr/day, 5 days/week for 13 weeks at 0, 5000, 15,000, and 50,000 ppm. Subgroups of rats and mice were held for a 1-month recovery period. A functional observational battery (FOB) was conducted on rats at 0, 4, 13, and 16 weeks. Clinical pathology evaluations were conducted at 7, 13, and 17 weeks. Thirteen or 17 weeks after study initiation, rats and mice underwent gross and microscopic evaluation, and livers were evaluated for hepatic beta-oxidation activity. In addition, groups of female rats and rabbits were exposed to HCFC-124 by inhalation during gestation to 0, 5000, 15,000, or 50,000 ppm. Exposure of rats and mice to HCFC-124 caused minimal compound-related effects. Compound-related changes occurred in several clinical pathology parameters in rats and mice. Hepatic beta-oxidation activity was significantly higher in 5000, 15,000, and 50,000 ppm male mice; however, there were no compound-related effects on beta-oxidation activity in rats. During the daily exposures, rats, mice, and rabbits exposed to 50,000 ppm were less responsive to auditory stimuli or less active compared to controls. At the 13-week FOB, male rats exposed to 15,000 or 50,000 ppm had decreased arousal. There were no compound-related effects on mortality, clinical signs, ocular tissues, hematology parameters, organ weights, and tissue morphology at any concentration in rats or mice. Maternal toxicity in rats was evident by a significant decrease in weight gain and food consumption at 50,000 ppm. Similarly, 50,000 ppm pregnant rabbits had lower food consumption. However, for both rats and rabbits, there was no evidence of fetal toxicity at any concentration.

A proposed design for conducting target animal safety studies for developing new veterinary pharmaceuticals.

Target animal safety studies are required for the regulatory approval of any new veterinary pharmaceutical registered almost anywhere in the world. Both the United States and Europe have extensive guidelines for conducting these studies. Due to new veterinary drug labeling requirements in the United States, changes in the overall design of target animal safety studies will be required for the United States. The guideline changes proposed in this paper are more evolutionary than revolutionary. While many new ideas are suggested, a number of existing practices that seem to work well were retained. Several new abbreviated drug development pathways are also proposed, an idea missing from current guidelines. This proposal, and others like it, should serve as a basis for representatives from the U.S. Food and Drug Administration, the European Commission on Veterinary Medical Products, the veterinary pharmaceutical industry, and practicing veterinarians to develop internationally harmonized target animal safety study guidelines that are clear in their intent, scientifically sound, and practical.

Two-year inhalation toxicity study in rats with hydrochlorofluorocarbon 123.

The potential chronic toxicity and oncogenicity of hydrochlorofluorocarbon 123 (HCFC-123) was evaluated by exposing male and female rats to 0, 300, 1000, or 5000 ppm HCFC-123 for 6 hr/day, 5 days/week, for 2 years. Clinical pathology was evaluated at 6, 12, 18, and 24 months. An interim termination and measurements of hepatic cell proliferation and beta-oxidation activity were conducted at 12 months. The terminal euthanization occurred at 24 months. Males and females exposed to 5000 ppm and females exposed to 300 or 1000 ppm had lower body weights and body weight gains. Serum triglyceride and glucose concentrations were significantly decreased at all exposure concentrations in both sexes. Serum cholesterol was also lower in 300, 1000, and 5000 ppm females and in 5000 ppm males. Alterations in serum protein concentrations occurred at 300, 1000, and 5000 ppm. Survival was higher in 1000 and 5000 ppm males and females. At 24 months, increased relative liver weight occurred in 5000 ppm males, and decreased absolute kidney weight occurred in 5000 ppm males and in 1000 and 5000 ppm females. Benign hepatocellular adenomas were increased in 5000 ppm males and in all test groups of females. Hepatic cholangiofibromas were also increased in 5000 ppm females. Pancreatic acinar cell adenomas were increased in all test groups of males, and acinar cell hyperplasia was increased in the 1000 and 5000 ppm males and females. Benign testicular interstitial adenomas and focal interstitial cell hyperplasia were also increased in all male test groups compared to controls. Diffuse retinal atrophy was increased in all male and female test groups, but it was considered to be an indirect compound-related effect. Hepatic beta-oxidation activity (peroxisome proliferation) was higher in 300, 1000 and 5000 ppm males and 1000 and 5000 ppm females. Compound-related differences in the rate of hepatic cell proliferation were not observed at any exposure concentration. Decreased incidences of a variety of age-related lesions occurred at 1000 and 5000 ppm.

