Commonwealth of Australia 2010
Toxicity reference values (TRVs)
Download 2.29 Mb.
|
- Navigate this page:
- 55.1.2Selection of Toxicity Reference Values (TRVs)
- 57.Avian toxicity data
- 58.Mammalian oral toxicity data
- 59.Aquatic toxicity data
Toxicity reference values (TRVs)55.1.1The TRV approach and summary of the TRVs determinedOral Toxicity Reference Values (TRVs) derived for the assessment of risks from exposure to cyanide by mammals and birds and the aquatic TRV have been presented in Table 9.. TRV is equivalent in general terminology to the term Predicted No Effect Concentration (PNEC), but the United States Army Center for Health Promotion and Preventative Medicine (USACHPPM, 2000) reference discussed below uses the term TRV. A high level of confidence of wildlife health protection is afforded by the derived TRVs for cyanide. Rationales for these values are presented in the following section. No published wildlife TRV for cyanide based on recent research was available. In an ecological risk assessment of a site in the United States, Tetra Tech EM Inc (2002) adopted an avian TRV of 0.04 mg/kg bw based on the acute oral toxicity data for American kestrel of Wiemeyer et al. (1986). In that instance, the acute LD50 of 4 mg/kg bw was divided by an uncertainty factor of 100 (as in Table 9.). However, as discussed later, this factor is considered to be unnecessarily conservative. A TRV for reptiles could not be derived due to a lack of toxicity data, however the derived wildlife TRV is probably also protective of reptiles. The approach used by the United States Army Center for Health Promotion and Preventative Medicine (USACHPPM, 2000) has been adapted for this assessment for both mammals and birds. Using the Approximation Approach, TRVs may be derived by extrapolating from a range of toxicity test endpoints (LC50, acute LOAELs) to derive two TRVs: a chronic NOAEL-based TRV and a chronic LOAEL-based TRV. As TRVs that are protective of wildlife health are required, only NOAEL-based TRVs have been derived for this assessment.
AF = Assessment Factor (refer Table 9.). USACHPPM (2000) has recommended the assessment factor (AFs i.e. safety or uncertainty factors) for use to derive NOAEL-based TRVs listed in Table 9.. The methodology is scientifically based, and is an important internationally published guide that describes the rationale and methods for deriving wildlife TRVs, particularly for situations such as site environmental risk assessments. Assessment (or uncertainty, or safety) factors are intended to account for potential differences in response between species, and differences in response due to exposure duration (e.g. acute vs chronic) and endpoints (e.g. LD50 vs NOAEL). They also account for differences between the controlled laboratory conditions under which toxicity tests are conducted and the generally harsh environmental conditions in which wildlife live. However, the assessment factors proposed in USACHPPM (2000) are applicable to small sets of toxicity data (the example given in the reference has a single bird toxicity study), whereas the avian and mammalian toxicity data for cyanide are somewhat larger and include a diversity of studies, though not large enough for probabilistic assessment and not all reliably meeting standard guidelines. It is therefore considered that lower assessment factors can be argued on a weight of evidence basis. Furthermore, cyanide is a highly acute poison and the acute to chronic toxicity ratio is likely to be relatively low compared to many other toxicants. Also, some of the species for which there is greatest concern (migratory birds) are likely to receive acute rather than chronic exposure. In general, the capacity for cyanide metabolism in animals means that a single, lethal dose of cyanide may not have lethal effects if that same quantity is ingested in smaller portions over time (Section 29.1.3). It is likely that a greater cumulative dose would be required for lethality. However, sublethal effects could also harm the bird’s subsequent survival, e.g. through the energy effects determined in biochemical studies and the effects on bird flight observed with pigeons (Section 34.1.1), or through greater predation susceptibility. For this assessment, it has been assumed for worst case exposure assessment that the animals evaluated consume all of their daily water requirements in one event. This is considered relevant to at least some species or individual animals in the environment. Table 9.. NOAEL-based AFs for use in deriving wildlife TRVs
For aquatic organisms, the freshwater and marine trigger values from the Australian water quality guidelines for the protection of aquatic ecosystems (ANZECC/ARMCANZ, 2000a, Section 81.3.1) will be used as the TRVs. 55.1.2Selection of Toxicity Reference Values (TRVs)56.Relevant forms of cyanideThe forms of cyanide relevant to this environmental assessment include free and metallocyanide complexes (e.g. WAD forms) of varying biological availability. It is considered relevant to environmental conditions that the derived TRVs for mammals and birds for cyanide are based on toxicity studies where cyanide was administered in a salt form rather than as organically-bound cyanide (e.g. diets of plants containing cyanogenic glycosides). The avian and mammalian TRVs presented below refer to doses administered in free cyanide forms (e.g. NaCN), and free cyanide is recognised as the most toxic and readily available form of cyanide. Very limited reliable toxicity data were available for metallocyanide complexes (e.g. WAD forms), which is particularly relevant due to the predominance of these types of complexes in tailings discharges at mining operations. Although there may be some environmental conditions where cyanide is present in metallocyanide complexes where the cyanide concentrations may not be as readily correlated to hazard due to lower bioavailability and delayed effects, many metallocyanide complexes (e.g. WAD cyanide) have been demonstrated to be biologically available to varying degrees and thus of toxicological relevance. Therefore, the extrapolation of TRVs based on free cyanide to environmental conditions involving free and available forms of metallocyanide complexes is considered applicable in the absence of chemical or site-specific data. In some instances, data suggest that the WAD CN component of tailings may be comprised of very high proportions of free CN (refer Section 23.6.4), but free CN content in the original water sampled may be significantly overestimated due to sampling and methodology difficulties (Schulz, pers. comm. 2006). Exposure to the dissolved metal may pose an additional hazard in these circumstances, but this review is focused on cyanide toxicity, and not on that of other components. 