Page 9: Guidelines for Canadian Drinking Water Quality: Guideline Technical Document – Tetrachloroethylene
8.0 Kinetics and metabolism
8.1 Absorption
Experimental absorption of tetrachloroethylene via the ingestion route has not been measured in humans; however, a poisoning case study in which tetrachloroethylene was measured in blood after being ingested by a 6-year-old boy (Koppel et al., 1985) indicated that tetrachloroethylene was absorbed following oral exposure.
Initial uptake of tetrachloroethylene via inhalation is rapid in humans, but declines as blood and body tissues become saturated (Fernandez et al., 1976; Monster, 1979). In six male volunteers exposed by inhalation to tetrachloroethylene at concentrations of 72 and 144 ppm, absorption was > 90% at the beginning of exposure and fell to approximately 50% after 8 hours (Monster, 1979). Steady-state levels of tetrachloroethylene in blood were reached after 50-100 minutes in five subjects exposed by inhalation to 100 ppm, and retention was 78-89% (Benoit et al., 1985). Another study found that peak tetrachloroethylene levels in venous blood were measured at the end of a 6-hour inhalation exposure to 1 ppm and declined thereafter (Chiu et al., 2007). Absorption of tetrachloroethylene by inhalation is also substantial in experimental animals (Schumann et al., 1980; Dallas et al., 1994).
Dermal absorption of tetrachloroethylene is less extensive than that via the ingestion and inhalation routes. Percutaneous absorption of tetrachloroethylene can occur in humans when skin is exposed to the compound in liquid form (Stewart and Dodd, 1964; Aitio et al., 1984; Kezic et al., 2001), but dermal absorption of tetrachloroethylene in vapour form is negligible compared with absorption following inhalation exposure (Riihimaki and Pfaffli, 1978). Stewart and Dodd (1964) suggested that dermal absorption of tetrachloroethylene is unlikely to be hazardous under normal working conditions. Dermal absorption of liquid tetrachloroethylene has been measured in mice (Tsuruta, 1975), rats (Tsuruta, 1977) and guinea pigs (Bogen et al., 1992), and absorption of tetrachloroethylene vapour was explored in mice (Tsuruta, 1989). Rats have shown a skin uptake of 3.5% following dermal exposure to tetrachloroethylene vapour at a concentration of 12 500 ppm (McDougal et al., 1990).
8.2 Distribution
Although no studies have measured the distribution of tetrachloroethylene in humans following oral exposure, distribution of tetrachloroethylene is not expected to be dependent on the route of exposure, as demonstrated in rats (Pegg et al., 1979).
Only a small fraction (~1-4%) of inhaled and ingested tetrachloroethylene has been observed to remain in the carcass of rats (Pegg et al., 1979; Frantz and Watanabe, 1983). The majority of the tetrachloroethylene that is stored tends to be distributed to fatty tissues, which reflects tetrachloroethylene's lipophilicity. Tissues with the highest concentrations of tetrachloroethylene are those with higher lipid content in both humans (Lukaszewski, 1979; Levine et al., 1981; Garnier et al., 1996) and experimental animals (Savolainen et al., 1977; Pegg et al., 1979). For example, tissue analyses of humans following fatal inhalation exposures showed the highest tetrachloroethylene concentrations in the liver, brain and kidney and the lowest concentrations in the lung tissue (Lukaszewski, 1979; Levine et al., 1981; Garnier et al., 1996), consistent with the lipophilic properties of tetrachloroethylene. Similarly, the highest concentrations of tetrachloroethylene in exposed rats were found in fat, kidneys and liver, and the lowest concentrations were in lung, heart and adrenals (Savolainen et al., 1977; Pegg et al., 1979). Some discrepancies were found related to storage in the brain, as tetrachloroethylene was not detected in the brain in one study (Pegg et al., 1979), but was detected in the brain in another study (Savolainen et al., 1977), with higher concentrations in the cerebrum than in the cerebellum.
Tetrachloroethylene is also distributed to tissues of relevance for developing humans and animals. After exposure of pregnant rats to tetrachloroethylene, the compound was measured in fetal blood and amniotic fluid (Ghantous et al., 1986; Szakmáry et al., 1997). Tetrachloroethylene has also been found in breast milk in humans (Bagnell and Ellenberger, 1977; Schreiber et al., 2002) and experimental animals (Byczkowski and Fisher, 1994), with partitioning to breast milk likely higher in animals than in humans due to differences in milk fat content (Byczkowski and Fisher, 1994).
8.3 Metabolism
The metabolism of tetrachloroethylene in experimental animals and humans has been thoroughly reviewed by Anders et al. (1988), Lash and Parker (2001) and U.S. EPA (2012c).
