Council Regulation (EC) No 440/2008 of 30 May 2008 laying down test methods pursuant to Regulation (EC) No 1907/2006 of the European Parliament and of the Council on the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) (Text with EEA relevance)

For strongly hydrophobic substances (log K OW > 5 and a solubility below ~ 0,01-0,1 mg/l), testing via aqueous exposure may become increasingly difficult. Reasons for constraints may be that the aqueous concentration cannot be maintained at a level that is considered to be sufficiently constant (e.g. due to sorption to the glass of exposure containers or rapid uptake by the fish) or that the aqueous concentrations to be applied are so low that they are in the same range as or below the analytical limit of quantification ([^E0084]). For these highly hydrophobic substances the dietary test is recommended, provided that the test is consistent with the relevant regulatory framework and risk assessment needs.

For surfactants it should be considered whether the aqueous bioconcentration test is feasible, given the substance properties, otherwise the dietary study is probably more appropriate. Surfactants are surface acting agents, which lower the interfacial tension between two liquids. Their amphiphilic nature (i.e. they contain both a hydrophilic and a hydrophobic part) causes them to accumulate at interfaces such as the water-air interface, the water-food interface, and glass walls, which hampers the determination of their aqueous concentration.

The dietary test can circumvent some of the exposure aspects for complex mixtures with components of differing water solubility limits, in that comparable exposure to all components of the mixture is more likely than in the aqueous method (cf. (20)).

It should be noted that the dietary approach yields a dietary biomagnification factor (BMF) rather than a bioconcentration factor (BCF) ([^E0085]). Approaches are available to estimate a kinetic bioconcentration factor (BCFK) from data generated in the dietary study (as discussed in Appendix 8, but these approaches should be used with caution. In general, these approaches assume first order kinetics, and are only applicable to certain groups of compounds. It is unlikely that such approaches can be applied for surfactants (see paragraph 12).

A minimised aqueous exposure test set-up with fewer sampling points to reduce the number of animals and/or resources (cf. paragraph 83 onwards) should only be applied to those substances where there is reason to expect that uptake and depuration will follow approximately first order kinetics (i.e. in general non-ionized organic substances, cf. paragraph 88).

C.13 - I:   AQUEOUS EXPOSURE BIOCONCENTRATION FISH TEST

PRINCIPLE OF THE TEST

The test consists of two phases: the exposure (uptake) and post-exposure (depuration) phases. During the uptake phase, a group of fish of one species is exposed to the test substance at one or more chosen concentrations, depending on the properties of the test substance (cf. paragraph 49). They are then transferred to a medium free of the test substance for the depuration phase. A depuration phase is always necessary unless uptake of the substance during the uptake phase has been insignificant. The concentration of the test substance in/on the fish (or specified tissue thereof) is followed through both phases of the test. In addition to the exposed group, a control group of fish is held under identical conditions except for the absence of the test substance, to relate possible adverse effects observed in the bioconcentration test to a matching control group and to obtain background concentrations of test substance ([^E0086]).

In the aqueous exposure test, the uptake phase is usually run for 28 days. The duration can be lengthened if necessary (cf. paragraph 18), or shortened if it is demonstrated that steady-state has been reached earlier (see Appendix 1, definitions and units). A prediction of the length of the uptake phase and the time to steady-state can be made from equations in Appendix 5. The depuration period is then begun when the fish are no longer exposed to the test substance, by transferring the fish to the same medium but without the test substance in a clean vessel. Where possible the bioconcentration factor is calculated preferably both as the ratio of concentration in the fish (C

1) and depuration (k 2) assuming first order kinetics ([^E0087]).

If a steady-state is not achieved within 28 days, either the BCF is calculated using the kinetic approach (cf. paragraph 38) or the uptake phase can be extended. Should this lead to an impractically long uptake phase to reach steady-state (cf. paragraphs 37 and 38, Appendix 5), the kinetic approach is preferred. Alternatively, for highly hydrophobic substances the conduction of a dietary study should be considered ([^E0088]), provided that the dietary test is consistent with the relevant regulatory framework.

The uptake rate constant, the depuration (loss) rate constant (or constants, where more complex models are involved), the bioconcentration factor (steady-state and/or kinetic), and where possible, the confidence limits of each of these parameters are calculated from the model that best describes the measured concentrations of test substance in fish and water (cf. Appendix 5).

The increase in fish mass during the test will result in a decrease of test substance concentration in growing fish (so-called growth dilution), and thus the kinetic BCF will be underestimated if not corrected for growth (cf. paragraphs 72 and 73).

The BCF is based on the total concentration in the fish (i.e. per total wet weight of the fish). However, for special purposes, specified tissues or organs (e.g. muscle, liver), may be used if the fish are sufficiently large or the fish may be divided into edible (fillet) and non-edible (viscera) fractions. Since, for many organic substances, there is a clear relationship between the potential for bioconcentration and hydrophobicity, there is also a corresponding relationship between the lipid content of the test fish and the observed bioconcentration of such substances. Thus, to reduce this source of variability in test results for those substances with high lipophilicity (i.e. with log K OW > 3), bioconcentration should be expressed as normalised to a fish with a 5 % lipid content (based on whole body wet weight) in addition to that derived directly from the study. This is necessary to provide a basis from which results for different substances and/or test species can be compared against one another. The figure of 5 % lipid content has been widely used as this represents the average lipid content of fish commonly used in this test method (21).

INFORMATION ON THE TEST SUBSTANCE

In addition to the properties of the test substance given in the Introduction (paragraph 3), other information required is the toxicity to the fish species to be used in the test, preferably the asymptotic LC50 (i.e. time-independent) and/or toxicity estimated from long-term fish tests (e.g. TMs C.47 (22), C.15 (23), C.14 (24)).

An appropriate analytical method, of known accuracy, precision, and sensitivity, for the quantification of the substance in the test solutions and in biological material should be available, together with details of sample preparation and storage. The analytical quantification limit of the test substance in both water and fish tissues should also be known. When a radiolabelled test substance is used, it should be of the highest purity (e.g. preferably > 98 %) and the percentage of radioactivity associated with impurities should be known.

VALIDITY OF THE TEST

For a test to be valid the following conditions apply:

The water temperature variation is less than ± 2 °C, because large deviations can affect biological parameters relevant for uptake and depuration as well as cause stress to animals;

The concentration of dissolved oxygen does not fall below 60 % saturation;

The concentration of the test substance in the chambers is maintained within ± 20 % of the mean of the measured values during the uptake phase;

The concentration of the test substance is below its limit of solubility in water, taking into account the effect that the test water may have on effective solubility ([^E0089]);

The mortality or other adverse effects/disease in both control and treated fish is less than 10 % at the end of the test; where the test is extended over several weeks or months, death or other adverse effects in both sets of fish should be less than 5 % per month and not exceed 30 % in all. Significant differences in average growth between the test and the control groups of sampled fish could be an indication of a toxic effect of the test substance.

REFERENCE SUBSTANCES

The use of reference substances of known bioconcentration potential and low metabolism would be useful in checking the experimental procedure, when required (e.g. when a laboratory has no previous experience with the test or experimental conditions have been changed).

DESCRIPTION OF THE METHOD

Apparatus

Care should be taken to avoid the use of materials — for all parts of the equipment — that can dissolve, sorb or leach and have an adverse effect on the fish. Standard rectangular or cylindrical tanks, made of chemically inert material and of a suitable capacity in compliance with loading rate (cf. paragraph 43), can be used. The use of soft plastic tubing should be minimised. Polytetrafluoroetheylene, stainless steel and/or glass tubing should be used. Experience has shown that for test substances with high adsorption coefficient, such as the synthetic pyrethroids, silanised glass may be required. In such situations the equipment should be discarded after use. It is preferable to expose test systems to concentrations of the test substance to be used in the study for as long as is required to demonstrate the maintenance of stable exposure concentrations prior to the introduction of test organisms.

Water

Natural water is generally used in the test and should be obtained from uncontaminated and uniform quality source. Yet, reconstituted water (i.e. demineralised water with specific nutrients added in known amounts) may be more suitable to guarantee uniform quality over time. The dilution water, which is the water that is mixed with the test substance before entering the test vessel (cf. paragraph 30), should be of a quality that will allow the survival of the chosen fish species for the duration of the acclimation and test periods without them showing any abnormal appearance or behaviour. Ideally, it should be demonstrated that the test species can survive, grow and reproduce in the dilution water (e.g. in laboratory culture or a life-cycle toxicity test). The dilution water should be characterised at least by pH, hardness, total solids, total organic carbon (TOC ([^E0090])) and, preferably also ammonium, nitrite and alkalinity and, for marine species, salinity. The parameters which are important for optimal fish well-being are not fully known, but Appendix 2 gives recommended maximum concentrations of a number of parameters for fresh and marine test waters.

The dilution water should be of constant quality during the period of a test. The pH value should be within the range 6,0 to 8,5 at test start, but during a given test it should be within a range of ± 0,5 pH units. In order to ensure that the dilution water will not unduly influence the test result (for example, by complexation of the test substance) or adversely affect the performance of the stock of fish, samples should be taken at intervals for analysis, at least at the beginning and end of the test. Determination of heavy metals (e.g. Cu, Pb, Zn, Hg, Cd, and Ni), major anions and cations (e.g. Ca²⁺, Mg²⁺, Na⁺, K⁺, Cl^–, and SO4 ^2–), pesticides (e.g. total organophosphorous and total organochlorine pesticides), total organic carbon and suspended solids should be conducted, for example, every three months where dilution water is known to be relatively constant in quality. If dilution water quality has been demonstrated to be constant over at least one year, determinations can be less frequent and intervals extended (e.g. every six months).

The natural particle content as well as the total organic carbon of the dilution water should be as low as possible to avoid adsorption of the test substance to organic matter, which may reduce its bioavailability and therewith result in an underestimation of the BCF. The maximum acceptable value is 5 mg/l for particulate matter (dry matter, not passing a 0,45 μm filter) and 2 mg/l for total organic carbon (cf. Appendix 2). If necessary, the dilution water should be filtered before use. The contribution to the organic carbon content in test water from the test fish (excreta) and from the food residues should be kept as low as possible (cf. paragraph 46).

Test Solutions

Prepare a stock solution of the test substance at a suitable concentration. The stock solution should preferably be prepared by simply mixing or agitating the test substance in the dilution water. An alternative that may be appropriate in some cases is the use of a solid phase desorption dosing system. The use of solvents and dispersants (solubilising agents) is not generally recommended (cf. (25)); however, the use of these materials may be acceptable in order to produce a suitably concentrated stock solution, but every effort should be made to minimise the use of such materials and their critical micelle concentration should not be exceeded (if relevant). Solvents which may be used are acetone, ethanol, methanol, dimethyl formamide and triethylene glycol; dispersants that have been used are Tween 80, methylcellulose 0,01 % and HCO-40. The solvent concentration in the final test medium should be the same in all treatments (i.e. regardless of test substance concentration) and should not exceed the corresponding toxicity thresholds determined for the solvent under the test conditions. The maximum level is a concentration of 100 mg/l (or 0,1 ml/l). It is unlikely that a solvent concentration of 100 mg/l will significantly alter the maximum dissolved concentration of the test substance which can be achieved in the medium (25). The solvent's contribution (together with the test substance) to the overall content of organic carbon in the test water should be known. Throughout the test, the concentration of total organic carbon in the test vessels should not exceed the concentration of organic carbon originating from the test substance, and solvent or solubilising agent ([^E0091]), if used, by more than 10 mg/l (± 20 %). Organic matter content can have a significant effect on the amount of freely dissolved test substance during flow-through fish tests, especially for highly lipophilic substances. Solid-phase microextraction (cf. paragraph 60) can provide important information on the ratio between bound and freely dissolved compounds, of which the latter is assumed to represent the bioavailable fraction. The test substance concentration should be below the solubility limit of the test substance in the test media in spite of the use of a solvent or solubilising agent. Care should be taken when using readily biodegradable solvents as these can cause problems with bacterial growth in flow-through tests. If it is not possible to prepare a stock solution without the use of a solubilising agent, consideration should be given to the appropriateness of an aqueous exposure study as opposed to a dietary exposure study.

For flow-through tests, a system which continuously dispenses and dilutes a stock solution of the test substance (e.g. metering pump, proportional diluter, saturator system) or a solid phase desorption dosing system is required to deliver the test concentrations to the test chambers. Preferably allow at least five volume replacements through each test chamber per day. The flow-through mode is to be preferred, but where this is not possible (e.g. when the test organisms are adversely affected) a semi-static technique may be used provided that the validity criteria are satisfied (cf. paragraph 24). The flow rates of stock solutions and dilution water should be checked both 48 hours before and then at least daily during the test. Include in this check the determination of the flow-rate through each test chamber and ensure that it does not vary by more than 20 % either within or between chambers.

