In a controlled blood-surrogate experiment, adding glucose greatly increased ethanol produced by Candida albicans. The findings show why substrate and specimen conditions matter. They do not establish a fixed postmortem ethanol amount for a person with diabetes or for any specimen with an elevated glucose measurement.
The starting material was manipulated human blood
Velivasi and colleagues pooled blood from healthy donors whose glucose ranged from 0.9 to 1.3 g/L. The blood was heated at 50°C for approximately two hours to denature proteins. This was intended as a surrogate for postmortem blood, rather than an intact-body model or a clinical study of people with diabetes.
Half was acidified to pH 5.5 ± 0.2; the remainder had a pH of 7.2 ± 0.2 after heating. For each pH condition, five series of ten tubes containing 2 mL blood received 0, 1, 3, 7 or 11 g/L additional glucose. “Additional” matters: the zero-addition series was not necessarily glucose-free.
Each tube was inoculated with C. albicans ATCC 10231 to approximately 10⁶ colony-forming units/mL. Cultures ran at 25°C for seven days, with an initial aerobic phase followed by anaerobic conditions. Two tubes from each series were analyzed on days 0, 2, 4, 5 and 7. These are repeated culture samples, not corresponding numbers of independent autopsy cases.
The ethanol ranges changed with added substrate
Without added glucose, ethanol ranged from 0.01 to 0.08 g/L in acidic blood and 0.01 to 0.07 g/L in neutral blood. With 3 g/L added glucose, the respective ranges were 0.82–1.19 and 0.99–1.25 g/L.
At 11 g/L added glucose, acidic cultures reached a reported range of 1.98–4.72 g/L; neutral cultures ranged from 1.99 to 3.78 g/L. These are experimental ranges over the sampled period, not expected concentrations in every similarly glucose-rich forensic specimen.
The authors characterized ethanol production as increasing with glucose content. They did not provide a validated patient-specific equation converting a glucose concentration or a diabetes diagnosis into the amount of ethanol formed after death.
pH was one factor within a larger system
Acidic and neutral cultures were tested in parallel, and higher-alcohol maxima were broadly comparable between them. The observed results do not support a rule that acidification invariably raises every alcohol or produces a fixed percentage increase.
A separate dextrose-dilution arm also changed the blood matrix and carbohydrate supply together. It should not be interpreted as an isolated test of dilution alone. The study’s authors identified temperature, time, glucose, pH and changes in medium composition as relevant influences.
What the analytical method established
Samples were measured in duplicate by headspace GC–FID. Ethanol had a reported detection limit of 0.1 mg/dL and quantification limit of 0.5 mg/dL, with a calibration range of 0.5–400 mg/dL. The reported calibration R² exceeded 0.999 for the measured alcohols. The two methyl-butanol isomers were measured separately but reported as their sum.
These data concern quantitative chemistry in inoculated cultures. There was no impairment assessment, controlled predeath drinking history or determination of a cause of death in this experimental arm. The extreme experimental yields cannot be presented as evidence that a high ethanol result in a particular death was necessarily microbial.
The highest reported ethanol yields exceeded the method’s stated upper calibration concentration of 400 mg/dL. The paper does not detail the dilution or extended-range handling for every such result. Those high values should retain that analytical-documentation limitation rather than be treated as independently reproducible case-specific predictions.
Why transfer to casework requires restraint
The same paper tested fungal models against 93 selected autopsy samples. Broth-derived models succeeded in only six cases across two different sets of three, using the authors’ ±40% agreement criterion. Microbial burden and glucose were unknown in those six cases. Good performance in a deliberately favorable culture therefore did not establish broad applicability to authentic blood.
Review measured substrate, specimen preparation and microbial evidence rather than inferring formation from a medical label. Our article on EtG and EtS interpretation addresses a complementary evidentiary question about exposure; it does not supply a glucose-based subtraction formula.
Primary source
Velivasi G, Sakkas H, Kourkoumelis N, Boumba VA. Modeling postmortem ethanol production by C. albicans: Experimental study and multivariate evaluation. Forensic Science International. 2021;324:110809. doi:10.1016/j.forsciint.2021.110809.
Leave a Reply