Important Facts About EtG (ethylglucuronide) and EtS (ethylsulfate) Testing
Both EtG and EtS are minor but important metabolites of alcohol (ethanol) (from a monitoring perspective), because they are more slowly eliminated from the body than alcohol itself (or than other markers of recent alcohol use) and can thus be used to better document abstinence and detect alcohol relapse earlier. In other words, urine EtG and EtS are the best markers available currently to monitor and document alcohol abstinence. Both EtG and EtS have similar time spectra. One standard drink, in an average person, can be detectable for about 24 hours (using the lowest cutoffs of 100ng/ml for EtG and 25ng/ml for EtS), whereas, urine alcohol will disappear following one drink within about 4 – 6 hours. After binge drinking EtG and/or EtS can be detected in urine for several days or more.
The detected level of EtG or EtS in urine is not meaningful to estimate amount of drinking because the level varies greatly depending on numerous factors including: the amount of alcohol intake, the duration of time over which the alcohol was consumed, the duration since the last drink, the individuals metabolism (genetic activity of their liver enzyme systems - influenced by medications that increase enzyme activity, pregnancy, etc), the amount of water in the urine (roughly proportional to amount of creatinine in urine), and other factors (i.e. body size, muscle mass). Because of the large number of variables it is impossible to calculate the amount of alcohol consumed from the urine levels of EtG or EtS. In other words, Urine EtG and EtS cannot and should not be used to determine the amount of drinking. A positive EtG and/or EtS is a qualitative result (as opposed to quantitative). It means alcohol was present. Future studies may help understand any relationship between urine levels and approximate amount of drinking but these studies have not been performed.
The enzyme systems that metabolize ethanol to EtG or EtS are very different. Some individuals have more or less amounts of these enzymes and thus make more or less of these markers. One reason to test for both EtG and EtS is to increase sensitivity (and specificity) of detecting recent alcohol use, by reducing the chance that the person being testing simply doesn’t make much of one or the other of these metabolites. Thus, if someone is negative for EtG after drinking they might well be positive for EtS and vice versa. Using both tests makes for more sensitive detection and more absolute proof of abstinence.
Overall, EtS is probably a superior marker compared to EtG. EtG came along first and is therefore more commonly performed. EtS is probably better because of findings that EtG is degraded in urine when certain bacteria are present (which can cause a false negative test) whereas EtS has not been shown to be degraded by bacteria. EtG has also been recently reported to be created in urine if alcohol is present in the urine along with similar bacteria. Thus, EtG can be created and/or destroyed in urine by bacteria but EtS has not been shown to be vulnerable in these ways. Therefore, EtS is possibly both more sensitive and specific and that is why it may be a better test.
Figure 1:
Chart: Comparing EtG to EtS
Analyte EtG EtS
Average duration of positive following one drink 20.6 h 21.2 h
Lab cutoffs offered 100 to 500ng/ml 25 to 100ng/ml
Stability: Degraded in presence of bacteria Yes No
In vitro synthesis by bacteria Yes No
LC/MS/MS Yes Yes
Number of labs performing test (4/2008) 20 2-3*
* Coventry Diagnostics (Michigan), Forensic Lab (Denver) and USDTL (Chicago)
As an example of the superior sensitivity of EtS in one study both tests were performed on 98 urine samples. 27 were positive for EtG and 20 positive for EtS suggesting a higher sensitivity for EtS. The authors concluded by recommending that both tests be performed when possible to improve sensitivity.(Wurst, et al, 2006) Further evidence regarding sensitivity and specificity came from two studies from Sweden, which demonstrated that EtG, but not EtS, can be destroyed in urine when certain bacteria are present (Helander, et al, 2005) and later the same authors found that EtG, but not EtS, is not only destroyed but can also be created in urine when alcohol and certain bacteria are present (Healander, et al, 2007). “Urine specimens with confirmed growth of Escherichia coli, Klebsiella pneumoniae, or Enterobacter cloacae were stored at room temperature in the presence of ethanol (either added to the samples or generated by inoculation with the fermenting yeast and glucose as substrate) and EtG but not EtS was later found. They concluded that the presence of EtG in urine is not therefore a unique indicator of recent drinking, but can originate from postcollection synthesis by bacteria. Given the associated risks for false identification of alcohol consumption and the risk of false-negative EtG results due to bacterial degradation, they recommended that EtG testing always be combined with EtS testing, or if only one test is feasible, EtS is superior.
