Multimycotoxin Analysis by LC-MS/MS in Cereal Food and Feed: Comparison of Different Approaches for Extraction, Purification, and Calibration.

Twelve different approaches commonly used for the simultaneous LC tandem MS (MS/MS) determination of mycotoxins (deoxynivalenol, aflatoxins, ochratoxin A, T-2 and HT-2 toxins, fumonisins, and zearalenone) were tested in cereals and feed materials. They comprised different extraction solvents, types of cleanup [solid-phase extraction, QuEChERS, and immunoaffinity (IMA)], and calibration approaches (external or matrix-matched). The percentage of mycotoxins with acceptable recovery, according to Regulation (EC) No. 401/2006, ranged from 9 to 100%. The approach giving the highest percentage of acceptable results was selected and further tested for corn, rice, and feed spiked at three different mycotoxin levels (low, medium, and high). The method is based on extraction with MeOH-water (70 + 30, v/v) and cleanup with two multiantibody IMA columns. For corn and rice spiked at low mycotoxin levels, a significant matrix effect was observed and was compensated by using 13C calibration. At higher mycotoxin levels (medium and high), matrix effects were negligible as no significant differences were observed for the majority of recovery results calculated by 13C calibration and external calibration. Although the proposed method still needs improvement in terms of accuracy and, to a lesser extent, precision, it was successfully tested with four proficiency tests in buckwheat, corn, rice, and feed, giving acceptable z-scores for 97% (34 out of 35) of results.