Ochratoxin A in raisins and currants: basic extraction procedure used in two small marketing surveys of the occurrence and control of the heterogeneity of the toxins in samples.

A basic extraction procedure for analysis of ochratoxin A (OTA) in currants and raisins is described, as well as the occurrence of OTA and a control of heterogeneity of the toxin in samples bought for two small marketing surveys 1999/2000 and 2001/02. Most samples in the surveys were divided into two subsamples that were individually prepared as slurries and analysed separately. The limit of quantification for the method was estimated as 0.1 microg kg(-1) and recoveries of 85, 90 and 115% were achieved in recovery experiments at 10, 5 and 0.1 microg kg(-1), respectively. Of all 118 subsamples analysed in the surveys, 96 (84%) contained ochratoxin A at levels above the quantification level and five samples (4%) contained more than the European Community legislation of 10 microg kg(-1). The OTA concentrations found in the first survey were in the range < 0.1-19.0 microg kg(-1) with a median concentration of 0.9 microg kg(-1). In the 2001/02 study, the range was < 0.1-34.6 microg kg(-1) with a median of 0.2 microg kg(-1). Big differences were often achieved between individual subsamples of the original sample, which indicate a wide heterogeneous distribution of the toxin. Data from the repeatability test as well as recovery experiments from the same slurries showed that preparation of slurries as described here seemed to give a homogeneous and representative sample. The extraction with the basic sodium bicarbonate-methanol mixture used in the surveys gave similar or somewhat higher OTA values on some samples tested in a comparison with a weak phosphoric acid water-methanol extraction mixture.