Use of statistical approaches to evaluate a pesticide multiresidue method for grapefruit samples.

PAREJA, LUCIA; REPETTI, MARÍA ROSA; COLAZZO, MARCOS; BESIL, NATALIA; GARCÍA, SILVIA; MICHLIG, NICOLÁS; DEMONTE, LUISINA; CESIO, VERÓNICA; HEINZEN, HORACIO; BELDOMÉNICO, HORACIO

Dublin

Workshop; 10th European Pesticide Residue Workshop; 2014

Citric fruit production chains are relevant in Argentina and Uruguay. Concerned on export markets they introduced advanced technology at crop fields and industry levels. Pesticide residues are a primary food safety topic worldwide. The analytical issue to determine pesticide residues in such complex matrices has been always a challenge, being grapefruit studies one of the less reported in literature. Grapefruit is complex to analyze as many co-extractives (polyphenols - methylated flavones and flavanones-, antranillic acid, coumarins, limonoids, etc.), can degrade pesticides or hamper its extractability, interfering the identification and quantification steps.In previous work, an experimental design (DOE) was employed to study how the main variables (factors) of the QhEChERS citrate method (QC) influenced the recovery rates of pesticides from grapefruit matrix, aiming to find the optimized operating conditions. A fractional factorial design of three factors studied at three levels resulting in 9 different experiments were adopted. The factors were: sample amount (fA), shaking time (fB) and C-18 amount (fC), on a grapefruit matrix spiked at 100 μg/kg with 13 different fungicides and insecticides intensively used in citrus production. QuEChERS extracts were then analyzed by UHPLC-MS/MS. Recovery percentage was selected as the response to be studied. Results were evaluated using ANOVA and Response Surface Methodology (RSM). Pareto charts and graphics of the surfaces applying the regression equation of the fitted model were performed. Software Statistica 7.0® and Statgraphics® XV were used for statistical analysis. The optimizing analysis using the fitted model gave the combined optimal values of fA=10 g, fB= 4 min and fC=200 mg, to yield appropriate optimal recovery levels. In this communication, further evaluation and validation of the procedure using these optimized values is presented. Satisfactory performance was obtained, complying with SANCO/12571/2013 guidelines for all compounds, including those that in the experimental assay showed worst performance (azinphos methyl, metalaxil and propiconazole). Linearity and specificity conditions were satisfactory in all cases. Matrix effects assay showed a variable influence of the matrix on ESI behavior of target ions, significant ion suppression (>30%) was verified for carbendazim, imidacloprid, azinphos-methyl and a significant signal enhancement for methamidofos. Trueness and precision was demonstrated through recovery study with values within 70-120% and RSD