Selection of species

Important criteria in the selection of species are that they are readily available, can be obtained in convenient sizes and can be satisfactorily maintained in the laboratory. Other criteria for selecting fish species include recreational, commercial, ecological importance as well as comparable sensitivity, past successful use etc. Recommended test species are given in Appendix 3. Other species may be used but the test procedure may have to be adapted to provide suitable test conditions. The rationale for the selection of the species and the experimental method should be reported in this case. In general, the use of smaller fish species will shorten the time to steady-state, but more fish (samples) may be needed to adequately analyse lipid content and test substance concentrations in the fish. In addition it is possible that differences in respiration rate and metabolism between young and older fish may hamper comparisons of results between different tests and test species. It should be noted that fish species tested during a (juvenile) life-stage with rapid growth can complicate data interpretation.

Holding of fish (relevant for aqueous and dietary exposure)

The stock population of fish should be acclimated for at least two weeks in water (cf. paragraph 28) at the test temperature and feed throughout on a sufficient diet (cf. paragraph 45). Both water and diet should be of the same type as those to be used during the test.

Following a 48-hour settling-in period, mortalities are recorded and the following criteria applied:

• — Mortalities exceeding 10 % of the population in seven days: reject the entire batch;

• — Mortalities of between 5 and 10 % of the population in seven days: acclimate for seven additional days — if more than 5 % mortality during the second seven days, reject the entire batch;

• — Mortalities below 5 % of the population in seven days: accept the batch.

Fish used in tests should be free from observable diseases and abnormalities. Any diseased fish should be discarded. Fish should not receive treatment for disease in the two weeks preceding the test, or during the test.

PERFORMANCE OF THE TEST

Preliminary test

It may be useful to conduct a preliminary experiment in order to optimise the test conditions of the definitive test, e.g. selection of test substance concentration(s), duration of the uptake and depuration phases, or to determine whether a full test need be conducted. The design of the preliminary test should be such as to obtain the information required. It can be considered if a minimised test may be sufficient to derive a BCF, or if a full study is needed (cf. paragraphs 83-95 on the minimised test).

Conditions of Exposure

Duration of uptake phase

A prediction of the duration of the uptake phase can be obtained from practical experience (e.g. from a previous study or an accumulation study on a structurally related substance) or from certain empirical relationships utilising knowledge of either the aqueous solubility or the octanol/water partition coefficient of the test substance (provided that uptake follows first order kinetics, cf. Appendix 5).

The uptake phase should be run for 28 days unless it can be demonstrated that steady-state has been reached earlier (see Appendix 1, definitions and units). A steady-state is reached in the plot of test substance in fish (C

f made on samples taken at intervals of at least two days are within ± 20 % of each other, and there is no significant increase of C f in time between the first a last successive analysis. When pooled samples are analysed, at least four successive analyses are required. For test substances which are taken up slowly the intervals would more appropriately be seven days. If steady-state has not been reached by 28 days, either the BCF is calculated using only the kinetic approach, which is not reliant on steady-state being reached, or the uptake phase can be extended, taking further measurements, until steady-state is reached or for 60 days, whichever is shorter. Also, the test substance concentration in the fish at the end of the uptake phase needs to be sufficiently high to ensure a reliable estimation of k 2 from the depuration phase. If no significant uptake is shown after 28 days, the test can be stopped.

Duration of the depuration phase

For substances following first order kinetics, a period of half the duration of the uptake phase is usually sufficient for an appropriate (e.g. 95 %) reduction in the body burden of the substance to occur (cf. Appendix 5 for explanation of the estimation). If the time required to reach 95 % loss is impractically long, exceeding for example twice the normal duration of the uptake phase (i.e. more than 56 days) a shorter period may be used (e.g. until the concentration of test substance is less than 10 % of steady-state concentration). However, longer depuration periods may be necessary for substances having more complex patterns of uptake and depuration than are represented by a one-compartment fish model that yields first order kinetics. If such complex patterns are observed and/or anticipated, it is advised to seek advice from a biostatistician and/or pharmacokineticist to ensure a proper test set-up. As the depuration period is extended, numbers of fish to sample may become limiting and growth differences between fish can influence the results. The period will also be governed by the period over which the concentration of the test substance in the fish remains above the analytical limit of quantification.

Numbers of test fish

Select the numbers of fish per test concentration such that a minimum of four fish are available at each sampling point. Fish should only be pooled if analysis of single fish is not feasible. If higher precision in curve fitting (and derived parameters) is intended or if metabolism studies are required (e.g. to distinguish between metabolites and parent substance when using radiolabelled test substances), more fish per sampling point will be necessary. The lipid content should be determined on the same biological material as is used to determine the concentration of the test substance. Should this not be feasible, additional fish may be needed (cf. paragraphs 56 and 57).

If adult (i.e. sexually mature) fish are used, they should not be in a spawning state or recently spent (i.e. already spawned) either before or during the test. It should also be reported whether male or female, or both are used in the experiment. If both sexes are used, differences in growth and lipid content between sexes should be documented to be non-significant before the start of the exposure, in particular if it is anticipated that pooling of male and female fish will be necessary to ensure detectable substance concentrations and/or lipid content.

In any one test, select fish of similar weight such that the smallest are no smaller than two-thirds of the weight of the largest. All should be of the same year-class and come from the same source. Since weight and age of a fish may have a significant effect on BCF values (12) these details should be recorded accurately. It is recommended that a sub-sample of the stock of fish is weighed shortly before the start of the test in order to estimate the mean weight (cf. paragraph 61).

Loading

High water-to-fish ratios should be used in order to minimise the reduction in the concentration of the test compound in water caused by the addition of the fish at the start of the test and also to avoid decreases in dissolved oxygen concentration. It is important that the loading rate is appropriate for the test species used. In any case, a fish-to-water loading rate of 0,1-1,0 g of fish (wet weight) per litre of water per day is normally recommended. Higher fish-to-water loading rates can be used if it is shown that the required concentration of test substance can be maintained within ± 20 % limits, and that the concentration of dissolved oxygen does not fall below 60 % saturation (cf. paragraph 24).

In choosing appropriate loading regimes, take into account the normal habitat of the fish species. For example, bottom-living fish may demand a larger bottom area of the aquarium for the same volume of water compared to pelagic fish species.

Feeding

During the acclimation and test periods, feed an appropriate diet of known lipid and total protein content to the fish in an amount sufficient to keep them in a healthy condition and to maintain body weight (some growth is allowed). Feed daily throughout the acclimation and test periods at a set level depending on the species used, experimental conditions and calorific value of the food (for example for rainbow trout between approximately 1 to 2 % of body weight per day). The feeding rate should be selected such that fast growth and large increase of lipid content are avoided. To maintain the same feeding rate, the amount of feed should be re-calculated as appropriate, for example once per week. For this calculation, the weight of the fish in each test chamber can be estimated from the weight of the fish sampled most recently in that chamber. Do not weigh the fish remaining in the chamber.

