Biological test method for determining toxicity of sediment using luminescent bacteria: chapter 7


Section 5: Procedure for Testing a Reference Toxicant

The routine use of a reference toxicant is necessary to assess, under standardized test conditions, the relative sensitivity of the bacteria after reconstitution, the accuracy of dilution techniques, and other factors affecting the precision and reliability of data produced by laboratory personnel using the Solid Phase Reference Method (EC, 1992). Reference toxicant(s) must be tested at least once a month during periods when solid-phase tests for sediment toxicity using V. fischeri are being run, and initially upon first use of a new shipment or batch of Bacterial Reagent.

One or more samples of positive control sediment (see Section 4.4) must be used as the reference toxicant(s) for this reference method. When performing each reference toxicity test, the test procedures and conditions defined in Sections 4.5 and 4.6 must be followed.

For performing a reference toxicity test according to the test procedures and conditions defined herein, use of a standard contaminated sediment such as a NRCC Harbour Marine Reference Material (e.g., HS-5 or HS-6), which can be purchased from the National Research Council of Canada, is recommended.Footnote 12 Alternatively, a spiked control sediment may be used as a reference toxicant, after suitable spiking procedures are standardized. In this regard, Environment Canada’s guidance document on using control sediment spiked with a reference toxicant should be consulted (EC, 1995b).

When undertaking a reference toxicity test, a series of concentrations should be chosen which, based on preliminary and/or previous tests performed using the same conditions and procedures, will enable calculation of an IC50 for light emission by V. fischeri (see Section 6.1), with acceptably narrow 95% confidence limits. The selected test concentrations should bracket the predicted IC50.

It is the responsibility of laboratory personnel to demonstrate their ability to obtain consistent, precise results with the reference toxicant before definitive sediment assays are conducted using this reference method. To meet this responsibility, the laboratory personnel should initially determine their intralaboratory precision, expressed as percent coefficient of variation (% CV), by performing five or more reference toxicity tests with different lots of Bacterial Reagent (Section 2.1), using the same reference toxicant and the procedures and conditions defined herein. This should be conducted to gain experience with the test procedure, and as a point of reference for future tests (EC, 1998). While routinely performing this reference toxicity test with each lot of Bacterial Reagent, laboratory personnel should continue to follow this same procedure. Once sufficient data are available (EC, 1995b), IC50s derived from these tests must be plotted successively on a reference toxicant-specific warning chart, and examined to determine whether the results are within ±2 SD of values obtained in previous tests using the same reference toxicant and test procedure. A separate warning chart must be prepared and updated for each reference toxicant used as part of this reference method. The warning chart should plot logarithm of concentration on the vertical axis against date of the test or test number on the horizontal axis. Each new IC50 for the reference toxicant must be compared with the established limits of the chart; the IC50 is acceptable if it falls within the warning limits. All calculations of mean and standard deviation should be made based on log(IC50).

The logarithm of concentration (including IC50) should be used in all calculations of mean and standard deviation, and in all plotting procedures. This simply represents continued adherence to the assumption by which each IC50 was estimated based on logarithms of concentrations. The warning chart may be constructed by plotting the logarithmic values of the mean and ±2 SD on arithmetic paper, or by converting them to arithmetic values and plotting those on the logarithmic scale of semi-log paper. If it were demonstrated that the IC50s failed to fit a log-normal distribution, an arithmetic mean and SD might prove more suitable. The mean of the available values of log(IC50), together with the upper and lower warning limits (±2 SD), should be recalculated with each successive IC50 until the statistics stabilize (EC, 1995b; 2002b).

If a particular IC50 fell outside the warning limits, the sensitivity of the test organisms and the performance and precision of the test would be suspect. Since this might occur 5% of the time due to chance alone, an outlying IC50 would not necessarily indicate abnormal sensitivity of the lot of Bacterial Reagent or unsatisfactory precision of toxicity data. Rather, it would provide a warning that there might be a problem. A thorough check of all test conditions and procedures should be carried out. Depending on the findings, it might be necessary to repeat the reference toxicity test, or to obtain a new lot of Bacterial Reagent for evaluating the toxicity of the samples of test material (together with a new reference toxicity test using the new lot of test organisms).

Results that remained within the warning limits might not necessarily indicate that a laboratory was generating consistent results. Extremely variable data for a reference toxicant would produce wide warning limits; a new data point could be within the warning limits but still represent undesirable variation in test results. A coefficient of variation of no more than 30%, and preferably 20% or less, is suggested as a reasonable limit by Environment Canada (1995b). For this reference method, the coefficient of variation for mean historic data derived for the reference toxicity tests performed at a testing facility should not exceed 30%.

Page details

Date modified: