A strong relationship between postmortem urine and blood ethanol does not establish a universal conversion for an individual. In a Lithuanian series of 378 selected autopsies, urine ethanol was usually higher and the two matrices were strongly correlated. The study did not validate a case-specific blood estimate or determine when each person drank.
Which deaths entered the study?
Okulevičiūtė and colleagues retrospectively selected 378 nonconsecutive cases from 1,504 autopsies performed during 2016–2023. Cases were selected for ethanol findings across body fluids or tissues; cases with ethanol detected only in blood or only in urine were excluded. People younger than 18 and those treated in hospital before death were excluded.
The cohort contained 319 males and 59 females. Signs of decay were recorded in 53 cases, approximately 14%. These inclusion rules matter: the results cannot estimate the frequency of urine–blood disagreement among all deaths, or describe hospital-treated patients without further evidence.
What the paired blood and urine results showed
Table 1 reports mean blood ethanol of 1.86 ± 1.27‰ and mean urine ethanol of 2.49 ± 1.40‰. The median concentrations were 1.76‰ and 2.54‰, respectively. Here the per-mille symbol, ‰, is retained as reported; it must not be silently replaced with percent or g/dL.
Urine exceeded blood in 86% of cases, while blood exceeded urine in 14%. The authors reported a significant difference between the matrices, p < 0.05, and Spearman’s rank correlation r = 0.93, p < 0.05.
The narrative reports a mean difference of 0.51‰ and median difference of 0.46‰. Its mean difference does not equal the 0.63‰ difference between the displayed group means. The publication does not reconcile that discrepancy; neither figure should be turned into a fixed individual adjustment.
Correlation is not a validated prediction interval
Spearman’s r describes how concentrations rank together across the selected cases. It is not a regression slope, a urine-to-blood ratio or an error bound for a new decedent. An r of 0.93 does not mean that dividing every urine value by 0.93 reconstructs blood ethanol.
The study did not provide an externally validated conversion with individual prediction limits. Nor did it establish an antemortem blood reference for each case. Even an accurate estimate of the collected postmortem blood specimen would be a different target from the concentration at an earlier event during life.
The specimen methods limit further inferences
The methods describe collection during the postmortem examination and headspace gas chromatography for ethanol. They do not specify the actual blood vessel for each case, a uniform death-to-collection interval, or a detailed case-specific preservative and storage history. Recommendations about femoral blood and preservation in the introduction should not be mistaken for documented methods in every included case.
The paper does not supply complete ethanol-method detection limits, quantification limits or validation performance. Its LC–MS methods refer to drug testing and should not be presented as mass-spectrometric confirmation of every ethanol result.
Do the results identify the drinking phase?
The authors discuss absorption, elimination and sequential urine sampling, but their primary dataset is a retrospective autopsy comparison. It did not follow controlled drinking or serial urine voids in living participants. The finding that urine was higher in most cases cannot date the last drink or prove an elimination phase in a particular decedent.
Likewise, the authors found no statistically significant decay-group difference in blood or urine ethanol, with p = 0.09 and p = 0.21, respectively. Those null results do not prove that microbial formation was absent: the design did not directly measure how much ethanol microbes produced.
For a different cross-specimen issue, see hospital serum and forensic whole-blood ethanol. Matrix-specific interpretation must precede conclusions about impairment or cause of death.
Primary source
Okulevičiūtė A, Chmieliauskas S, Laubner-Sakalauskienė G, Badaras R, Laima S, Vasiljevaitė D, Stasiūnienė J. Postmortem Analysis of Ethyl Alcohol Concentration in Blood, Urine, Muscle and Bile. Acta Medica Lituanica. 2024;31(2):207–216. doi:10.15388/Amed.2024.31.2.6.
Leave a Reply