Postmortem Blood Alcohol Concentrations – Are they Reliable?

Dead manPostmortem blood alcohol concentrations (BACs) tend to be less reliable and stable than BACs obtained from drinking drivers for the following reasons:

  • Postmortem blood is not sterile, over half of postmortem blood samples were found to contain bacteria and fungi (WOA70202)
  • Postmortem blood can have a much higher glucose concentration of 7 to 10X that of antemortem blood (WOA70201)
  • Postmortem diffusion of alcohol from the gut can falsely elevate the BAC (WOA70301).

Putrefaction

A high glucose concentration and bacteria/fungi/yeast found in postmortem blood makes an ideal environment for the production of alcohol as the sugar is converted eventually into alcohol causing a falsely high BAC:

Glucose = Acetaldehyde =  Alcohol

Yeasts, the most efficient fermenters, can convert approximately 100 mg of glucose into 40 to 50 mg of alcohol.  Bacteria and fungi typically can only convert 100 mg of glucose into 10 to 20 mg of alcohol.

During fermentation, other volatiles such as n-propanol and acetaldehyde are produced which can be used as markers of putrefaction.  Other samples such as urine and vitreous humor are more stable than blood and can also indicate putrefaction.  This is illustrated in the following postmortem case in which I was the forensic toxicologist (WOA70211).

Case report

A 31 year old man was found dead several hours after he apparently committed suicide via a fall from a great height.  At autopsy the next day, peritoneal cavity blood and urine were collected.  The blood was placed in a jar without preservatives and the urine in a tube with 1% NaF.  Nineteen days later, the samples were received at the laboratory and a headspace gas chromatographic analysis showed a BAC of 0.096 g/100mL.

This result raised all sorts of questions.  Where did the victim drink?  Was the BAC related to the fall?  Did he stumble due to the BAC and the fall was an accident and not a suicide?

Fortunately, analysis of the urine and a more detailed analysis of the blood showed:

Volatile Detected Blood (g/100mL) Urine (g/100mL)
Alcohol (Ethanol) 0.096 0
N-propanol 0.004 0
Acetaldehyde 0.003 0

It has been suggested that the concentration of alcohol produced postmortem is about 20 to 25X that of the n-propanol produced (WOA70208, WOA702U2), which would indicate that all the alcohol was produced postmortem.  Combined with the zero urine alcohol concentration and elevated blood acetaldehyde concentration, it can reasonably be concluded that the victim was alcohol-free at the time of death.  This case also illustrates that, especially for postmortem alcohol analysis, t-butanol should be used as the internal standard and not n-propanol.

V.D. Pleuckhahn

My favourite researcher and writer in this field is the Australian pathologist V.D. Pleuckhahn, who published several important studies nearly 50 years ago on postmortem blood alcohol levels.  His conclusions, however, are as valid today (WOA70202):

      1. Blood alcohol levels at autopsy are valid up to 48 hours after death when simple principles are observed in the collection and storage of samples
      2. Alcohol levels in samples of blood taken from the intact heart are as significant as levels of blood from the femoral veins
      3. False blood alcohol levels greater than 0.200% can be generated in autopsy blood samples which are not correctly stored.
      4. High blood alcohol levels may develop during putrefaction and levels up to 0.200% do not necessarily indicate that alcohol was imbibed before death.
      5. Significant false blood alcohol levels do not develop during incineration in absence of putrefaction.

Pleuckhahn also stated that the more specific GC method be employed for postmortem alcohol analysis rather than the chemical methods typically used at the time.

What complicates the postmortem blood alcohol issue even further is that, not only can putrefaction increase the BAC, but microorganisms can also decrease the BAC by up to 50% (WOA70207).

In my next blog on the reliability of postmortem BACs, I will discuss the effects of postmortem diffusion of alcohol on BAC.

 


Comments

Postmortem Blood Alcohol Concentrations – Are they Reliable? — 34 Comments

  1. Hi I have a question.

    Someone passed away in an industrial accident. If blood samples were taken in under 48 hours, but the official autopsy and blood tests will only be done after 5 days post death. Will the alcohol levels still reflect in the tests?

    • The longer the time between death and the time the blood samples are collected for alcohol, the greater the risk for an increase in BAC by putrefaction. If the body is refrigerated during the 5 days storage it would lessen the risk somewhat. The samples collected under 48 hours would be the best for alcohol analysis. Hopefully peripheral and central blood were collected as well as urine and vitreous humor as they would confirm that no putrefaction occurred.

