Higher mean ethanol concentrations in an autopsy group do not, by themselves, demonstrate that ethanol increased after death. A 2024 study compared 166 deceased people with 483 living cases. These were different groups, rather than paired measurements collected before and after death from the same people.
What population did the study describe?
Issa and Khattab included 649 alcohol-positive cases: 483 living cases and 166 autopsies. Records covered 2017–2020. Living cases included people seeking emergency care and people referred following legal or traffic-related events. Inclusion required positive alcohol findings, so this was not a prevalence survey of all hospital patients or all deaths.
Autopsy generally occurred within 24 hours of death. The methods excluded putrefied bodies, environmental contamination, open wounds that might hasten bacterial invasion, delayed collection beyond 24 hours and harsh environmental exposure. These restrictions matter: the dataset was not designed to measure ethanol production in advanced decomposition.
The measurements had different collection contexts
Living blood and urine were collected on hospital arrival. Autopsy blood came from the femoral vein; urine was obtained from the bladder by suprapubic needle extraction. Blood tubes contained a reported 1% sodium fluoride preservative. Vitreous was collected into dry tubes and urine into sterile containers. The paper describes refrigeration at 2–8°C and prompt vitreous analysis, with refrigeration if postponed until the next day.
Ethanol quantification used headspace gas chromatography with flame-ionization detection. This procedure was separate from screening and other instrumental drug testing described in the study. Identification or quantification of ethanol in a collected specimen cannot independently establish its origin.
What differences were statistically significant?
Table 5 reports a blood comparison of 166 postmortem and 483 living cases, with t = 2.465, df = 69.443 and p = 0.008. The urine comparison used 127 postmortem and 483 living cases, with t = 2.460, df = 35.640 and p = 0.009. Both met the study’s stated significance threshold of p ≤ 0.01, and the reported mean was higher in the postmortem group.
These tests address group differences under the analysis used. They do not identify why the groups differed. Drinking history, selection into hospital or autopsy casework, elapsed time and other characteristics could differ between groups. No paired within-person predeath/postmortem ethanol series was supplied for this comparison.
Nor does a significant p value quantify a microbial ethanol increment. It cannot be used to calculate a universal amount or multiplier to remove from an individual autopsy result.
Reporting inconsistencies limit numerical reuse
The main methods and group counts total 649 cases, but the discussion calls the sample size 549. Table 1’s printed age-group counts also exceed its stated postmortem denominator of 166. These inconsistencies should not be silently repaired by inventing a different population.
Table 5 does not label the units beside its concentration means, while other sections use percentage BAC and mg/dL. Assigning a unit to those unlabeled means, converting them or deriving a concentration ratio for practical correction requires clarification. The statistical comparison can be described without presenting those numbers as a validated correction factor.
What did the analytical reporting establish?
The assay section reports calibration spanning 10–400 mg/dL, a 10 mg/dL cutoff and dilution followed by reanalysis for results above 400 mg/dL. Elsewhere it lists eight calibration concentrations despite describing a six-point curve. The paper defines detection and quantitation limits conceptually without supplying unambiguous matrix-specific numerical values for both. Its reporting cutoff should not be relabeled as either limit.
These analytical qualifications are distinct from the study-design issue. Even flawless measurement would not transform two different groups into a longitudinal redistribution or microbial-production experiment.
What conclusion is justified?
The study reported higher means in selected alcohol-positive autopsy cases than in selected living cases. It did not measure the amount by which a particular person’s ethanol changed after death. An opinion about drinking, impairment or causal contribution requires the actual case evidence.
For a study that did observe repeated measurements in one specimen, see serial postmortem ethanol testing. That addresses a different question from a between-group comparison.
Primary source
Issa SY, Khattab A. Ethanol concentrations in various biological specimens: Living and postmortem forensic toxicology analysis and comprehensive literature review. Journal of Forensic and Legal Medicine. 2024;106:102737. doi:10.1016/j.jflm.2024.102737.
Leave a Reply