Auto Brewery Syndrome – Does it affect BAC Test Results in Diabetics?

endogenous BAC in diabeticsThe issue of auto brewery syndrome (endogenous BAC) has occasionally been raised in court as a defense against drunk driving. It is also thought that diabetics may be more prone to high endogenous BAC due to the metabolic disturbances, high blood ketones and sugar concentrations caused by this disease. The recently published paper I have chosen for my second Research Update is:

Simic, M., Ajdukovic, M., Veselinovic, I., Mitrovic, M., and Djurendic-Brensel, N., “Endogenous Ethanol Production in Patients with Diabetes Mellitus as a Medicolegal Problem”, Forensic Science International, 216: 97-100, 2012

In this study, blood samples were collected from 100 patients (of both genders) diagnosed with diabetes mellitus in the early morning after an overnight fast and analysed for alcohol by the specific headspace gas chromatography (HS-GC) method. In another 30 diabetic patients, blood and urine samples were analysed for alcohol by the less specific Widmark method (wet chemical potassium dichromate oxidation) and by HS-GC. Blood samples were also collected from 30 healthy subjects (controls) and analysed for alcohol by both the Widmark and HS-GC methods. Urine alcohol concentrations were also determined by both methods in 30 diabetic patients and 15 controls.

The mean endogenous BACs detected were as follows:

Mean BAC (g/100mL) as determined by Widmark Method Mean BAC (g/100mL) as determined by HS-GC Method
Control Subjects 0.0007 0.00004
Diabetic Patients 0.0027 0.00027

Note:  All BACs as determined by the HS-GC method were less than 0.001 g/100mL. The highest UAC detected by HS-GC in the diabetic patients was 0.0026 g/100mL.

There are several interesting points that can be seen from the Table. One, is that there are a lot of zeros after the decimal point indicating extremely low endogenous BACs. So low, in fact, that they would normally not be detected routinely in forensic laboratories. With that said, diabetic patients do tend to have a higher endogenous BAC than the control subjects. However, the endogenous BACs determined by the less specific Widmark method are at least 10X greater than by the HS-GC method. This may explain some of the high endogenous BACs reported by older studies which used the Widmark method to analyse alcohol.

Blood glucose concentrations were also measured in the diabetic patients and there was no correlation between the blood sugar concentration and the endogenous BAC (i.e. auto-production of alcohol does not depend on blood glucose concentrations). This can be seen from the next Table which lists some of the results in the diabetic patients.

Blood Glucose Concentration (mg/dL) Endogenous BAC (g/100mL)
90 0.00026
216 0.00010
252 0.00010
432 0.00038
666 0.00010

Speculative Science – not in court please!

The extremely low endogenous BACs detected have been confirmed by other studies, which I have summarized in my book in section 2.05 on Endogenous BACs (WOA20501-WOA20506). One study lists four criminal cases of the auto brewery syndrome that occurred in the UK (WOA20506). Needless to say, the science was very speculative and self-serving.  One defense expert had never heard of the Widmark factor and calculated BACs based on plasma volume (about 4L) not on total body water (about 40 L). In another case the charged drinking driver did a poorly controlled reconstruction experiment in which he consumed approximately 5 grams of glucose and obtained an apparent endogenous BAC of 0.019 g/100mL, one hour later. This is in spite of the fact that the complete fermentation of 5 g of glucose would only result in a maximum BAC of 0.005 g/100mL and that is not taking into account the elimination of alcohol in that hour. It was strongly suspected that “there had been an attempt to falsify the scientific evidence submitted for analysis”. In none of the cases did the autobrewery defense result in an acquittal.

This article also calculates that large quantities of carbon dioxide (approximately 6 L/hour) would have to be generated in order to produce significant endogenous BACs by auto- fermentation and state, “As this gas is only slightly soluble, removal would have to be by eructation or flatus at a rate around 100 mL per minute. In a patient who is strictly monitored, such a phenomenon should have been observed and worthy of mention.”


The authors of this featured Research Update concluded:

“From the forensic point of view, such low values of ethanol cannot affect brain function and the ability of drivers of motor vehicles to safely participate in traffic.”

The paper can be obtained from the Forensic Science International website at




Auto Brewery Syndrome – Does it affect BAC Test Results in Diabetics? — 4 Comments

  1. Pingback: Drinking and Driving - Not Just a Night Thing!

  2. I normally see this defense as claiming that higher levels of glucose or dextrose of a diabetic, in the blood will ferment due to one of many yeasts or bacteria after the antemortem blood was drawn, thus elevating the blood results from the time of driving. Of course it also is based upon a claim of either missing or innefective preservative in the sample.

  3. I am a forensic scientist. I am type 2 Diabetic.
    I do conduct research myself on myself on several occasions.
    I have noticed that if I take 2 tots of whisky and measure the level next day it goes down to abt 5.4 m/mol.
    When I double the dose by taking 4 tots the result is abnormally high producing a level of 9m /mol glucose…. with same type and amount of food.
    This observation may give a lead to researchers to see the way forward

  4. Even if ante mortem endogenous alcohol production were a significant phenomenon, it wouldn’t “affect BAC test results.” The test results show the amount of alcohol in the blood, not where it came from. If the results are accurate, then the source of that alcohol is irrelevant to the BAC results. They are what they are. Source is also irrelevant to the impairment produced by alcohol. Whether produced in the body or imbibed from outside the body, alcohol is alcohol, and the resulting impairment is the same. DUI laws aren’t designed to prohibit drinking; they are designed to prohibit impaired driving by alcohol, so the source of alcohol isn’t really material to the charge or to the threat to road safety. In fairness, if endogenous alcohol were significant (it clearly is not, but if it were), then it should be a mitigating factor on the first offense, since the subject might not have been aware of it. But once proven to occur in a given individual, then that individual should be legally required to take that into consideration when deciding to drive (knowing that they cannot drink as much before driving as a “normal” person can). People who know they need corrective lenses to see clearly are not excused for driving without them just because their vision impairment is “natural”; they are required to compensate for that natural impairment in their abilities, and if they cannot, then they are not permitted to drive. It would be the same if someone were found to produce significant endogenous alcohol. But all this of course is only philosophical musing, because research has clearly demonstrated that it does NOT occur.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>