Increasing incubation temperature from 25°C to 37°C did not uniformly increase ethanol production in a three-species bacterial experiment. The response depended on the organism and on which alcohol was measured. Those results limit any simple rule that a warmer postmortem specimen must contain a proportionately larger microbial ethanol contribution.
The experimental environment was defined
Velivasi and colleagues studied individual reference strains of Klebsiella pneumoniae ATCC 13883, Staphylococcus aureus ATCC 25923 and Enterococcus faecalis ATCC 29212. Each was inoculated separately into Brain Heart Infusion broth at approximately 106 colony-forming units/mL.
The main 25°C cultures were followed for 23 days. One arrangement was strictly anaerobic. The other began aerobically for five days, then used a paraffin overlay and anaerobic jars. Parallel controls were incubated at 37°C for 48 hours and 4°C for 23 days. Two tubes per sampling point from each culture set were removed for alcohol measurement. These repeated culture specimens were not independent human deaths.
Temperature effects differed by species
K. pneumoniae produced the most ethanol under the main experimental conditions, followed by S. aureus and E. faecalis. Ethanol generally rose rapidly during the first two days at 25°C and then plateaued or, for S. aureus, declined slowly.
The maximum ethanol concentrations for K. pneumoniae and E. faecalis at 37°C were comparable to those at 25°C. In contrast, S. aureus produced lower ethanol concentrations at 37°C than at 25°C. The paper describes these patterns but does not provide a formal statistical demonstration of equivalence between the temperature conditions.
The other alcohols did not all move with ethanol. At 37°C, K. pneumoniae produced more 1-butanol but less 1-propanol, isobutanol and methyl-butanol than at 25°C, while its ethanol maximum was similar. A temperature effect on one congener therefore cannot be substituted automatically for an effect on ethanol itself.
Oxygen history was another experimental variable
Strictly anaerobic incubation yielded slightly higher ethanol concentrations than the aerobic-to-anaerobic arrangement for all three species. The transition was intended to approximate declining oxygen availability after death. It did not reproduce the complete microbial community, nutrient changes or temperature history of a human body.
The species also differed in their dominant non-ethanol alcohol. At 25°C, 1-propanol predominated for K. pneumoniae and S. aureus, whereas 1-butanol predominated for E. faecalis. This matters when interpreting a congener pattern: identical ethanol origins need not produce one fixed companion-alcohol ratio across all organisms and conditions.
What did the refrigerated controls show?
No ethanol or other studied volatiles were produced in the 4°C broth controls during the 23-day incubation under either aeration arrangement. That is useful evidence for the tested strains, medium, inoculum and duration. It does not establish that every refrigerated autopsy specimen is unchanged or that refrigeration removes ethanol already formed before collection.
The study separately inoculated pooled human blood heated at 50°C for about two hours to denature proteins. The resulting alcohol patterns differed from broth, and E. faecalis produced negligible alcohols in that medium. Even this substitute for postmortem blood was not equivalent to an intact decomposing body.
What was measured, and what remains unknown?
Headspace GC–FID measured ethanol and five other alcohols. Table 1 reports ethanol calibration of 0.005–4.0 g/L, a detection limit of 0.001 g/L and a quantitation limit of 0.005 g/L. Its ethanol quantitation limit differs from a figure in the prose, an internal reporting issue relevant to method reproduction. Analytical performance does not supply a human postmortem production rate.
The 25°C and 37°C experiments also had different observation periods. They should not be used to invent a universal temperature coefficient, hourly production rate or amount to subtract from a person’s result. The findings establish neither drinking, impairment nor ethanol’s role in death.
For evidence from a different design, see serial changes in a preserved human postmortem specimen.
Primary source
Velivasi G, Kourkoumelis N, Sakkas H, Boumba VA. Modeling microbial ethanol production by S. aureus, K. pneumoniae, and E. faecalis under aerobic/anaerobic conditions — applicability to laboratory cultures and real postmortem cases. International Journal of Legal Medicine. 2021;135:2555–2565. doi:10.1007/s00414-021-02638-4.
Leave a Reply