      Also, if it is a traumatic death and the stomach is ruptured there is an additional risk of alcohol and microorganisms diffusing through the central body region and a greater risk of artefactual BACs.

  2. Maybe you can assist me in the following case:
    A 45 year-old man was found dead in a wine fermentation tank. The body was found half-immersed in the second-day of grape fermentation. The death occured about 10 hours before the body was discovered. The tank had a high CO2 concentration and some alcohol already formed in the fermentation material. The autopsy report revealed no signs of physical violence or putrefaction. The lungs had large amount of fluid. The femoral blood ethanol concentration was 1.5 g/dl. Our interpretation was that the man was drunk and fell into the tank. He died due to CO2 inhalation. Some colleagues argued that the blood ethanol concentations could be related to TGI and skin absortion inside the tank. I do not buy that, because the blood etanol levels were too high. Am I correct?

    Thanks,
    Gilberto

    • If the postmortem femoral blood alcohol concentration was 1.5 g/100mL and not 0.15 g/100mL, then that would indicate that the BAC was an artefact and probably caused by postmortem diffusion of alcohol through the skin after death. No significant BACs can be obtained in living subjects from dermal absorption because the intact adult skin is not very permeable to alcohol and the liver eliminates rapidly any alcohol that does enter the blood. At death, however, the skin does not retain its structure and any alcohol that does enter the body is not removed by the liver but builds up. I have some studies cited in my book WOA70307 and WOA70308 about postmortem diffusion of alcohol through the skin.

  3. This is fascinating! I am involved in a case in which two men were killed in a car accident after a night of heavy drinking in two bars. An autopsy was performed on the driver of the vehicle on February 21, 2012 at 11:30 a.m. The time of death was 2:00 a.m. on Feb. 19, 2012 – the cause was drowning. The vehicle, after going off the road, overturned into a drainage chute. The driver’s toxicology report reveals ethanol .18 (specimen: one vial fluoridated heart); and ethanol .19 (specimen: vitreous humor). The ME sent the specimens for the toxicology evaluation and the toxicology report says registered on Feb. 23, 2012, four days after death. Am I correct in assuming that, from these results and the lag of time, we can tell nothing about the driver’s blood alcohol at the time of death?

    • In regards to postmortem putrefaction, the important time is the time from death to autopsy. In this case it appears to be less than 2 days and so is not problematic. Once the blood is in a fluoridated tube it is not important how long it takes for the forensic lab to analyze for alcohol.

      It is unfortunate that heart blood was collected rather than peripheral blood which is less susceptible to postmortem changes. You should also check to see that n-propanol was not used as the internal standard for the GC analysis as it is a marker of postmortem changes.

      The vitreous humor should be OK. However, using a mean ratio of 0.84, a VHAC of .19 would indicate a BAC of 0.14. This would indicate that the driver died while still in the absorbing phase of the alcohol curve and there was recent drinking prior to death.

  4. Need help with this toxicology report…husband died in tragic car accident. his toxicology report came back stating: Ethanol 0.22g/100ml and then whole blood sample 0.19g/100ml Does that mean he was DUI…don’t understand two reports and why different reports..

    Thank you…Michelle

    • Hi Michelle, As is often the case, more information is required for analysis. Did the husband die instantly or did he survive long enough such that the blood samples were taken in hospital? Exactly where were the blood samples collected, was the one sample serum? or urine? The ratio between the 2 samples is 1.16 so I suspect the other sample is serum and the 2 agree, but we need more information before being definite. If you can provide further information, I will endeavour to assist.

  5. Family member recently passed in boating accident. Our best guess was that he died somewhere between 2 and 4 a.m. Body was not discovered until around 5 p.m. that same day with blood taken. 3 tubes of femoral and one tube of urine. Analysis stated ethyl alcohol result by gas chromotography with result 0.274 (+/-.010)grams per 100 ml.

    Body was exposed to the elements for approximately 15 to 13 hours.Boat was caught up in the marsh.

    In your professional opinion, would the BAC be materially overstated given these circumstances? Please note it is known that the family member was drinking, but the question is how much?

    • Hi Charles, I need a little more information in order to make a reliable analysis. Can you provide:

      Office cause of death? ie. drowning, hypothermia, etc.
      Environmental temperature at the time of the accident?
      Is he a type 1 diabetic?
      The urine and blood alcohol concentrations separately- not just 0.274 g1/00mL?

