An ethanol assay’s analytical uncertainty does not automatically include ethanol formation before or after postmortem collection. Taylor and colleagues quantified uncertainty for a validated headspace GC–FID method used in road-traffic blood analysis. That study helps define measurement performance, but did not validate reconstruction of a living person’s ethanol concentration from decomposed blood.
What the method validation actually studied
The 2022 investigation used gas-tight syringe headspace sampling, two chromatographic columns and two flame-ionization detectors. Calibrators, quality controls and specimens were prepared in duplicate, with 100 µL sample added to 1 mL internal-standard solution containing tertiary butanol and sodium metabisulphite.
Materials included aqueous standards, defibrinated horse blood, volunteer human blood and proficiency-testing samples. The paper also reported results from 30 blood samples analyzed over approximately 18 months in the method’s applications. It did not describe a validation cohort of decomposed autopsy blood, with controlled collection sites, postmortem intervals and microbial activity.
A direct comparison with pressure-balance headspace sampling was not performed. The results therefore should not be described as proof that gas-tight syringe sampling outperformed that alternative.
Precision and bias depended on concentration
Calibration ranged from 10 to 400 mg/100 mL; both columns had R² values above 0.999 across 11 validation batches. Quality controls were studied at 20, 80 and 200 mg/100 mL.
The study found no significant bias at 80 or 200 mg/100 mL, but did find significant bias at 20 mg/100 mL. Its variance analysis also distinguished the lowest control level from the higher levels. A broad statement that the method was unbiased at every tested concentration would omit a material result.
What the reported uncertainty numbers mean
The authors reported expanded analytical uncertainties at 99.73% confidence of 5.31% at 20 mg/100 mL, 3.64% at 80 mg/100 mL and 1.95% at 200 mg/100 mL. At 80 mg/100 mL, the reported absolute expanded uncertainty was 2.89 mg/100 mL.
Those figures describe the study’s uncertainty calculation for the analytical procedure and tested levels. They are not a probability that a person’s ethanol was consumed before death, an allowance for microbial production, or an uncertainty interval for the concentration at an earlier event.
The authors also noted that later quality-control behavior appeared more variable than the initial validation suggested, with particularly narrow limits at the highest control. Validation performance and continuing casework performance should therefore be reviewed together.
Detection, quantification and origin remain separate
Ethanol was qualitatively detected at 5 mg/100 mL in four separate extractions on different days. The quantification limit was set at the lowest calibrator, 10 mg/100 mL. A signal below that quantification limit should not be presented as if it had the same demonstrated quantitative performance as the principal calibration range.
Dual-column chromatography contributes evidence about analytical identity and interference. It does not distinguish two ethanol molecules according to whether they originated from drinking or microbes. Both processes can yield the same measured analyte.
Applying the evidence to postmortem testing
First establish whether the actual laboratory’s method was validated for the specimen matrix and whether its reported uncertainty covers the concentration at issue. Then examine collection, preservation, temperature history and evidence relevant to formation or loss. The uncertainty of the chemical measurement cannot substitute for evidence about those separate processes.
The study’s road-traffic application does not validate a postmortem correction or establish impairment or cause of death from a concentration. It also does not make its 2022 discussion of legal practices a rule for another case or jurisdiction.
Our discussion of hospital and forensic ethanol comparisons illustrates another source of difference—matrix and method bias—that should remain distinct from postmortem change.
Primary source
Taylor L, Saskőy L, Brodie T, et al. Development of a Gas-Tight Syringe Headspace GC-FID Method for the Detection of Ethanol, and a Description of the Legal and Practical Framework for Its Analysis, in Samples of English and Welsh Motorists’ Blood and Urine. Molecules. 2022;27:4771. doi:10.3390/molecules27154771.
Leave a Reply