Uneaten food and faeces should be siphoned daily from the test chambers shortly after feeding (30 minutes to one hour). The chambers should be kept as clean as possible throughout the test to keep the concentration of organic matter as low as possible (cf. paragraph 29), since the presence of organic carbon may limit the bioavailability of the test substance (12).

Since many feeds are derived from fishmeal, it should be ensured that the feed will not influence the test results or induce adverse effects, e.g. by containing (traces of) pesticides, heavy metals and/or the test substance itself.

Light and temperature

A 12- to 16-hour photoperiod is recommended and the temperature (± 2 °C) should be appropriate for the test species (cf. Appendix 3). The type and characteristics of illumination should be known. Caution should be given to the possible phototransformation of the test substance under the irradiation conditions of the study. Appropriate illumination should be used avoiding exposure of fish to unnatural photoproducts. In some cases it may be appropriate to use a filter to screen out UV irradiation below 290 nm.

Test concentrations

The test was originally designed for non-polar organic substances. For this type of substance, the exposure of fish to a single concentration is expected to be sufficient, as no concentration effects are expected, although two concentrations may be required for the relevant regulatory framework. If substances outside this domain are tested, or other indications of possible concentration dependence are known, the test should be run with two or more concentrations. If only one concentration is tested, justification for the use of one concentration should be given (cf. paragraph 79). Also, the tested concentration should be as low as is practical or technically possible (i.e. not close to the solubility limit).

In some cases it can be anticipated that the bioconcentration of a substance is dependent on the water concentration (e.g. for metals, where the uptake in fish may be at least partly regulated). In such a case it is necessary that at least two, but preferably more, concentrations are tested (cf. paragraph 49) which are environmentally relevant. Also for substances where the concentrations tested have to be near the solubility limit for practical reasons, testing at least two concentrations is recommended, because this can give insight into the reliability of the exposure concentrations. The choice of the test concentrations should incorporate the environmentally realistic concentration as well as the concentration that is relevant to the purpose of the specific assessment.

The concentration(s) of the test substance should be selected to be below its chronic effect level or 1 % of its acute asymptotic LC50, within an environmentally relevant range and at least an order of magnitude above its limit of quantification in water by the analytical method used. The highest permissible test concentration can also be determined by dividing the acute 96 h LC50 by an appropriate acute/ chronic ratio (e.g. appropriate ratios for some substances are about three, but a few are above 100). If a second concentration is used, it should differ from the one above by a factor of ten. If this is not possible because of the toxicity criterion (that limits the upper test concentration) and the analytical limit (that limits the lower test concentration), a lower factor than ten can be used and use of radiolabelled test substance (of the highest purity, e.g. preferably > 98 %) should be considered. Care should be taken that no concentration used is above the solubility limit of the test substance in the test media.

Controls

One dilution water control or if relevant (cf. paragraphs 30 and 31), one control containing the solvent should be run in addition to the test series.

Frequency of Water Quality Measurements

During the test, dissolved oxygen, TOC, pH and temperature should be measured in all test and control vessels. Total hardness and salinity (if relevant) should be measured in the control(s) and one vessel. If two or more concentrations are tested, measure these parameters at the higher (or highest) concentration. As a minimum, dissolved oxygen and salinity (if relevant) should be measured three times — at the beginning, around the middle and end of the uptake period — and once a week in the depuration period. TOC should be measured at the beginning of the test (24 h and 48 h prior to test initiation of uptake phase) before addition of the fish and at least once a week during both uptake and depuration phases. Temperature should be measured and recorded daily, pH at the beginning and end of each period and hardness once each test. Temperature should preferably be monitored continuously in at least one vessel.

Sampling and Analysis of Fish and Water

Fish and water sampling schedule

Water should be sampled from the test chambers for the determination of test substance concentration before addition of the fish and during both uptake and depuration phases. The water should be sampled the before feeding, at the same time as the fish sampling. More frequent sampling may be useful to ensure stable concentrations after introduction of the fish. During the uptake phase, the concentrations of test substance should be determined in order to check compliance with the validity criteria (paragraph 24). If water sample analyses at the beginning of the depuration phase show that the test substance is not detected, this can be used as a justification not to measure test and control water for the test substance for the remainder of the depuration phase.

Fish should be sampled on at least five occasions during the uptake phase and on at least four occasions during the depuration phase for test substance. Since on some occasions it will be difficult to calculate a reasonably precise estimate of the BCF value based on this number of samples (especially when other than simple first order uptake and depuration kinetics are indicated), it may be advisable to take samples at a higher frequency in both periods (cf. Appendix 4).

The lipid content should be determined on the same biological material as is used to determine the concentration of the test substance at least at the start and end of the uptake phase and at the end of the depuration phase. Should this not be feasible, at least three independent fish should be sampled to determine lipid content at each of the same three time-points. The number of fish per tank at the start of the experiment should be adjusted accordingly ([^E0092]). Alternatively, if no significant amounts of the test substance are detected in control fish (i.e. fish from the stock population), the control fish from the test can be analysed for lipid content only and test substance analysis in the test group(s) (and the related uptake rate constant, depuration rate constant and BCF values) can be corrected for changes according to control group lipid content during the test ([^E0093]).

Dead or diseased fish should not be analysed for test substance or lipid concentration.

An example of an acceptable sampling schedule is given in Appendix 4. Other schedules can readily be calculated using other assumed values of K OW to calculate the exposure time for 95 % uptake (refer to Appendix 5 for calculations).

Sampling should be continued during the uptake phase until a steady-state has been established (see Appendix 1, definitions and units) or the uptake phase is otherwise terminated (after 28 or 60 days, cf. paragraphs 37 and 38). Before beginning the depuration phase, the fish should be transferred to clean vessels.

Sampling and sample preparation

Water samples should be obtained for analysis e.g. by siphoning through inert tubing from a central point in the test chamber. Neither filtration nor centrifuging appears always to separate the non-bioavailable fraction of the test substance from that which is bioavailable. If a separation technique is applied, a justification for, or validation of, the separation technique should always be provided in the test report given the bioavailability difficulties (25). Especially for highly hydrophobic substances (i.e. those substances with a log K OW > 5) (12) (26), where adsorption to the filter matrix or centrifugation containers could occur, samples should not be subjected to those treatments. Instead, measures should be taken to keep the tanks as clean as possible (cf. paragraph 46) and the content of total organic carbon should be monitored during both the uptake and depuration phases (cf. paragraph 53). To avoid possible issues with reduced bioavailability, sampling by solid phase microextraction techniques may be used for poorly soluble and highly hydrophobic substances.