The majority of inhaled tetrachloroethylene does not undergo metabolism in humans and is excreted unchanged (Monster, 1979; Benoit et al., 1985; Chiu et al., 2007). For the minor amount of the compound that is metabolized, there are two major pathways for transformation. In both humans and experimental animals, tetrachloroethylene undergoes mainly oxidative metabolism and will proceed to reductive metabolism once the enzymes associated with oxidative metabolism are saturated (Lash and Parker, 2001).
Oxidative metabolism is driven by cytochrome P450 (CYP) enzymes, with CYP2E1 as the major relevant isoform for tetrachloroethylene metabolism (Lash and Parker, 2001). The primary site for oxidative metabolism is the liver, with smaller amounts of metabolism occurring in other organs. Although renal metabolism is relevant to rats (Cummings et al., 1999), the role of renal metabolism is less clear in humans, as the human kidney expresses some CYP isoforms, but not CYP2E1 (Cummings et al., 2000). The major end products of oxidative metabolism are TCA, carbon monoxide, carbon dioxide and—primarily in rats—oxalic acid. Reaction of tetrachloroethylene with CYP2E1 first results in a tetrachloroethylene-iron oxide intermediate, which can then generate 1,1,2,2-tetrachloroethylene oxide (a hypothesized epoxide), oxalic acid and trichloroacetyl chloride. The epoxide is thought to generate trichloroacetyl chloride, ethanedioyl chloride (which eventually can be metabolized to carbon monoxide and carbon dioxide) and chloral hydrate, and reaction of the epoxide with microsomal epoxide hydrase forms oxalic acid. The major metabolite, TCA, is formed from trichloroacetyl chloride. There is conflicting evidence stating that DCA, the major metabolite from the glutathione transferase (GST) pathway, might be generated from TCA by conversion in gut microflora; however, the primary source of DCA is thought to be from the GST pathway (U.S. EPA, 2012c).
The second pathway for metabolism, the GST pathway, becomes predominant when tetrachloroethylene levels are high enough for saturation of CYP2E1; however, it can still operate prior to the saturation of the oxidative pathway (Chiu and Ginsberg, 2011). Saturation of oxidative metabolism occurs in humans at inhalation exposures of approximately ≥ 100 ppm, as metabolite excretion measured in workers in industrial and dry cleaning settings was shown to plateau at higher concentrations (Ikeda, 1977; Ohtsuki et al., 1983; Seiji et al., 1989); however, there are still uncertainties regarding the extent of GST metabolism in humans (Chiu and Ginsberg, 2011). Although the GST pathway is responsible for less tetrachloroethylene metabolism at low exposure levels, it is still considered of importance, because it produces several reactive metabolites.
Through glutathione conjugation, further biotransformation and processing occur; GST mediates this production of water-soluble compounds for excretion (U.S. EPA, 2012c). Conjugation begins primarily in the liver, where GST converts tetrachloroethylene to S-(1,2,2-trichlorovinyl) glutathione (TCVG), and early products of the metabolism—including TCVG and the product it generates with the catalysts gamma-glutamyltransferase and dipeptidases, S-(1,2,2-trichlorovinyl) cysteine (TCVC)—travel to the kidney; however, a smaller amount of metabolism also occurs in the kidney (Lash et al., 1998; U.S. EPA, 2012c). N-Acetyl transferase can catalyze the conversion of TCVC to the excretory product N-acetyl-S-(1,2,2-trichlorovinyl)-L-cysteine (NAcTCVC), but other enzymes can lead to the production of the reactive metabolites TCVC sulphoxide (via flavin monooxygenase-3 and CYP) or the unstable 1,2,2-trichlorovinylthiol (via β-lyases). DCA, the major end-product of the GST pathway, can be formed from the latter reactive metabolite (U.S. EPA, 2012c).
Variability in tetrachloroethylene metabolism has been studied in human populations. Some variability in metabolism may be related to ethnicity, as demonstrated by studies showing lower levels of urinary TCA and higher exhalation of tetrachloroethylene in Chinese workers compared with Japanese workers (Seiji et al., 1989) and in Asians compared with Caucasians (Jang and Droz, 1997). It has been suggested that variations in the oxidative and reductive metabolic pathways due to genetic polymorphisms in populations may be responsible for the differences in metabolism of tetrachloroethylene among individuals and ethnic groups (Lash et al., 2007; U.S. EPA, 2012c).
Although the conversion of tetrachloroethylene to TCA is the main route of metabolism in humans, rats and mice, species differences in the metabolism of tetrachloroethylene have been identified. Mice metabolize tetrachloroethylene more extensively than rats, and rats metabolize it more extensively than humans (Schumann et al., 1980; Völkel et al., 1998; Lash and Parker, 2001). Moreover, although saturation of oxidative metabolism occurs at similar levels in rats and humans, oxidative metabolism does not become saturated in mice, as demonstrated by the continued increase in blood TCA concentrations at inhalation exposure levels of up to 400 ppm (Odum et al., 1988).
