Urine is to date the most classically used matrix for the assessment of human exposure to pesticides. Although this matrix allows for the detection of several compounds, mainly metabolites, the kinetics of urinary compounds makes it representative of recent exposure only for most chemicals. Thanks to progress in analytical sciences which led to significant improvement in methods sensitivity, growing interest is currently observed in hair analysis for the detection of organic pollutants. Hair enables to detect both metabolites and parent compounds, with windows of detection generally corresponding to several months, depending on hair sample length. Due to the different nature of urine and hair, which are liquid and solid respectively, and also partially to the prejudice on the compounds detectable therein, no comparison of the respective possibilities of both matrices with regard to the biomonitoring of human exposure to organic pollutants has been performed so far. In order to fill this gap, the purpose of the present work was therefore to develop multi-residue methods, as similar as possible, for the analysis of the same compounds in hair and in urine.

The list of pollutants investigated here consisted of 58 compounds, including organochlorines, organophosphates, pyrethroids, carbamates, other pesticides and PCBs. The list included both parent compounds and metabolites. Two different approaches were necessary for the analysis of non-polar compounds (mainly parent) and polar (mainly metabolites). In the final procedure, extraction from hair (50mg) was carried out with acetonitrile/ water after sample decontamination and pulverization. Extract was split into two fractions, which were analyzed directly with Solid phase microextraction (SPME) injection for non-polar compounds and after derivatization in liquid injection for polar compounds. In urine, non-polar compounds were analyzed directly using SPME. Polar compounds were analyzed after acidic hydrolysis, liquid-liquid extraction with acetonitrile-cyclohexane, derivatization and liquid injection. The volume of urine used was 0.5mL. Analysis was carried out with gas chromatography tandem mass spectrometry (GC-MS/MS) for all the compounds (non-polar and polar) in the two matrices.

In hair, limits of quantification (LOQ) ranged from 0.02pg/mg for trifluralin to 5.5pg/mg for diethyl phosphate (an organophosphate metabolite). In urine, LOQ ranged from 0.4pg/mL for a-endosulfan to 4ng/mL for dimethyldithiophosphate. The analysis of samples supplemented with standards confirmed that all the compounds were analyzable in both hair and urine. The levels of sensitivity reached with these methods were quite satisfactory with regard to previously published studies, and also considering the number of compounds investigated.

The approach presented here is the first one allowing for the analysis of a common list of compounds in two different biological matrices. The limits of quantification were compatible with the ranges of concentration generally reported in these matrices for the general population. The application of the latter methods on urine and hair samples collected from the same individuals will provide information on the respective relevance of these matrices in the assessment of human exposure to organic pollutants.