Interlaboratory comparison of two AOAC liquid chromatographic fluorescence detection methods for paralytic shellfish toxin analysis through characterization of an oyster reference material.

ABSTRACT

An interlaboratory ring trial was designed and conducted by the Centre for Environment, Fisheries, and Aquaculture Science to investigate a range of issues affecting the analysis of a candidate Pacific oyster paralytic shellfish toxin reference material. A total of 21 laboratories participated in the study and supplied results using one or more of three instrumental methods, specifically precolumn oxidation (Pre-COX) LC with fluorescence detection (FLD; AOAC Official Method 2005.06), postcolumn oxidation (PCOX) LC-FLD (AOAC Official Method 2011.02), and hydrophilic interaction LC/MS/MS. Each participant analyzed nine replicate samples of the oyster tissue in three separate batches of three samples over a period of time longer than 1 week. Results were reported in a standardized format, reporting both individual toxin concentrations and total sample toxicity. Data were assessed to determine the equivalency of the two AOAC LC methods and the LC/MS/MS method as well as an assessment of intrabatch and interbatch repeatability and interlaboratory reproducibility of each method. Differences among the results reported using the three methods were shown to be statistically significant, although visual comparisons showed an overlap between results generated by the majority of tests, the exception being the Pre-COX quantitation of N-hydroxylated toxins in post ion-exchange fractions. Intralaboratory repeatability and interlaboratory reproducibility were acceptable for most of the results, with the exception of results generated from fractions. The results provided good evidence for the acceptable performance of the PCOX method for the quantitation of C toxins. Overall the study showed the usefulness of interlaboratory analysis for the characterization of paralytic shellfish poisoning matrix reference materials, highlighting some issues that may need to be addressed with further method assessment at individual participant laboratories.