Safety of albendazole in developing bovine fetuses.

Albendazole, administered orally at a dose rate of 25 mg/kg of body weight to presumed pregnant cows or heifers on days 21, 31, 41, 51, and 61 of gestation, did not induce toxicosis in embryos or fetuses, and all calves born were structurally normal. Albendazole administration at a rate of 25 mg/kg to cows at 7 and/or 14 days of gestation decreased the apparent conception rate (ie, embryolethality), but did not have a teratogenic effect. Apparent embryolethality was greater in cows administered 25 mg/kg only on day 14, compared with those administered the drug only on day 7. Single dosage of 25 mg/kg given in the final 3 months of gestation did not induce abortion. There were no adverse effects of albendazole at a dosage of 10 or 15 mg/kg on developing embryos or fetuses when administered to presumed pregnant cows at various times in early gestation.

Inhalation toxicity of an isomeric mixture of hydrochlorofluorocarbon (HCFC) 225 in male rats.

A blend of the hydrochlorofluorocarbon isomers HCFC-225ca and HCFC-225cb has been proposed as a potential substitute for CFC-113, an important solvent and cleaning agent. The toxicity following repeated inhalation of an HCFC-225 isomer mixture was assessed in male Crl:CDBR rats. Three groups of 10 male rats were exposed to the test compound in air at design concentrations of 500, 5000, and 13,000 ppm. Rats were exposed 6 hr/day, 5 days/week for 2 weeks. A control group of 10 male rats was exposed to air only. Decreased serum cholesterol, triglycerides, and glucose; dose-related increased mean absolute and relative liver weights; and microscopic hepatocellular hypertrophy were present at all exposure concentrations. Hepatocellular hypertrophy correlated ultrastructurally to proliferation of peroxisomes. Clinical chemical parameters and organ weight and morphologic changes in the liver were reversible following 14 days of recovery.

Effect of circulating neutrophil depletion on lung injury induced by inhaled silica particles.

Polymorphonuclear leukocytes (PMNs) recruited into the alveolar region during inflammation may injure the lung parenchyma by releasing cytotoxic oxygen radicals and proteases. Because brief exposures to crystalline silica elicit recruitment of PMNs into the alveolar region, which is strongly correlated with parameters of cytotoxicity, increased alveolar epithelial permeability, and lysosomal enzyme release, we sought to evaluate the potential role of PMNs in silica-induced lung injury. Rats were depleted of PMNs by administration of an anti-rat PMN antiserum prior to exposure to silica. Pulmonary inflammatory responses to silica in this group were compared to responses in normal silica-exposed rats as well as sham-exposed normal or PMN-depleted rats. Bronchoalveolar lavage fluids from normal, silica-exposed rats contained 9.7 x 10(6) PMNs immediately after exposure for 3 days, compared to 0.01 x 10(6) PMNs for both normal or PMN-depleted, sham-exposed rats. Bronchoalveolar lavage fluids from successfully PMN-depleted, exposed rats contained significantly fewer (0.7 x 10(6)) PMNs compared to normal silica-exposed rats. In both groups of silica-exposed rats, a variety of biochemical indicators of lung injury were increased significantly compared to measurements from both sham-exposed groups, but there were no differences between PMN-depleted and normal silica-exposed groups. The results suggest that recruitment of PMNs into the alveolar region is not a necessary prerequisite for the observed increases in biochemical indicators of silica-induced acute lung injury.

Interpreting clinical laboratory data in safety assessment studies.

Traditionally, clinical laboratory data in safety assessment studies have been analyzed statistically to determine potential treatment-related effects. This has often led to inappropriate and/or controversial conclusions about clinical laboratory findings that are statistically significant but considered biologically unimportant. To overcome these problems, a decision-making process that integrates statistical analyses and sound medical judgement is required. Understanding the reasons why false-positive and false-negative results occur is the first step toward both better test selection and more accurate decisions on test material effects.

Clinical pathology testing recommendations for nonclinical toxicity and safety studies. AACC-DACC/ASVCP Joint Task Force.

Clinical pathology testing in nonclinical toxicity and safety studies is an important part of safety assessment. In recent years, clinical laboratory testing has rapidly expanded and improved. Some government regulatory agencies provide guidelines for clinical pathology testing in nonclinical toxicity and safety studies. To improve these testing guidelines and the resultant safety assessments, the American Association for Clinical Chemistry's Division of Animal Clinical Chemistry and the American Society for Veterinary Clinical Pathology formed a joint committee to provide expert recommendations for clinical pathology testing of laboratory species involved in subchronic and chronic nonclinical toxicity and safety studies. These recommendations include technical recommendations on blood collection techniques and hematology, serum chemistry, and urinalysis tests.

Complement facilitates macrophage phagocytosis of inhaled iron particles but has little effect in mediating silica-induced lung inflammatory and clearance responses.

Complement-mediated mechanisms are known to play a role in pulmonary inflammation and clearance responses to some types of inhaled particles. The present studies were undertaken to investigate the role of complement in mediating pulmonary inflammation and/or phagocytosis as a function of particle clearance in rats exposed to silica or carbonyl iron (CI) particles. Both particle types were shown to be weak activators of serum complement in vitro. In these studies, normal and complement-depressed (CVF-treated) rats were exposed to aerosols of CI or silica particles for 6 hr at 100 mg/m3. Following exposure, alveolar fluids and cells from sham and dust-exposed animals were recovered by bronchoalveolar lavage (BAL) at several time periods postexposure and measured for a variety of biochemical and cellular indices. In addition, pulmonary macrophages were cultured and studied for morphology and phagocytosis. Our results showed that CI exposure did not produce cellular or biochemical indices of pulmonary inflammation, either in normal or complement-depleted rats. However, fewer phagocytic macrophages were recovered from the lungs of CVF-treated, CI-exposed rats than from normal exposed animals. In contrast, silica inhalation produced a sustained PMN inflammatory response in the lungs of exposed rats, measured up through 1 month postexposure, along with significant increases in BAL fluid levels of LDH, protein, and alkaline phosphatase (P less than 0.05) and deficits in pulmonary macrophage phagocytic functions. Cobra venom factor (CVF) treatment prior to exposure in rats had no significant effect upon the silica-induced parameters, suggesting that complement may not play an important role in the acute pulmonary response to silica. The results indicate that complement may play a role in mediating CI-related macrophage clearance responses but has little effect upon sustained silica-induced pulmonary inflammatory parameters.

Modification of an automated method for determining plasma and erythrocyte cholinesterase activity in laboratory animals.

By altering the analytical parameters on an automated analyzer, analytical precision for measuring cholinesterase (ChE) activity in hemolysates was markedly improved in samples from several species. Manual and automated spectrophotometric analyses of plasma and erythrocyte ChE activity were optimized for use in rats, mice and dogs. Replicate ChE analyses were performed on plasma samples and on hemolysates made from whole blood or packed erythrocytes to determine the precision of the manual ChE method and 4 modifications of the automated method. Large method-related differences in precision were observed for the erythrocyte assay, but not the plasma assay. The addition of a nonionic detergent to make hemolysates was beneficial in determining erythrocyte ChE activity in the rat, but not in the mouse or dog. Species specific temperature conversion factors were necessary for comparing results from methods using different analytical temperatures. Analysis of whole blood hemolysates provided similar or better precision for determining erythrocyte ChE activity compared to using hemolysates made from packed erythrocytes. Comparisons of erythrocyte ChE results obtained from assays with even minor methodological differences should be approached with caution because of the many analytical factors which can affect results.

Four-week inhalation toxicity study with Ludox colloidal silica in rats: pulmonary cellular responses.

This study was designed to complement a traditional subchronic inhalation toxicity study with Ludox colloidal silica. CD rats were exposed nose-only for 2 or 4 weeks at concentrations of 0, 10, 50, and 150 mg/m3 Ludox (dried SiO2). Additional groups of rats exposed for 4 weeks were given a 3-month recovery period. Following exposure and/or recovery, fluids and cells were recovered from the lungs by bronchoalveolar lavage (BAL) and measured for cellular and biochemical parameters. Additional groups of animals were processed for cell labeling studies or lung deposition studies. Inhaled doses of Ludox colloidal silica were measured after 4-week exposures and were found to be 489 micrograms/lung (10 mg/m3 group), 2418 micrograms/lung (50 mg/m3), and 7378 micrograms/lung (150 mg/m3), respectively. Results showed that exposures to 150 mg/m3 Ludox for 2 or 4 weeks produced pulmonary inflammation along with increases in BAL protein, LDH, and alkaline phosphatase values (p less than 0.05) and reduced macrophage phagocytosis. Inflammatory responses, evidenced by increased numbers of neutrophils, were also measured in the lungs of the 50 mg/m3 group following 2 and/or 4 weeks of exposure. Most biochemical parameters for all groups returned to control values following a 3-month recovery period. Autoradiographic studies demonstrated that the labeling indices of terminal bronchiolar and lung parenchymal cells were generally increased in the 50 and 150 mg/m3 groups after 2 and 4 weeks of exposure but, with one exception, returned to normal levels following a 3-month postexposure period. No significant alterations in any measured parameters were detected in rats exposed to 10 mg/m3 Ludox at any time postexposure. The determination of a no-observable-effect level (NOEL) of 10 mg/m3 was consistent with results obtained by conventional toxicology methods and affirms the utility of these biochemical, cellular, and autoradiographic techniques for providing a predictive screen to assess the toxicity of inhaled particles.