The sampled fish should be euthanised instantly, using the most appropriate and humane method (for whole fish measurements, no further processes than rinsing with water (cf. paragraph 28) and blot drying the fish should be done). Weigh and measure total length ([^E0094]). In each individual fish, the measured weight and length should be linked to the analysed substance concentration (and lipid content, if applicable), for example using a unique identifier code for each sampled fish.

It is preferable to analyse fish and water immediately after sampling in order to prevent degradation or other losses and to calculate approximate uptake and depuration rate constants as the test proceeds. Immediate analysis also avoids delay in determining when a plateau (steady-state) has been reached.

Failing immediate analysis, the samples should be stored by an appropriate method. Before the beginning of the study, information should be obtained on the proper method of storage for the particular test substance — for example, deep-freezing, holding at 4 °C, extraction, etc. The duration of storage should be selected to ensure that the substance has not degraded while in storage.

Quality of analytical method

Since the whole procedure is governed essentially by the accuracy, precision and sensitivity of the analytical method used for the test substance, check experimentally that the accuracy, precision and reproducibility of the substance analysis, as well as recovery of the test substance from both water and fish are satisfactory for the particular method. This should be part of preliminary tests. Also, check that the test substance is not detectable in the dilution water used. If necessary, correct the values of test substance concentration in water and fish obtained from the test for the recoveries and background values of controls. The fish and water samples should be handled throughout in such a manner as to minimise contamination and loss (e.g. resulting from adsorption by the sampling device).

Analysis of fish samples

If radiolabelled materials are used in the test, it is possible to analyse for total radiolabel (i.e. parent and metabolites) or the samples may be cleaned up so that the parent substance can be analysed separately. If the BCF is to be based on the parent substance, the major metabolites should be characterised, as a minimum at the end of the uptake phase (cf. paragraph 6). Major metabolites are those representing ≥ 10 % of total residues in fish tissues, those representing ≥ 5 % at two consecutive sampling points, those showing increasing levels throughout the uptake phase, and those of known toxicological concern. If the BCF for the whole fish in terms of total radiolabelled residues is ≥ 500, it may be advisable — and for certain categories of substances such as pesticides strongly recommended — identifying and quantifying major metabolites. Quantification of such metabolites may be required by some regulatory authorities. If degradates representing ≥ 10 % of total radiolabelled residues in the fish tissue are identified and quantified, then it is also recommended to identify and quantify degradates in the test water. Should this not be feasible, this should be explained in the report.

The concentration of the test substance should usually be determined for each weighed individual fish. If this is not possible, pooling of the samples on each sampling occasion may be done but pooling does restrict the statistical procedures which can be applied to the data, so an adequate number of fish to accommodate the desired pooling, statistical procedure and power should be included in the test. References (27) and (28) may be used as an introduction to relevant pooling procedures.

BCF should be expressed as normalised to a fish with a 5 % lipid content (based on wet weight) in addition to that derived directly from the study (cf .paragraph 21), unless it can be argued that the test substance does not accumulate primarily in lipids. The lipid content of the fish should be determined on each sampling occasion if possible, preferably on the same extract as that produced for analysis for the test substance, since the lipids often have to be removed from the extract before it can be analysed chromatographically. However, analysis of test substances often requires specific extraction procedures which might be in contradiction to the test methods for lipid determination. In this case (until suitable non-destructive instrumental methods are available), it is recommended to employ a different strategy to determine the fish lipid content (cf. paragraph 56). Suitable methods should be used for determination of lipid content (20). The chloroform/methanol extraction technique (29) may be recommended as standard method (30), but the Smedes-method (31) is recommended as an alternative technique. This latter method is characterised by a comparable efficiency of extraction, high accuracy, the use of less toxic organic solvents and ease of performance. Other methods for which accuracy compares favourably to the recommended methods could be used if properly justified. It is important to give details of the method used.

Fish growth measurement

At the start of the test, five to ten fish from the stock population need to be weighed individually and their total length measured. These can be the same fish used for lipid analysis (cf. paragraph 56). The weight and length of fish used for each sampling event from both test and control groups should be measured before chemical or lipid analysis is conducted. The measurements of these sampled fish can be used to estimate the weight and length of fish remaining in the test and control tanks (cf. paragraph 45).

DATA AND REPORTING

Treatment of results

The uptake curve of the test substance should be obtained by plotting its concentration in/on fish (or specified tissues) in the uptake phase against time on arithmetic scales. If the curve has reached a plateau, that is, become approximately asymptotic to the time axis, the steady-state BCF (BCFSS) should be calculated from:

Image omitted; see official source

The development of C f may be influenced by fish growth (cf. paragraphs 72 and 73). The mean exposure concentration (Cw) is influenced by variation over time. It can be expected that a time-weighted average concentration is more relevant and precise for bioaccumulation studies, even if variation is within the appropriate validity range (cf. paragraph 24). A time weighted average (TWA) water concentration can be calculated according to Appendix 5, section 1.

The kinetic bioconcentration factor (BCFK) should be determined as the ratio k 1/k 2, the two first order kinetic rate constants. Rate constants k 1 and k 2 and BCFK can be derived by simultaneously fitting both the uptake and the depuration phase. Alternatively, k 1 and k 2 can be determined sequentially (see Appendix 5 for a description and comparison of these methods). The depuration rate constant (k 2) may need correction for growth dilution (cf. paragraphs 72 and 73). If the uptake and/or depuration curve is obviously not first order, then more complex models should be employed (see references in Appendix 5 and advice from a biostatistician and/or pharmacokineticist sought.

Fish weight/length data

Individual fish wet weights and total lengths for all sampling intervals are tabulated separately for test and control groups during the uptake (including stock population for start of uptake) and depuration phases. In each individual fish the measured weight and length should be linked to the analysed chemical concentration, for example using a unique identifier code for each sampled fish. Weight is the preferred measure of growth for the purposes of correcting kinetic BCF values for growth dilution (see paragraph 73 and Appendix 5 for the method used to correct data for growth dilution).