57.Avian toxicity dataDrinking water toxicity studies where birds were exposed to NaCN solutions at similar pH to gold mine process water were provided and would have been the studies of choice for selecting a TRV if they had not been so unreliable (Section 34.1.1). Similarly, a study with exposure to a range of cyanide concentrations in actual process water has been conducted, but the study report was not available and may or may not have been reliable. No other acceptable studies with dietary exposure of birds were available. The most useful studies for birds were acute oral exposure studies, which were generally conducted to a standard guideline, but are considered as acceptable rather than reliable, due to deficiencies in their design and reporting (Section 32.1.1). These covered a range of seven species, including two standard test species, three raptors, and two species known to be relatively insensitive to chemicals. Of these, mallard ducks were found to be the most sensitive. Data were available from additional studies with mallard ducks exposed to a single dose of tap or effluent pond water with various concentrations of cyanide or provided with contaminated water for a short period. Too few study details are available to be certain of the results and their interpretation and therefore, considered unreliable. Because of the availability of data for several species and additional data for the most sensitive species, and the nature of toxicity of cyanide as an acute toxin, it is considered that a satisfactory assessment factor for the avian toxicity endpoint is 10, rather than 100, as would be required according to Table 9.. Use of the 1.4 mg/kg bw LD50 value from Henny et al. (1994) and an AF of 10 for the purposes of this risk assessment results in the derivation of an avian TRV of 0.14 mg/kg bw. Sublethal toxic effects from the available acute oral toxicity studies have been studied in special studies with mallard ducks and pigeons, as well as being reported for LD50 studies. Ma and Pritsos (1997) indicated that tissue taken from mallard ducks killed two hours after they were dosed with 0.25 mg KCN/kg bw (0.1 mg CN/kg bw) showed statistically significant biochemical effects (i.e. brain and liver tissue ATP reduction) compared to untreated birds. However, these biochemical data cannot be related directly to effects of ecological relevance, and differences evident at 2 h post dosing were no longer evident at 24 h. Work by Cooper (2003) indicated that homing pigeon flight time was significantly increased when pigeons received a single oral dose of 1.25 mg KCN/kg bw ( 0.5 mg CN/kg bw) when compared to untreated control pigeons. Cooper (2003) also indicated that a single oral dose of 0.5 mg KCN/kg bw ( 0.2 mg CN/kg bw) produced no observed signs of toxicity in mallards and pigeons, whereas there was an early onset of such signs at doses ≥ 1.0 mg KCN/kg bw (≥ 0.4 mg CN/kg bw), in mallards at least. Hence, based on observable signs, the lowest reported NOEL value was 0.2 mg CN/kg bw for mallards, the most sensitive species of those tested based on acute oral toxicity. The TRV derived above is below the lowest NOEL for observable effects from this sequence of studies, though comparable to levels causing short term biochemical effects.
Acute oral studies with mammals were available for 6 species with NaCN, including a marsupial species (brushtail possum) as well as mice, white-footed mice, rats, and foxes. Further studies with some of these species were available with KCN and HCN. Thus there is a good body of data for acute oral exposure, though not enough species to warrant a probabilistic assessment approach. The lowest acute oral toxicity value is the LD50 for KCN to rabbits of 2.3 mg/kg bw (Ballantyne, 1987). Because of the availability of data for several species and the nature of toxicity of cyanide as an acute toxin where recovery occurs from sublethal doses, it is considered that a satisfactory assessment factor for the mammal toxicity endpoint is 10, rather than 100, as would be required according to Table 9.. Use of the 2.3 mg/kg bw LD50 value and an AF of 10 for the purposes of this risk assessment results in the derivation of a mammalian TRV of 0.23 mg/kg bw. 59.Aquatic toxicity dataAustralian water quality guidelines for the protection of aquatic ecosystems were published in 2000 (ANZECC/ARMCANZ, 2000a). The guidelines provide a decision-tree framework for the assessment of surface water quality, with provision for screening-level and more detailed levels of investigation, if required, to assess risks from substances in surface waters. The guidelines provide trigger values for free cyanide for the protection of aquatic life from screened acute toxicity data from tests conducted at different pHs (6.5 to 8.6) and temperatures (5°C to 30°C). The toxicity data used by ANZECC/ARMCANZ (2000a) to derive water quality trigger values were subjected to rigorous quality checks and review prior to adoption. Minimum data requirements were specified, leading to reliability estimates for the trigger values derived. All toxicity values were first converted to concentration as un-ionised HCN using the reported pH and temperature. The Australian freshwater and marine trigger values for cyanide of 7 µg CN/L and 4 µg CN/L, respectively, were calculated from LC50 values using a statistical distribution method allowing for a theoretical protection level of 95% of species, and an acute to chronic ratio (ACR) of 8.45 (rather than an AF of 10). These trigger values, which are appropriate for screening-level environmental assessments, are considered by ANZECC/ARMCANZ (2000a) to be of moderate reliability for protecting aquatic ecosystems. It is noted that both these values are above natural background levels (Section 70.1), as would be expected for them to be practicable. These values have therefore been adopted as the aquatic TRV values, as the exposure scenarios relate to situations where such screening level assessments are made, such as environmental impact assessments for NaCN manufacturing facilities and TSFs. The corresponding values to protect 99% of species were 4 μg CN/L and 2 μg CN/L: these values provide more protection for sub-lethal effects (see Table 9.) and could be used for protection of highly valued aquatic ecosystems or species.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||