Determination of eight volatile nitrosamines in thirty cured meat products with capillary gas chromatography-high-resolution mass spectrometry: the presence of nitrosodiethylamine and the absence of nitrosopyrrolidine.
Five samples each of six cured meat products were analysed for the presence of eight volatile nitrosamines. A combined steam distillation-extraction step at pH 8-9 plus a concentration step were used for preparation of extracts. These were analysed, without further clean-up, using a coupled capillary GC-HRMS system at a resolution of 4 000. The MS was used in the fixed mass mode, but after the retention time of each nitrosamine the mass adjustment was tuned to the mass of the next nitrosamine. This combination allowed not only an unambiguous identification of the nitrosamines but also their direct quantitative determination. The products investigated were cooked ham, fermented sausage, raw bacon, mildly fried bacon, fried minced mean and smoked meat. Only one nitrosamine, NDEA, was sometimes found to be present in relatively large amounts. It occurred in two of the bacon samples; the concentration increased during mild frying from 4 and 13 mug/kg to 16 and 43 mug/kg, respectively. It was also present in two of the smoked meat samples in amounts of 7 and 91 mug/kg, respectively. Its identity was further confirmed on a second and third capillary column also coupled to the MS, and by direct introduction of the extracts into the MS at a resolution of 15 000. A silicon compound with practically the same retention time and mass as NDMA, which originates from the antifoam agents used as shown by other investigators, was found in the extracts in amounts corresponding to abuot 2 mug/kg of product. Nevertheless, seven samples were found positively to contain NDMA in amounts of 2-6 mug/kg. NMEA, NMPA, NMBA and NDPA could not be detected in any of the samples, nor could NPip or NPy. That no NPy (i.e., less than 50 mug/kg) was found in the fried bacon samples may be due to the mild frying method used. The use only of capillary GC with a nitrogen-selective detector, without further sample clean up, would have led to many false-positive results due to the presence of other nitrogen-containing compounds.