Growth-Dilution Correction and Lipid Normalisation

Fish growth during the depuration phase can lower measured chemical concentrations in the fish with the effect that the overall depuration rate constant (k 2) is greater than would arise from removal processes (e.g. respiration, metabolism, egestion) alone. Kinetic bioconcentration factors should be corrected for growth dilution. A BCFSS will also be influenced by growth, but no agreed procedure is available to correct a BCFSS for growth. In cases of significant growth, the BCFK, corrected for growth (BCFKg), should also be derived as it may be a more relevant measure of the bioconcentration factor. Lipid contents of test fish (which are strongly associated with the bioaccumulation of hydrophobic substances) can vary enough in practice such that normalisation to a set fish lipid content (5 % w/w) is necessary to present both kinetic and steady-state bioconcentration factors in a meaningful way, unless it can be argued that the test substance does not primarily accumulate in lipid (e.g. some perfluorinated substances may bind to proteins). Equations and examples for these calculations can be found in Appendix 5.

To correct a kinetic BCF for growth dilution, the depuration rate constant should be corrected for growth. This growth-corrected depuration rate constant (k 2g) is calculated by subtracting the growth rate constant (k g, as obtained from the measured weight data) from the overall depuration rate constant (k 2). The growth-corrected kinetic bioconcentration factor is then calculated by dividing the uptake rate constant (k 1) by the growth-corrected depuration rate constant (k 2g) (cf. Appendix 5). In some cases this approach is compromised. For example, for very slowly depurating substances tested in fast growing fish, the derived k 2g may be very small and so the error in the two rate constants used to derive it becomes critical, and in some cases kg estimates can be larger than k

2 should still be reported even if this alternative approach is used.

Kinetic and steady-state bioconcentration factors should also be reported relative to a default fish lipid content of 5 % (w/w), unless it can be argued that the test substance does not primarily accumulate in lipid. Fish concentration data, or the BCF, are normalised according to the ratio between 5 % and the actual (individual) mean lipid content (in % wet weight) (cf. Appendix 5).

If chemical and lipid analyses have been conducted on the same fish, then individual fish lipid normalised data should be used to calculate a lipid-normalised BCF. Alternatively, if the growth in control and exposed fish is similar, the lipid content of control fish alone may be used for lipid-correction (cf. paragraph 56). A method for calculating a lipid-normalised BCF is described in Appendix 5.

Interpretation of results

The results should be interpreted with caution where measured concentrations of test solutions occur at levels near the detection limit of the analytical method.

Average growth in both test and control groups should in principle not be significantly different to exclude toxic effects. The growth rate constants or the growth curves of the two groups should be compared by an appropriate procedure ([^E0095])).

Clearly defined uptake and depuration curves are an indication of good quality bioconcentration data. For the rate constants, the result of a χ² goodness-of-fit-test should show a good fit (i.e. small measurement error percentage (32)) for the bioaccumulation model, so that the rate constants can be considered reliable (cf. Appendix 5). If more than one test concentration is used, the variation in uptake/depuration constants between the test concentrations should be less than 20 % ([^E0096]). If not, concentration dependence could be indicated. Observed significant differences in uptake/ depuration rate constants between the applied test concentrations should be recorded and possible explanations given. Generally, the 95 % confidence limit of BCFs from well-designed studies approach ± 20 % of the derived BCF.

If two or more concentrations are tested, the results of both or all concentrations are used to examine whether the results are consistent and to show whether there is concentration dependence. If only one concentration is tested to reduce the use of animals and/or resources, justification of the use of one concentration should be given.

The resulting BCFSS is doubtful if the BCFK is significantly larger than the BCFSS, as this can be an indication that steady-state has not been reached or growth dilution and loss processes have not been taken into account. In cases where the BCFSS is very much higher than the BCFK, the derivation of the uptake and depuration rate constants should be checked for errors and re-evaluated. A different fitting procedure might improve the estimate of BCFK (cf. Appendix 5).

Test Report

Apart from the test substance information indicated in paragraph 3, the test report includes the following information:

Test substance:

Physical nature and, where relevant, physicochemical properties;

• — Chemical identification data, such as IUPAC or CAS name, CAS number, SMILES or InChI code, structural formula, purity, chemical identity of impurities as appropriate and practically feasible, etc. (including the organic carbon content, if appropriate).

• — For multi-constituent substances and UVCB (chemical substances of Unknown or Variable composition, Complex reaction products and Biological materials) description, as far as possible, of the chemical identity of the individual constituents and, for each, of its percentage of the total mass of the substance. How the analytical method used in the test reflects a measure of the concentration of the substance should be summarised; all analytical procedures should be described including the accuracy of the method, method detection limit, and limit of quantification.

• — If radiolabelled, the precise position of the labelled atom(s) and the percentage of radioactivity associated with impurities.

• — Information on the test substance toxicity to fish (ideally the test species). The toxicity should be reported as an acute 96 h LC50 and a NOAEC & LOAEC from a chronic study (i.e., an early life stage test or a full life cycle test, if available).

• — Storage conditions of the test chemical or test substance and stability of the test chemical or test substance under storage conditions if stored prior to use.

Test species:

Scientific name, strain, source, any pre-treatment, acclimation, age, sex (if relevant), size-range (weight and length), etc.

Test conditions:

• — Test procedure used (e.g. flow-through or semi-static); regular study or minimised design (including rationale and justification).

• — Type and characteristics of illumination used and photoperiod(s).

• — Test design (e.g. number and size of test chambers, water volume replacement rate, loading rate, number of replicates, number of fish per replicate, number of test concentrations, length of uptake and depuration phases, sampling frequency for fish and water samples).

• — Method of preparation of stock solutions and frequency of renewal (the solvent, its concentration and its contribution to the organic carbon content of test water should be given, when used) or description of alternative dosing system.

• — The nominal test concentrations, the means of the measured values and their standard deviations in the test vessels and the method and frequency by which these were attained.

• — Source of the dilution water, description of any pre-treatment, results of any demonstration of the ability of test fish to live in the water, and water characteristics: pH, hardness, temperature, dissolved oxygen concentration, residual chlorine levels (if measured), total organic carbon, suspended solids, salinity of the test medium (if appropriate) and any other measurements made.

• — Water quality within test vessels, pH, hardness, TOC, temperature and dissolved oxygen concentration; methods used and frequency of measurements.

• — Detailed information on feeding, e.g. type of food(s), source, composition (at least lipid and protein content if possible), selected feeding rate, amount given and frequency;

• — Information on the treatment of fish and water samples, including details of preparation, storage, extraction and analytical procedures (and precision) for the test substance and lipid content.

• — Methods used for treatment randomisation and assignment of fish to test vessels.

• — Date of introduction of test organisms to test solutions and test duration.

• — Description of range-finding tests and results, if available.

Results:

• — Results from any preliminary study performed.

• — Mortality of the control fish and the fish in each exposure chamber and any observed abnormal behaviour.

• — Information on any adverse effects observed.

• — Complete description of all chemical analysis procedures employed including limits of detection and quantification, variability and recovery.