8.4 Excretion
For humans and experimental animals, many quantitative studies have shown that the majority of tetrachloroethylene is eliminated unmetabolized in exhaled air, regardless of the route of exposure (Yllner, 1961; Daniel, 1963; Fernandez et al., 1976; Monster, 1979; Pegg et al., 1979).
Tetrachloroethylene exhibits a multiphasic elimination curve in humans, with a rapid phase followed by slower phases (Fernandez et al., 1976; Monster, 1979). The rapid phase likely represents the elimination of tetrachloroethylene from the blood immediately after exposure, and the slow phase is indicative of elimination from adipose tissue (Monster, 1979). Pulmonary excretion was estimated from modelling to occur in three different phases: a rapid phase of elimination from blood vessel-rich tissues (brain, heart, hepatoportal system, kidneys, endocrine glands); a slower elimination from muscle, skin and blood vessel-poor tissues (connective tissue, lung tissue); and a very slow elimination from adipose tissues (Guberan and Fernandez, 1974). Quantitative estimates of the rapidity of elimination in humans vary, with empirical studies demonstrating half-lives of 12-16, 30-40 and 55 hours, depending on where along the concentration curve they were calculated (20, 50 and 100 hours, respectively) (Monster, 1979), and 79 minutes (Benoit et al., 1985) and a model providing an estimate of 71.5 hours (Guberan and Fernandez, 1974); this phenomenon is likely representative of the different phases of elimination from different tissues. Calculated half-lives in rats ranged from 6.94 to 7.43 hours (Pegg et al., 1979). Elimination occurs more rapidly via exhalation than via urinary excretion, as demonstrated by elimination half-lives of 65 hours and 144 hours, respectively (Ikeda, 1977, as calculated from data from Stewart et al., 1970). Men appear to eliminate tetrachloroethylene more rapidly than women, with half-lives of urinary metabolites of 123 hours in males and 190 hours in females, estimated from occupational exposures (Ikeda and Imanura, 1973); however, this differed from rats, where there were no differences in elimination half-lives between males and females (Ikeda and Imanura, 1973). Elimination over a short time period is also more pronounced in individuals with lower body weights than in those with higher body weights (Monster et al., 1983).
Tetrachloroethylene can also be excreted in breast milk, as demonstrated in a case study. Breast milk concentrations in a woman exposed to tetrachloroethylene while visiting her husband at a dry cleaning establishment were 1 mg/dL 1 hour after her visit and 0.3 mg/dL 24 hours afterwards; the blood concentration of tetrachloroethylene measured 2 hours after her exposure was 0.3 mg/dL (Bagnell and Ellenberger, 1977).
The majority of tetrachloroethylene (80-100%) is excreted unchanged via exhalation in humans (Monster, 1979; Benoit et al., 1985; Chiu et al., 2007); however, in animals—particularly mice—metabolism is greater (Yllner, 1961; Daniel, 1963; Pegg et al., 1979; Schumann et al., 1980), leading to levels of unchanged tetrachloroethylene in breath as low as 12% in mice (Schumann et al., 1980). Only minor amounts of metabolites are excreted in humans exposed by inhalation to 72-200 ppm (Fernandez et al., 1976; Monster, 1979), with slightly higher levels in rats exposed by gavage to 1-9000 mg/kg bw per day for up to 12 administrations or by inhalation to 10-600 ppm for 6 hours (Daniel, 1963; Pegg et al., 1979; Schumann et al., 1980) and much higher levels in mice exposed by gavage to 100-1000 mg/kg bw per day for up to 12 administrations or by inhalation to 10-600 ppm for 6 hours (Schumann et al., 1980).
The major metabolite in humans is TCA (Völkel et al., 1998), which is excreted in urine (Monster, 1979; Birner et al., 1996; Völkel et al., 1998; Schreiber et al., 2002; Chiu et al., 2007); however, studies demonstrate that only ≤ 1-3% of tetrachloroethylene intake is excreted as TCA (Monster, 1979; Chiu et al., 2007). The half-life of TCA has been measured at 45.6-65 hours in urine (Monster et al., 1983; Völkel et al., 1998) and 75-90 hours in blood (Monster, 1979; Monster et al., 1983), and the highest excretion has been measured to occur 24-48 hours after exposure (Fernandez et al., 1976). TCA excretion is more rapid in experimental animals than in humans (Völkel et al., 1998). Other metabolites that have been measured in human urine after tetrachloroethylene exposure include NAcTCVC (Birner et al., 1996; Völkel et al., 1998; Schreiber et al., 2002) and trichloroethanol (Monster, 1979; Birner et al., 1996; Schreiber et al., 2002); however, the latter compound is not expected to be a metabolite of tetrachloroethylene and might therefore result from simultaneous trichloroethylene exposures or be an analytical artifact (U.S. EPA, 2012c). DCA has been detected in the urine of rats, but not humans (Völkel et al., 1998).
