A process used to prepare the test portion of ground wheat from the whole grain laboratory sample for ochratoxin A (OTA) analysis using dry comminution with homogenisation and sub-sampling via a rotary sample divider was developed and evaluated. With respect to OTA content, the developed process produced a homogeneous sample of ground wheat from 10 kg of whole grain. Relative standard deviations of the mean OTA concentration for five naturally contaminated wheat samples processed using the developed method ranged from 9% to 19% over a relevant concentration range of 1.7 to 7.6 mg/kg. Additional studies demonstrated that OTA was stable in ground wheat with moisture content between 12 to 13% for at least a year when stored at ambient temperatures. Further examination of the developed comminution and dividing procedure demonstrated that higher concentrations were measured in smaller sized particles, indicating that the accuracy and precision of OTA analyses could be affected by the particle size of ground wheat.

Randomly selected domestic and export shipments (n = 1907) of Canadian durum and other wheat that occurred between 1 January 2010 and 31 December 2012 were analysed for ochratoxin A (OTA). The majority of samples did not contain OTA above the LOQ of 1 μg kg⁻¹. Only 37% of samples analysed contained quantifiable OTA; the median OTA of the positive results was 2.10 μg kg⁻¹. Canada Western Amber Durum shipments contained OTA more frequently, and at slightly higher concentrations, than Canada Western Red Spring wheat. For both wheat classes the frequency of OTA occurrence and mean concentrations appeared to increase in the lower grades, but these increases were not statistically significant. A periodic trend of a late summer increase of mean monthly OTA concentrations in shipments appears tied to the cycle of producer deliveries of wheat to primary grain elevators.

The accuracy and precision of a commercially available system based on an indirect competitive immunoassay and planar waveguide technology was evaluated for the analysis of deoxynivalenol (DON), ochratoxin A (OTA), zearalenone (ZEAR), and T-2 toxin in wheat. The system generally performed well at the tested concentrations that were close to the regulatory limits of DON and OTA in wheat. The mean percent recovery of OTA from certified and in-house reference materials ranged from 90 to 111 %, with a relative standard deviation of 8-16 % (at 4.2, 4.9, and 7.0 μg/kg). Mean percent recoveries of DON ranged from 75 to 103 %, with a relative standard deviation of 14-20 % (at 610, 940, and 1300 μg/kg). As analyte concentrations approached the lower limits of the working range of 3 μg/kg OTA and 400 μg/kg DON, the mean percent recoveries and relative standard deviation increased for both DON and OTA. A lack of reference materials precluded a thorough evaluation of the method for the analysis of ZEAR and T-2. The particular strength of the technology was that multiple mycotoxins were analyzed simultaneously.