• — The lipid content of the fish, including the method used, and if derived, lipid normalisation factor (L n, factor to express results relative to fish lipid content of 5 %).

• — Tabulated fish weight (and length) data, linked to individual fish chemical concentrations (and lipid content, if applicable), both for control and exposure groups (for example using unique identifiers for each sampled fish) and calculations for derived growth rate constant(s).

• — Tabulated test substance concentration data in fish (C f, linked to individual fish) and water (C

f expressed in mg/kg wet weight of whole body or specified tissues thereof, e.g. lipid, and C w in mg/l). C w values for the control series (background should also be reported).

• — Curves (including all measured data), showing the following (if applicable, concentrations may be expressed in relation to the whole body and the lipid content normalised to 5 % of the animal or specified tissues thereof): — growth, i.e. fish weight vs. time or natural logarithm transformed weight vs. time (including the derived growth rate constant, k g); — the uptake and depuration of the test substance in the fish (on one graph); — the time to steady-state (if achieved); — natural logarithm transformed concentration vs. uptake time (including the derived uptake rate constant k 1); — natural logarithm transformed concentration (ln concentration) vs. depuration time (including the derived depuration rate constant k 2); and — both uptake and depuration phase curves, showing both the data and the fitted model.

• — If a visual inspection of a plot shows obvious outliers, a statistically valid outlier test may be applied to remove spurious data points as well as documented justification for their omission.

• — The steady-state bioconcentration factor, (BCFSS), if steady-state is (almost) achieved.

• — Kinetic bioconcentration factor (BCFK) and derived uptake and depuration rate constants k 1 and k 2, together with the variances in k 2 (slope and intercept) if sequential fitting is used.

• — Confidence limits, standard deviation (as available) and methods of computation/data analysis for each parameter for each concentration of test substance used. — Any information concerning radiolabelled test substance metabolites and their accumulation. — Growth rate constant(s) (including 95 % confidence interval(s)) and calculated growth-corrected depuration rate constant (k 2g), half-life and BCF (BCFKg) values. — Anything unusual about the test, any deviation from these procedures, and any other relevant information. — A summary table of relevant measured and calculated data, as hereafter: Substance Uptake and Depuration Rate Constants and Bioconcentration Factors (BCF) k g (growth rate constant; day^{– 1}): Insert Value (95 % CI) [^E0097] k 1 (overall uptake rate constant; l kg^{– 1} day^{– 1}): Insert Value (95 % CI) [^E0097] k 2 (overall depuration rate constant; day^{– 1}): Insert Value (95 % CI) [^E0097] k 2g (growth-corrected depuration rate constant; day^{– 1}): Insert Value (95 % CI) [^E0097] C f (chemical concentration in the fish at steady-state; mg kg^{– 1}): Insert Value ± SD [^E0098] C w (chemical concentration in the water; mg l^{– 1}): Insert Value ± SD [^E0098] L n (lipid normalisation factor): Insert Value (³) BCFSS (steady-state BCF; l kg^{– 1}) Insert Value ± SD [^E0098] BCFSSL (lipid normalised steady-state BCF; l kg^{– 1}): Insert Value (³) ± SD [^E0098] BCFK (kinetic BCF; l kg^{– 1}) Insert Value (95 % CI) [^E0097] BCFKg (growth-corrected kinetic BCF; l kg^{– 1}) Insert Value (95 % CI) [^E0097] t 1/2g (growth-corrected half-life; day): Insert Value (95 % CI) [^E0097] BCFKL (lipid-normalised kinetic BCF; l kg^{– 1}): Insert Value BCFKLG (lipid-normalised growth corrected kinetic BCF; l kg^{– 1}): Insert Value (¹) CI: confidence interval (where possible to estimate) (²) SD: Standard deviation (where possible to estimate) Results reported as ‘not detected/quantified at the limit of detection/quantification’ by pre-test method development and experimental design should be avoided, since such results cannot be used for rate constant calculations.

C.13 - II:   MINIMISED AQUEOUS EXPOSURE FISH TEST

INTRODUCTION

The growing experience that has been gained in conducting and interpreting the full test, both by laboratories and regulatory bodies, shows that — with some exceptions — first order kinetics apply for estimating uptake and depuration rate constants. Thus, uptake and depuration rate constants can be estimated with a minimum of sampling points, and the kinetic BCF derived.

The initial purpose of examining alternative designs for BCF studies was to develop a small test to be used in an intermediate testing step to refute or confirm BCF estimates based on K OW and QSARs and so eliminate the need for a full study for many substances, and to minimise cost and animal use via reduction in sampling and in the number of analytical sequences performed. While following the main design of the previous test method to allow integration of test results with existing BCF data, and to ease performance of testing and data interpretation, the aim was to provide BCF estimates of adequate accuracy and precision for risk assessment decisions. Many of the same considerations apply as in the full test, e.g. validity criteria (cf. paragraph 24) and stopping a test if insignificant uptake is seen at the end of the uptake phase (cf. paragraphs 16 and 38).

Substances that would be eligible for the minimised test design should belong to the general domain that this test method was developed for, i.e. non-polar organic substances (cf. paragraph 49). If there is any indication that the substance to be tested might show a different behaviour (e.g. a clear deviation from first-order kinetics), a full test should be conducted for regulatory purposes.

Typically, the minimised test is not run over a shorter period than the standard BCF test, but comprises less fish sampling (see Appendix 6 for the rationale). However, the depuration period may be shortened for rapidly depurating substances to avoid concentrations in the fish falling below the limit of detection/quantification before the end of the test. A minimised exposure fish test with a single concentration can be used to determine the need for a full test, and if the resulting data used to calculate rate constants and BCF are robust (cf. paragraph 93), the full test may be waived if the resulting BCF is far from regulatory values of concern.

In some cases, it may be advantageous to perform the minimised test design with more than one test concentration as a preliminary test to determine whether BCF estimates for a substance are concentration dependent. If the BCF estimates from the minimised test show concentration dependence, the performance of the full test will be necessary. If, based on such a minimised test, BCF estimates are not concentration dependent but the results are not considered definitive, then any subsequent full test could be performed at a single concentration, thereby reducing animal use in comparison to a two (or more) concentration full test.

