Simultaneous Quantitative Determination of Ethyl Glucuronide and Ethyl Sulphate in Human Urine using UPLC-MS/MS

This application note demonstrates to develop and validate a simple and rapid UPLC-MS/MS method for the simultaneous quantitative determination of ethyl glucuronide (EtG) and ethyl sulphate (EtS) in human urine.

• EtG and EtS are non-volatile, water soluble, direct metabolites of ethanol.
• Both EtG and EtS are formed shortly after ethanol consumption and can be detected much longer than ethanol itself (up to 80 hours for EtG, 30 hours for EtS)^1,2.
• EtG has been shown to be susceptible to post sampling synthesis³ and also bacterial degradation^4,5 which may lead to false positive or negative results respectively. These effects were not seen with EtS, making EtS a more reliable marker.
• EtG and EtS are formed by different metabolic pathways and therefore simultaneous determination has been found to increase sensitivity and reliability in detecting recent ethanol consumption⁶.
• EtG testing is now widely used in alcohol withdrawal programmes (to monitor abstinence), within workplace settings and for forensic cases such as post-mortems and drug-facilitated crimes.

Materials

Specimens

Validation was performed using human urine samples obtained from the Analytical Unit, St George’s - University of London (London, UK) and Wythenshawe Hospital (Manchester, UK). All samples (Sodium fluoride preserved) were stored at -20 °C until analysis. Synthetic blank urine (Surine, DYNA-TEK industries, USA) was used as the control material to prepare all the calibrators.

Internal standards

Deuterated analogues EtG-D5 and EtS-D5 (Lipomed, Switzerland) were used as the internal standards (IS). A mixed stock solution was prepared in water at 20 and 5mg/L respectively.

Sample Preparation

A simple urine dilution (1:20) was undertaken after centrifugation at 12000rpm (~11000xg) for 10 minutes.

This dilution also incorporated the addition of the IS. Briefly, IS (10μL) and 0.1% formic acid (940μL) were added to the human urine samples (50μL) before finally vortex mixing for 30 seconds.

The MRM conditions used for the measurement of EtG, EtS and their respective internal standards are summarised in Table 1. A calibration curve (0.25–100mg/L for EtG, 0.05-20mg/L for EtS) was prepared by adding EtG and EtS to synthetic blank urine. Calibrators and quality controls (QC) were diluted by the same procedure as previously described for the samples.

Figure 2 shows the MRM chromatograms obtained from a 10μL injection of a 0.5mg/L urine calibrator. The quantifier/qualifier ion ratios for both compounds were monitored for all calibrators, QC’s and samples and were found to be within ±20% of the target ion ratios.

Quantitation was performed by the integration of the area under the peak of the specific MRM chromatogram. Figure 3 shows a typical standard curve for EtG and EtS in urine. Calibrators were plotted using 1/x weighting and found to be linear for both compounds, over the investigated range (coefficient of determination r² = >0.996).

Limits of detection were 0.2 and 0.04mg/L for EtG and EtS respectively, which is below the cut-offs applied for this analysis i.e, 0.5 and 0.1mg/L respectively. 

Intra-assay precision and accuracy were assessed by adding the EtG and EtS to blank patient urine (n=5) at four QC concentrations (0.75, 2.5, 7.5 and 50mg/L for EtG and 0.15, 0.5, 1.5 and 10mg/L for EtS). Inter-day precision was assessed by analysing the QC samples in duplicate on five different days. Intra and interassay precision and accuracy was found to be good, with precision CV’s <10% and accuracy between 97-112%, as shown in Table 2.

The stability of prepared samples and standards was assessed over 24 hours. A prepared calibrator (2500/500mg/L, EtG/EtS) was stored at 5 °C in the dark in the ACQUITY Sample Manager with an injection performed every hour. No significant changes in absolute peak area were found for either compound over the investigated time period.

Matrix effects were assessed in 2 ways, firstly by spiking blank prepared patient samples (n = 6) with both compounds and comparing the absolute peak areas against the equivalent concentration of standard solution in solvent. The average matrix effects were found to be acceptable (-16% for EtG and -7% for EtS). Secondly, a post-column infusion of both compounds was performed during the injection of a solvent blank and prepared urine. Minimal matrix effects were observed with a simple urine dilution. An example shown in Figure 4.

Forensic samples (n = 39) collected from alleged drug- facilitated sexual assault (DFSA) cases which had been previously analysed for EtG using the Microgenics DRI EtG Enzyme Immunoassay⁷, were subsequently analysed using the newly developed UPLC-MS/MS method. There is currently no immunoassay test available for EtS. EtG and EtS levels are highly influenced by water intake⁸ therefore normalisation of EtG and EtS values to the creatinine concentration is recommended, but for the purpose of this comparison results were not normalised. Preliminary results showed that many samples contained EtG and EtS concentrations which were above the calibration range used. Therefore, all samples were re-prepared by dilution (1:100) with synthetic blank urine, as previously described and re-analysed. Table 3 shows the EtG and EtS results from the forensic case samples. The EtG results showed a good correlation (r²=0.978) but also showed an analytical bias, as shown in Figure 5. The bias will be investigated in future work by the analysis of an independent reference material. EtG and EtS were detectable in samples collected up to approximately 40 hours after the alleged DFSA.

EtG and EtS testing is becoming more widely used across the world within different settings such as alcohol withdrawal programs, clinical situations, forensic cases and the workplace to identify recent ethanol consumption or to verify abstinence.

The developed methodology has been shown to be accurate, precise and sensitive for the simultaneous quantitation of EtG and EtS and can provide rapid results in a single 4 minute chromatographic run.

The method has been successfully applied to the analysis of EtG in forensic samples with good correlation when compared to an established immunoassay. There is currently no immunoassay test available for EtS.

The speed and simplicity of the developed method make it the ideal solution for reliable, rapid, high-throughput EtG and EtS analysis.

1. Wurst FM, Skipper GE, Weinmann W. Addition. (2003) 98(S2):51–61
2. Dresen S, Weinmann W, Wurst FM. J. Am. Soc. Mass Spectrom. 2004;15:1644–8
3. Helander A, Olsson I, Dahl H. Clin Chem 2007:53:1855–7
4. Baranowski S, Serr A, Thierauf A, Weinmann W, GroβePerdekamp M, Wurst FM, Halter CC. Int J Legal Med. 2008 Jun 24. [Epub ahead of print]
5. Helander A, Dahl H. Clin Chem 2005:51;1728–30
6. Wurst FM, Dresen S, Allen JP, Wiesbeck G, Graf M, Weinmann W. Addiction 2006:101(2):204–11
7. Button J, Mathers C, Elms R, Dhillon J, Holt D. Ann Toxicol Anal. 2008;20(S1):68
8. Goll M. J Anal Toxicol 2002;26:262–5

Acknowledgements

Brian Keevil, Wythenshawe Hospital (Manchester, UK) for supplying blank patient urine samples for the validation process.