      • Thank you for responding to my initial request. Please note the following:

        1.)Official cause of death was head trauma. Boat struck beam supporting boardwalk to a dock.
        2.)I pulled the historical ambient temperatures by hour. It would appear that the temperature at estimated time of accident was 57.2 degrees F. We estimate that the accident took place between 2 and 4 a.m. By 9:00 a.m. the temperature was 62.8 degrees. Please note the following hourly temperatures:
        10:00 a.m. – 72.9 degrees
        11:00 a.m. – 75.2 degrees
        1:00 p.m. – 77.9 degrees
        2:00 p.m. – 81.5 degrees
        3:00 p.m. – 80.6 degrees
        4:00 p.m. – 79.2 degrees
        5:00 p.m. – 79.0 degrees
        We suspect blood and urine was taken around 5:00 p.m. when the boat was pulled back to the ramp. I can get the definitive time if that time is material to your evaluation. The toxicology report suggests that the blood analysis was done on November 4th.

        3.) Our loved one was not a diabetic.
        4.) The feedback we are getting is there was only an analysis of the blood for both alcohol and drugs.

        Thank you again for helping us to piece together those circumstances in the tragic loss of our beloved.

        • If the blood was collected from the femoral vein then I would expect the BAC to be valid. To make sure you should get the urine analysed as well to confirm the results. I would however, highly recommend obtaining the assistance of a personal injury or wrongful death lawyer especially if insurance money is involved.

      • Admin,

        Did you get my earlier replies to your questions? Both were displayed on your blog site earlier awaiting moderation, but neither is displayed for me now.

  6. I have an autopsy report that lists the BAC take from the vitreous at 0.314 and the BAC taken from the femoral also a 0.314. If the number from the vitreous should be divided by two, how can both sources be the exact same? This man showed absolutely no obvious signs of intoxication just 20 min’s prior to car wreck that took his life. The autopsy was performed approximately 8 hours after death.

    • The dividing of the vitreous humor alcohol concentration by 2 to obtain a blood alcohol concentration is used when you only have vitreous humor and no blood. It prevents the vitreous humor from overestimating BAC. In this case, you have femoral blood alcohol concentration which being from the leg vein and collected within 8 hours of death would indicate that the BAC of the driver at the time of death was 0.314 g/100mL. A more realistic ratio for VHAC to compare with BAC would be 1.17 (WOA70604). Thus the VHAC of 0.314 g/100mL would indicate a BAC of 0.268 g/100mL, which is still lower than the femoral BAC.

      This is due to the fact that the VHAC lags behind BAC and indicates the BAC at a prior time. Therefore, the VHAC compared to the femoral BAC indicates that drinking occurred recently prior to death.

      No obvious signs of intoxication at a BAC of 0.314 g/100mL is not uncommon in individuals with a high tolerance to alcohol (WOA50612) .

  7. I have an autopsy report stating an alcohol concentration 250mg per 100 grams of liver. No BAC or other samples reported
    Is this within a toxic level? How reliable is the measure? Can anything be concluded from this isolated result?

    • It is difficult to comment without more details of the case and possible factors which could have affected the liver alcohol concentration. The average ratio of liver alcohol concentration to blood is 0.56:1 which would indicate non toxic BAC of 140 mg/100mL. But the range of ratios is very large from 0.1 to 1.40:1 (WOA70809). Certainly a liver alcohol concentration in isolation may not be a reliable indicator of the BAC at the time of death.

      • Thanks for the response. Unfortunately, I don’t have more information other that is was a sample from a 62yr old male with no history of alcohol abuse who committed suicide in the subway. Probably the body suffered substantial trauma and it was impossible to obtain blood or urine samples. The report also mentions that all chromatography analyses were negative without specifying for which substances

        • You may be interested in a study of 211 subway deaths that occurred in NYC between 2003 and 2007. Fifty-two percent of the deaths were suicide- alcohol was detected in 14% of the victims and the mean BAC was 160 mg/100mL. Thirty-six percent of the deaths were accidental and 42% of these victims had a positive BAC (mean 200 mg/dL). It also found that certain physical injuries such as decapitation and torso transection were more indicative of suicide (WOA61506)

          • The title of the publication is LIN, P.T. AND J.R. GILL. “Subway Train-Related Fatalities in New York City: Accident versus Suicide.” Journal of Forensic Sciences, 54: 1414-18, 2009.