Incidental Exposure to Ethanol
Incidental exposure to ethanol can occur from many sources, since ethanol is common in our environment as it is used in cooking, is found in fermented products like wine vinegar or soy sauce, as a solvent in over-the-counter medications, in hygiene products, antibacterial gels, perfumes, bug spray, in gasohol (ethanol added to gasoline), communion wine, etc.. When fumes of ethanol are breathed, positive EtG tests can occur with levels as high as 700ng/ml in some cases. When testing individuals in monitoring using EtG or EtS it is important to warn them to avoid products containing ethanol (alcohol) and to give them a list of the products to avoid. Regarding fumes of alcohol from handgel or perfumes, recommending that they use simple effort to avoid excessive breathing of these products and/or that they use them in well ventilated areas is usually adequate to avoid positive tests. They should be warned to avoid foods, mouthwash, otc meds, and fumes of alcohol. If they must use alcohol gels for hand cleaning, they should avoid the fumes by holding their hands away from the face, and using the gel in a well-ventilated area, if possible. Despite all warnings, some individuals still have relatively low positives (less than 1,500ng/ml) from incidental exposure. Because more than one source of incidental exposure can occur, and because some individuals make much more EtG/EtS than others, and all the factors mentioned above can come into play, it is impossible at this time to know what the upper limit is for cutoff to be certain alcoholic beverages were consumed. The best approach is to confront the individual who’s had a positive test, no matter the level, with supportive encouragement to be honest and accept help early. More than half, in one series, admitted drinking. The likelihood of admitting drinking is probably related to the perceived consequences. If the individual believes they will not be harshly punished for admitting drinking they are much more likely to admit it.
What cutoff should be used for EtG and/or EtS? The answer is, it depends on at least two issues: 1. how much time and energy the program has to investigate lower level positives that could be from drinking but may be from incidental exposure and 2. the consequences and rigidity of programs to levy those consequences for lower values that could be from incidental exposure. Thus, if a program has flexibility to evaluate low positives and deal with more “gray zone” cases, in which it is difficult to know what caused the positive, but wishes to detect relapses as early as possible, then a low cutoff is preferable. It is not uncommon for a low positive EtG (such as, 110ng/ml) or EtS (such as, 30ng/ml) to elicit a confession of drinking and allow earlier intervention. Some labs have reported that up to ½ of their positive EtG tests are between 100-500ng/ml. Setting a higher cutoff may cause less stress and confusion but fewer drinking episodes will be detected. Each program must weigh their own policies, flexibility, and desire for early detection to determine what cutoff they wish to use. Cutoffs for EtG commonly available include: 100, 250, 500, and 1,000ng/ml.
The clinical situation with the participant should always be considered. If there have been signs or symptoms of drinking then a low positive EtG is probably more likely to truly represent proof of drinking. Nevertheless, if an individual firmly denies drinking, and it is important to “prove” whether or not the positive test was actually from drinking, programs can refer to specialized evaluation centers or individual practitioners to conduct a more thorough “addiction assessment” in which a more detailed history, interview with collateral sources of information, polygraph, testing possible sources of incidental exposure in a protected environment, can all be performed to seek to get a more definitive answer. In less critical situations, it’s often just is useful, and therapeutically more positive, to simply heighten vigilance and testing and if further drinking occurs it will likely become clear. Rarely is there serious risk to patients if vigilance is present.
Other ways to monitor individuals for alcohol use include: devices (such as the SCRAM device that is worn on the ankle) that measures transdermal alcohol, administering observed Antabuse (which would then cause an anatabuse reaction if there is drinking), etc.. However, because urine testing is so common, EtG and EtS have been valuable additions to our toxicology armamentarium.
References
Helander A, Dahl H. Urinary tract infection: a risk factor for false-negative urinary ethyl glucuronide but not ethyl sulfate in the detection of recent alcohol consumption. Clin Chem. 2005 Sep;51(9):1728-30
Helander A, Olsson I, Dahl H. Postcollection synthesis of ethyl glucuronide by bacteria in urine may cause false identification of alcohol consumption. Clin Chem. 2007 Oct;53(10):1855-7.
Wurst FM, Dresen S, Allen JP, Wiesbeck G, Graf M, Weinmann W. Ethyl sulphate: a direct ethanol metabolite reflecting recent alcohol consumption. Addiction. 2006 Feb;101(2):204-11