Development of a short-term inhalation bioassay to assess pulmonary toxicity of inhaled particles: comparisons of pulmonary responses to carbonyl iron and silica.

This paper describes a short-term inhalation bioassay for evaluating the lung toxicity of inhaled particulate materials. To validate the method, rats were exposed for 6 hr or 3 days to various concentrations of either aerosolized alpha-quartz silica or carbonyl iron particles. Cells and fluids from groups of sham- and dust-exposed animals were recovered by bronchoalveolar lavage (BAL). Alkaline phosphatase, lactate dehydrogenase (LDH), and protein values were measured in BAL fluids at several time points postexposure. Cells were identified, counted, and evaluated for viability. Pulmonary macrophages (PM) were cultured and studied for morphology, chemotaxis, and phagocytosis by scanning electron microscopy. The lungs of additional exposed animals were processed for histopathology and transmission electron microscopy. Brief exposures to silica elicited a sustained granulocytic inflammatory response (primarily neutrophils) with concomitant increases in alkaline phosphatase, LDH, and protein in the lavage fluids (p less than 0.05). In addition, PM functional capacity was depressed (p less than 0.05) and histopathologic lesions were observed within 1 month after exposure. In contrast, 6-hr or 3-day exposures to CI produced no cellular, cytotoxic, or alveolar/capillary membrane permeability changes at any time postexposure. PM function was either enhanced or unchanged from controls. These data demonstrate that short-term, high-dose inhalation exposures of silica produce effects similar to those previously observed using intratracheal instillation or chronic inhalation models, and lend support to this method as a reliable short-term bioassay for evaluating the pulmonary toxicity and mechanisms associated with exposures to new and untested materials.

Interpreting rodent clinical laboratory data in safety assessment studies: biological and analytical components of variation.

Evaluation of the biological and toxicological significance of clinical laboratory results obtained in safety assessment studies requires an understanding of factors unrelated to the treatment that may affect test results. Since the magnitude of the components of variation is usually unknown, the toxicologic significance of small, but statistically significant, differences between control and treatment group results can be difficult to assess. Over a 12-week period, components of variation were determined for a wide range of hematologic and clinical biochemical assays in clinically normal Sprague-Dawley-derived rats. Estimates of variance components and ratios were obtained for each test. While the intra-animal/inter-animal ratios (r ratios) revealed some important tests with a high or low degree of individuality, many hematologic and clinical biochemical tests had r ratios within the 95% confidence interval for equal variances. Analytical variance ratios revealed tests that are sensitive to the effects of experimental error and experimental design. Due to the diversity of the variance component patterns among clinical laboratory tests, complex experimental designs may be required to reduce the effects of analytical and biological variation on the statistical analysis of clinical laboratory data. The results of this study suggest that statistically significant clinical laboratory findings that are not biologically or toxicologically important will be present in many rodent safety assessment studies with a standard design. Overreliance on the result of standard prepackaged statistical analyses for determining the presence of toxicologically significant findings can lead to misinterpretation of clinical laboratory data. Sound medical judgment must be applied to clinical laboratory findings using appropriate statistical analyses as a tool for pattern recognition.

Subchronic inhalation study with vinyl fluoride: effects on hepatic cell proliferation and urinary fluoride excretion.

