Comparison of analytical approaches for liquid chromatography/mass spectrometry determination of the alcohol biomarker ethyl glucuronide in urine

Official guidelines originating from a European Union directive regulate requirements for analytical methods used to identify chemical compounds in biological matrices. This study compared different liquid chromatography/electropray ionization mass spectrometry (LC/ESI‐MS) and tandem mass spectrometry (LC/ESI‐MS/MS) procedures for accurate determination of the conjugated ethanol metabolite and alcohol biomarker ethyl glucuronide (EtG) in urine, and the value of combined EtG and ethyl sulfate (EtS) measurement. Analysis was carried out on 482 urines following solid‐phase extraction (SPE) sample cleanup or using direct injection of a diluted sample. SPE combined with LC/MS/MS was demonstrated to be the most selective and sensitive method and was chosen as reference method. The EtG results by different methods showed good correlation (r = 0.96–0.98). When comparing five reporting limits for EtG in the range 0.10–1.00 mg/L, the overall agreement with the reference method (frequency of true positives plus true negatives) was 82–97% for direct‐injection LC/MS/MS, 90–97% for SPE‐LC/MS, 86–98% for direct‐injection LC/MS, and 86–98% for direct‐injection LC/MS analysis of EtG and EtS. Most deviations were attributable to uncertainty in quantitation, when the value was close to a cutoff but the respective results were slightly above and below, or vice versa, the critical limit. However, for direct‐injection LC/MS/MS, despite earning 4 identification points, equally many negative results were due to a product ion ratio outside the ±20% deviation accepted by the guidelines. These results indicate that the likelihood of different analytical methods to provide reliable analytical results depends on the reporting limit applied. Copyright © 2010 John Wiley & Sons, Ltd.