  8. I have a Toxicology Report which indicates a liver ethanol concentration of 97 mmol/kg
    The victim was a 25 year old male who died in a car accident in which he received 4th degree burns involving 100% of total body surface. 90% of the surface of the liver was charred, and the depth of charring extended to a depth of 0.5 to 1 cm into the parenchyma. The conclusion of the toxicologist was that he was acutely intoxicated at the time of driving (0.17%). It is my opinion that to try to correlate the liver alcohol level to a blood alcohol level ante-mortem would not be prudent and it would just not be good science to use this sample to make such a conclusion. Thoughts anyone?

    • As you indicated there are numerous factors to consider. First is there any contamination of the liver sample by the fire and chemicals used to put the fire out. Then there is the potential for putrefaction causing the production of alcohol. Finally there is the wide variation of ratios to convert a liver alcohol concentration into a BAC. In one study, WOA70809, the mean liver:blood alcohol ratio was 0.56:1 but the SD was very high at 0.30. The range of ratios was 0.0 to 1.40:1 and r2 was low at 0.716. I would be reluctant to use the liver alcohol concentration alone to determine BAC. Other evidence such as the drinking pattern by the deceased could be useful to verify the liver alcohol concentration conversion.

  9. Hello.
    I have a toxicology report that lists the following results regarding the presence of ethanol. Is there a way to determine which is the most accurate?
    Thanks!

    Urine — 0.261%
    Bile — 0.214%
    Vitreous — 0.164%
    Brain — 0.176%
    Blood (Chest) — 0.157%

  10. I have a question….

    Post M Toxicology report says

    Blood Heart 0.05

    Blood Femoral 0.06

    Vitreous 0.09

    0.08 Is the legal limit In Maryland.
    Would the report for Vitreous mean he was over the limit?

    Thanks in advance

    • Hi Eli,

      Vitreous humor is only an indirect measure of blood alcohol concentration and has to be converted into a BAC. A study cited in my book – WOA70605 – states the the vitreous humor alcohol concentration should be divided by 2, which would indicate that the BAC would be 0.045 g/100mL, much below .08. So all three measures of alcohol indicate a BAC less than 0.08 g/100mL.

      Hope this helps.

  11. Whilst assisting in postmortems in the 1980s’, the main place we sampled blood was from the heart, namely because I was told to believe it was the best place to achieve a full blood sample as the
    human body absorbed the water that blood is mainly made of ?

  12. Pingback: Postmortem Diffusion of Alcohol and Trauma

  13. Another great blog entry, Jim. The microbiologist remnant within me, however, forces me to mention that yeasts are in fact members of Kingdom Fungi. Most of the familiar yeast genera (e.g., Candida & Saccaromyces) are in Phylum Ascomycota (sac fungi), but some others are instead found in Phylum Basidiomycota (filamentous fungi). So perhaps the statement about conversion of glucose to alcohol by “Bacteria and fungi” should be revised to “Bacteria and fungi (other than yeasts)”

    What I’ve learned about postmortem BAC levels pretty much jives with Pleuckhahn’s principles as you listed them, but I think it’s worth mentioning that if the GI tract is ruptured above the diaphragm, such that gastric contents are spilled into the thoracic cavity where they can bathe and perfuse the pericardium, it is possible for blood in even an intact heart to become contaminated by alcohol from the gastric contents. That’s why some medical examiners routinely prefer blood from the major vessels of the extremities versus heart blood, at least where alcohol is concerned.

    Regarding Kirk’s comment, another good sample that resists post-mortem changes is cerebrospinal fluid. Thus CSF is a viable alternative when vitreous is not available.

  14. peptides/protein hydrolysis also conduct to ethanol production by fermentation. aerobic flora makes deamination and fermentation

  15. Kirk

    Thank you for your comment about vitreous. I think that there are a lot of postmortem forensic toxicology labs are now doing blood, vitreous, and urine on most of their postmortem cases (collected in preservative/anticoagulant). This information helps to make the interpretation of possible alcohol ingestion a lot easier than nonpreserved blood samples alone.

  16. I did a paper in law school on this subject (1974). You are correct as far as you go. However, the vitreous in the eye is the last place in the human body to generate alcohol from body decay. It can be extracted using a syringe and tested in the same manner as blood. I interviewed the medical examiners in a number of major cities and none of them did this and some said it would be too difficult.

    I’m on the road but let me know if you are interested in the paper and I can send you a copy.

    Kirk

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>