Ensaio de proficiência para determinação de dioxinas e furanos em gordura de aves

Abstract

Dioxins and furans are highly lipophilic contaminants classified as persistent organic pollutants. Their presence in food is a public health topic due to toxic effects, including carcinogenic properties even at very low concentrations. Recent incidents involving Brazilian chicken meat included chicken fat among the priority matrices for monitoring. However, the control of these compounds depends on the reliability of analytical methods. Proficiency test (PT) are important tools for analytical quality, although rare. The objectives of the present study were the production of a reference material from naturally contaminated chicken fat, the evaluation of homogeneity and stability of the produced material and the organization of a PT for the determination of dioxins and furans. The conditions for the production of the material were optimized in a factorial design, involving potency and time of melting in the microwave, temperature of the thermostatic bath and speed of agitation, aiming to obtain a material of lower viscosity. The agitation speed was a non-significant factor (p> 0.05), while the other factors were significant (p <0.05). The samples were then melted, packed under heating and homogeneization in 80 g flasks, and stored under refrigeration. In the evaluation of homogeneity, ten flasks were analyzed in duplicate by gas chromatography (GC) coupled to high resolution mass spectrometry (HRMS). The batch was considered homogeneous for the ten analytes quantified, according to criteria established by ISO 13528 and IUPAC. In the evaluation of short-term stability, simulating stress conditions in the transport, the materials were exposed to temperatures of 45 °C ± 3 °C for 3, 6 and 9 days. Three samples were taken per treatment, which were analyzed in duplicate. No significant differences between the different treatments (p > 0.05) were observed for all the analytes, indicating stability of the material. The longterm stability was monitored by analyzing three samples, in duplicate, at five different times, for 120 days. This characteristic was demonstrated by regression analysis (p > 0.05) for nine analytes. The criteria recommended in the IUPAC document were more restrictive than those of ISO 13528 in relation to stability. Fourteen laboratories were invited, 12 reported interest in participating and 10 from four different countries effectively participated in the PT. CG-HRMS was the most used technique, followed by CG tandem mass spectrometry (GC-MS/MS) and bioassays, which only reported results for the sum TEQ. Designated values were estimated by the mean of the participants' results, after outliers treatment. The performance statistic was the z-score, and the target standard deviation was estimated by the Horwitz model and by the maximum deviation allowed by the European Legislation. For five of the eight analytes evaluated, all laboratories presented satisfactory performance in both target standard deviation criteria. For the congeners 1,2,3,7,8-PeCDF, 1,2,3,6,7,8-HxCDD, 1,2,3,6,7,8-HxCDF were found one, two and two laboratories with unsatisfactory results, respectively. For the TEQ summation, two of the four unsatisfactory results corresponded to laboratories that used bioassay. In this way, adequate performance of the participants was demonstrated for the determination of dioxins and furans congeners of greater toxicological relevance.