Occurrence of heterocyclic aromatic amines in the Swiss diet: analytical method, exposure estimation and risk assessment.

A total of 86 meat samples, prepared in restaurants or homes, ready to eat (including poultry and fish) and 16 commercial samples such as bouillon (cubes) were analysed for heterocyclic aromatic amines (HAA). The analytical method consisted of an acidic extraction, clean-up on a cation exchange cartridge followed by an analogous HPLC step to recover the following HAA: IQ, MeIQ, MeIQx, 4,8-DiMeIQx, PhIP and 7,8-DiMeIQx. The HAA containing HPLC-fractions were collected, the HAA identified and quantified using two RP-HPLC-systems of different retention properties (UV-detection). The limit of quantitation was in the range of 0.2-0.4 ng/g and the relative repeatability 6-15%. The recovery of PhIP was lower than for the other HAA analysed (less than 80%) and a correction factor was applied. No significant differences of the HAA-concentration were found in samples from homes and restaurants, half of the total samples contained HAA at the following frequencies: PhIP and MeIQx 33% (each), 4,8-DiMeIQx 11% and MeIQ 4%; 7,8-DiMeIQx and IQ were not detected. The frequencies in commercial products were for MeIQx 31%, 7,8-DiMeIQx 19%, IQ 13% and PhIP 6%; MeIQ and 4,8-DiMeIQx were not found. Based on these data, the average exposure of Swiss adults to HAA was estimated to be 5 ng/kg body mass per day, commercial products contributing less than 10%. The theoretical excess cancer risk due to this intake was estimated on the base of the carcinogenic potency of the HAA in long-term animal experiments by linear extrapolation. The resulting risk in the order of 10(-4) at the maximum is discussed in terms of Swiss epidemiological data.