Substances potentially eligible for the minimised test should:

• — Be likely to exhibit approximate first order uptake and depuration kinetics, e.g. derived from read-across with similar substances;

• — Have a log K OW < 6 unless rapid metabolism is expected ([^E0099]);

• — Be sufficiently water-soluble with respect to the analytical technique (cf. paragraph 24);

• — Be clearly quantifiable (i.e. concentrations should be at least one order of magnitude above the limit of quantification), both in fish and water, radioactive labelling is recommended (cf. paragraph 23); and

• — Have a depuration period greater than its predicted half-life (cf. Appendix 5 for calculations), or the duration of depuration should be adjusted accordingly (cf. paragraph 91). An exception to this rule is allowed if rapid metabolism of the substance is expected.

SAMPLING SCHEDULE FOR STUDIES FOLLOWING THE MINIMISED DESIGN

Fish sampling

Fish sampling is reduced to four sampling points:

• — At the middle and end of the uptake phase (the latter being the beginning of depuration as well), e.g. after 14 and 28 days (33).

• — At the middle of the depuration phase and at termination of the study (where substance concentration is < 10 % of the maximum concentration, or at least clearly past one half-life of the substance), e.g. after 7 and 14 days of depuration (33). If rapid depuration is expected or observed, it may be necessary to shorten the depuration period to avoid concentrations in the fish falling below the limit of quantification.

• — Lipid measurement as in full study.

• — Growth correction as in full study.

• — The BCF is calculated as a kinetic BCF.

Water sampling

For the minimised design, water is sampled as in full study (cf. paragraph 54) or at least five times equally divided over the uptake phase, and weekly in the depuration phase.

Design modifications

Taking into account the test substance properties, valid QSAR predictions and the specific purpose of the study, some modifications in the design of the study can be considered:

• — If greater precision is needed, more fish (6 or 8 instead of 4) could be used for the sample at the end of the uptake phase.

• — Inclusion of an ‘extra’ group of fish to be used if depuration at 14 days (or the predicted end of the depuration phase) has not been sufficient for adequate depuration (i.e. > 50 %). If the predicted duration of the depuration phase is shorter or longer than 14 days, the sampling schedule should be adapted (i.e. one group of fish at the predicted end of the depuration phase, and one group after half that time).

• — Use of two test concentrations to explore possible concentration dependence. If the results of the minimised test, conducted with two test concentrations, show that the BCF is not concentration dependent (i.e. differ less than 20 %), one test concentration may be considered sufficient in a full test, if it is conducted.

• — It seems likely that models of bioaccumulation processes such as those proposed by Arnot et al. (35) can be used to assist in planning the length of uptake and depuration phases (see also Appendix 5).

Calculations

The rationale for this approach is that the bioconcentration factor in a full test can either be determined as a steady-state bioconcentration factor (BCFSS) by calculating the ratio of the concentration of the test substance in the fish's tissue to the concentration of the test substance in the water, or by calculating the kinetic bioconcentration factor (BCFK) as the ratio of the uptake rate constant k 1 to the depuration rate constant k

Interpretation of the results

To assess the validity and informative value of the test, verify that the depuration period exceeds one half-life. Also, the BCFKm (kinetic BCF derived from a minimised test) should be compared to the minimised BCFSS value (which is the BCFSS calculated at the end of the uptake phase, assuming that steady-state has been reached. This can only be assumed, as the number of sampling points is not sufficient for proving this). If the BCFKm < minimised BCFSS, the minimised BCFSS should be the preferred value. If BCFKm is less than 70 % of the minimised BCFSS, the results are not valid, and a full test should be conducted.

If the minimised test gives a BCFKm in the region of any value of regulatory concern, a full test should be conducted. If the result is far from any regulatory value of concern (well above or below), a full test may not be necessary, or a single concentration full test may be conducted if required by the relevant regulatory framework.

If a full test is found to be necessary after a minimised test at one concentration, this can be conducted at a second concentration. If the results are consistent, a further full test at a different concentration can be waived, as the bioconcentration of the substance is not expected to be concentration dependent. If the minimised test has been conducted at two concentrations, and the results show no concentration dependence, the full test may be conducted with only one concentration (cf. paragraph 87).

Test report

The test report for the minimised test should include all the information demanded for the full test (cf. paragraph 81), except that which is not possible to elaborate (i.e. a curve showing the time to steady-state and the steady-state bioconcentration factor; for the latter the minimised BCFss should be given instead). Additionally, it should also include the reasoning for using the minimised test and the resulting BCFKm.

C.13 - III:   DIETARY EXPOSURE BIOACCUMULATION FISH TEST

INTRODUCTION

The method described in this section should be used for substances where the aqueous exposure methodology is not practicable (for example because stable, measurable water concentrations cannot be maintained, or adequate body burdens cannot be achieved within 60 days of exposure; see previous sections on the aqueous exposure method). It should be realised though that the endpoint from this test will be a dietary biomagnification factor (BMF) rather than a bioconcentration factor (BCF) ([^E0101]).

In May 2001 a new method for the bioaccumulation testing of poorly water soluble organic substances was presented at the SETAC Europe conference held in Madrid (36). This work built on various reported bioaccumulation studies in the literature using a dosing method involving spiked feed (e.g. (37)). Early in 2004 a draft protocol (38), designed to measure the bioaccumulation potential of poorly water soluble organic substances for which the standard water exposure bioconcentration method was not practicable, together with a supporting background document (39), was submitted to an EU PBT working group. Further justification given for the method was that potential environmental exposure to such poorly soluble substances (i.e. log K OW >5) may be largely via the diet (cf. (40) (41) (42) (43) (44)). For this reason, dietary exposure tests are referred to in some published chemicals regulations ([^E0102]). It should be realised however, that in the method described here exposure via the aqueous phase is carefully avoided and thus a BMF value from this test method cannot directly be compared to a BMF value from a field study (in which both water and dietary exposure may be combined).

This section of the present test method is based on this protocol (38) and is a new method that did not appear in the previous version of TM C.13. This alternative test allows the dietary exposure pathway to be directly investigated under controlled laboratory conditions.

Potential investigators should refer to paragraphs 1 to 14 of this test method for information on when the dietary exposure test may be preferred over the aqueous exposure test. Information on the various substance considerations is laid out, and should be considered before a test is conducted.

The use of radiolabelled test substances can be considered with similar considerations as for the aqueous exposure method (cf. paragraphs 6 and 65).

The dietary method can be used to test more than one substance in a single test, so long as certain criteria are fulfilled; these are explored further in paragraph 112. For simplicity the methodology here describes a test using only one test substance.

The dietary test is similar to the aqueous exposure method in many respects with the obvious exception of the exposure route. Hence many aspects of the method described here overlap with the aqueous exposure method described in the previous section. Cross-reference to relevant paragraphs in the previous section has been made as far as possible, but in the interests of readability and understanding a certain amount of duplication is unavoidable.

PRINCIPLE OF THE TEST