8.5 Physiologically based pharmacokinetic models
In human health risk assessment for tetrachloroethylene, physiologically based pharmacokinetic (PBPK) modelling is useful, because no appropriate toxicity data are available for humans ingesting tetrachloroethylene in drinking water, ingestion studies in experimental animals are limited and the metabolite generation over low to high parent compound exposures is non-linear. Several PBPK models have been developed for tetrachloroethylene; a detailed critical review of available models was performed by Clewell et al. (2005).
The basic structure of these models is based primarily on the initial Ramsey and Andersen (1984) styrene model, which compartmentalizes organs into liver, fat, rapidly perfused tissues and slowly perfused tissues. A separate kidney compartment was also included in many of the models (Gearhart et al., 1993; Covington et al., 2007; Qiu et al., 2010; Chiu and Ginsberg, 2011), and some models (Rao and Brown, 1993; Dallas et al., 1994; Qiu et al., 2010) included a brain compartment to enable prediction of tetrachloroethylene concentrations in target tissues in neurological studies. Although several earlier models either did not simulate metabolism (Rao and Brown, 1993) or predicted only overall tetrachloroethylene metabolism (Bois et al., 1996; Reitz et al., 1996), generation of TCA via the oxidative pathway has been included in models for rats (Chiu and Ginsberg, 2011), mice (Gearhart et al., 1993; Fisher et al., 2004; Sweeney et al., 2009; Chiu and Ginsberg, 2011) and humans (Gearhart et al., 1993; Covington et al., 2007; Chiu and Ginsberg, 2011). GST metabolism has been considered in only two models (Sweeney et al., 2009; Chiu and Ginsberg, 2011), wherein urinary concentrations of NAcTCVC and DCA were predicted in rats, mice and humans. The models were designed for exposures from the injection, inhalation and/or ingestion routes, and one model included exposure to tetrachloroethylene via both dermal and inhalation routes from bathing and showering (Rao and Brown, 1993). Monte Carlo simulation components have also been added to some models (Gearhart et al., 1993; Bois et al., 1996; Covington et al., 2007) to consider distributions of values for some of the model parameters (e.g., the ventilation over perfusion ratio, blood flows, volumes, partition coefficients and metabolism values).
The model by Gearhart and colleagues (1993) was used as the basis for the PBPK model developed to facilitate rodent to human and inhalation to ingestion extrapolation estimates for this assessment (Nong, 2013). The model quantitatively accounts for differences in metabolism between animals and humans. Minor adjustments to the model were applied using physiological and metabolic parameters from Reitz et al. (1996) and Clewell et al. (2005). Health Canada's model (Nong, 2013) allows for the estimation of concentration and area under the concentration-time curve (AUC) for tetrachloroethylene in the blood, liver and kidney; rates of metabolism in liver and kidney; and concentration and AUC of TCA in blood. The model, however, does not allow for estimates of TCA levels in tissues. Moreover, a brain compartment was not added to the model, as the blood-brain partition coefficient is similar to that for blood-liver and blood-kidney, and liver and kidney tetrachloroethylene concentrations and AUCs have been demonstrated to be appropriate proxies for brain levels (Dallas et al., 1994). The model was also extrapolated to describe pregnancy kinetics using similar assumptions as for the Fisher et al. (1989) model for trichloroethylene. The model also incorporated elements from Rao and Brown (1993) to allow for estimates of tetrachloroethylene exposure via the dermal (in liquid form—a conservative estimate because it is more highly absorbed than tetrachloroethylene in vapour form, which will form some of the exposure in bathing) and inhalation (in vapour form) routes from bathing and showering, which were used to calculate the litre-equivalent values (see Section 5.6). The Health Canada model (Nong, 2013) was validated using data for mice (Odum et al., 1988), rats (Dallas et al., 1994; Reitz et al., 1996) and humans (Fernandez et al., 1976; Monster, 1979; Rao and Brown, 1993; Völkel et al., 1998), with exposures from the ingestion (Gearhart et al., 1993; Dallas et al., 1994), inhalation (Fernandez et al., 1976; Monster, 1979; Odum et al., 1988; Dallas et al., 1994; Reitz et al., 1996; Völkel et al., 1998) and dermal (Rao and Brown, 1993) routes.
Page details
- Date modified: