Matrix Effect in the Quantitative Determination of Mandelic and Phenylglyoxylic Acid in Urine Samples by HPLC-MS/MS with Isotopic Dilution

Styrene is a commercially important chemical widely used, whose metabolic pathway in humans leads to mandelic (MA) and phenylglyoxylic (PGA) acids, whose total concentration is the dose biomarker suggested by ACGIH for occupational exposure. Non occupational exposure to styrene is uncommon, but nevertheless a background value is always found in human urine. Most analytical methods applied to workers biomonitoring use HPLC-UV, a non specific analytical technique. The reference value reported for non styrene exposed Italian population was determined by HPLCMS/ MS but without considering matrix effect, an alteration of ionization efficiency due to the presence of undetected coeluting substances that can affect the reliability of results. This work presents the validation of quantitative determination by HPLC-MS/MS for MA and PGA using the isotopic dilution method that makes possible the determination of urinary concentration compensating for the matrix effect, and the comparison of this results with those obtained with a traditional HPLC/UV method. LOD is 0.02 mg/l for MA and 0.015 mg/l for PGA, and LOQ 0.075 and 0.040 mg/l respectively. Accuracy is always higher than 82% and variability lower than 11% for both analytes, while a very strong matrix effect makes the use of the internal standard crucial, in order to achieve reliable results. Using this method, we showed that HPLC/UV method overestimates the biomarkers levels because of its poor specificity, while MS/MS detection presents the risk of underestimation because of the urinary matrix, if this is not correctly compensated. Keywords: Accuracy, biomonitoring, mandelic acid, matrix effect, occupational exposure, phenylglyoxylic acid, specificity, styrene