Vinyl fluoride is used widely in the manufacture of fluoropolymers. Based in part on the structural similarity of vinyl fluoride to the hepatocarcinogens vinyl chloride and vinyl bromide, a TSCA Section 4 test rule mandated the testing of vinyl fluoride for oncogenicity. This report presents the results of a 90-day inhalation study in rats and mice with vinyl fluoride designed to set test concentrations for a subsequent oncogenicity study. Groups of 15 male and female rats and mice were exposed 6 hr per day, 5 days per week for approximately 90 days to target concentrations of 0, 200, 2000, or 20,000 ppm vinyl fluoride. Clinical chemical, hematological, and urine analyses were performed on rats after 45 and 90 days of exposure. A hematological evaluation was performed on mice following 45 and 90 days of exposure. A complete gross and microscopic evaluation was conducted at the end of the study. After 93 days on test, groups of five rats and five mice per sex were implanted with osmotic minipumps containing [3H]thymidine and were exposed for an additional 5 days to measure cell proliferation in liver, kidney, lung, and nasal cavity tissues. Results of the histopathological, clinical chemical, and hematological evaluations showed no significant effects of vinyl fluoride exposure at any concentration following either 45 or 90 days of exposure. A concentration-related increase in fluoride ion in urine was observed in rats at 45 and 90 days of exposure. A plateau in urinary fluoride excretion was observed at approximately 2000 ppm, suggesting saturation of vinyl fluoride metabolism. Vinyl fluoride-related cell proliferation effects were largely restricted to liver. Hepatic cell proliferation in male and female rats and mice was elevated at all concentrations. The response was similar at concentrations of either 2000 or 20,000 ppm and was consistent with concentration-response relationships for other haloethylenes. Taken together, the urinary fluoride excretion and hepatic cell proliferation data suggest a mechanistic link between the two effects. On the basis of these findings and experience with other haloethylenes, concentrations of vinyl fluoride to be tested for oncogenicity should be chosen such that the full linear range of the concentration-response curve is evaluated. The present study demonstrates through example the value of incorporating cell proliferation studies in standard testing protocols.

Attenuation of perfluoropolymer fume pulmonary toxicity: effect of filters, combustion method, and aerosol age.

Thermal decomposition products of some perfluorinated polymers are toxic to experimental animals in small-scale combustion toxicity tests; the toxicity is dependent upon the heating procedure, combustion temperature, and other experimental conditions. In the current studies we investigated the time course of fume generation and exposure on pulmonary effects in rats following a 30-min exposure to perfluoropolymer decomposition products (i.e., fume concentration = 0.2 mg/m3 of tetrafluoroethylene/hexafluoropropylene copolymer (FEP)) pyrolyzed with either static or dynamic airflows. In the first set of experiments, five different groups of rats were exposed to FEP fumes in a static combustion toxicity test system. Three groups were exposed to unfiltered FEP fumes during 0- to 15-, 15- to 30-, and 0- to 30-min intervals, respectively, and one to a filtered (particle-free) atmosphere of combusted FEP for 30 min. Sham-exposed rats constituted the control group. Immediately after exposure, the rats were sacrificed and their lungs weighed and lavaged or perfused to assess indices of cytotoxicity. Our results showed that lung weights, markers of inflammation, and pulmonary hemorrhage and alkaline phosphatase, beta-glucuronidase, lactate dehydrogenase, and protein levels in bronchoalveolar lavage fluids were significantly elevated in all unfiltered FEP-exposed groups compared to those in either the rates exposed through filters or controls (P less than 0.01). In a second set of experiments using a dynamic pyrolysis toxicity test system, rats were exposed for 30 min to FEP-pyrolyzed fumes which were either freshly generated or aged for 1 or 5 min prior to delivery to the animal's breathing zone. Subsequently, lung cytotoxicity parameters were measured. Rats exposed directly to the fresh fumes demonstrated toxic effects consistent with those described above (P less than 0.01), but the pulmonary toxicity of aged (i.e., 1 or 5 min delay) FEP fumes was diminished in a time-dependent manner, suggesting that the toxicant was unstable. Histopathological studies correlated with biochemical results and revealed that inhalation of unfiltered or freshly generated FEP fumes produced a severe lung injury characterized by the development of alveolar and interstitial edema, intraalveolar hemorrhage, congestion, and fibrin deposition. Electron microscopy studies demonstrated severe damage to terminal bronchiolar cells and detachment of Type I epithelial and endothelial cells in pulmonary regions. The severity of pathology observed in lungs of rats exposed to 1-min aged fumes was intermediate between unfiltered/unaltered fume-exposed animals and sham controls. The results of these studies demonstrate that the lung toxicity of perfluoropolymer fumes is associated with the aerosol phase generated in perfluoropolymer pyrolysis.