A measured nitrous oxide concentration from a living driver cannot simply be used as a reference boundary for an autopsy. A Danish study detected the gas in 52 of 62 selected traffic-case blood samples, but reported positive cases qualitatively because the total loss before analysis was unknown. Its findings establish testing feasibility under a specified workflow, rather than a validated scale of impairment or fatality.
A living-person study with a selected denominator
Lindholm and colleagues retrospectively reviewed all traffic-case samples specifically requested for nitrous oxide analysis in Eastern Denmark from 1 January 2020 through 31 December 2022. The 62 cases were selected because police suspected recent exposure. They were not a random sample of drivers or autopsy cases.
Fifty-seven drivers were male and five female. Ages ranged from 15 to 37 years, with a median of 22. Sixty cases came from Greater Copenhagen and two from rural Zealand. A rising number of requests could reflect changing suspicion or testing practice as well as changing exposure.
The 52 positive cases, or 84%, therefore describe this referred group. The percentage is neither population prevalence nor a validation sensitivity estimate: the study did not independently establish the exposure status of every negative case.
How the specimens were handled
Medical staff drew blood into 5 mL tubes containing 12.5 mg sodium fluoride and 10 mg potassium oxalate. Samples reached the laboratory within a few days and were stored at −20°C or colder until analysis. The report did not establish one fixed exposure-to-draw interval or one anatomical collection site for all cases.
The assay transferred 500 µL of blood to a headspace vial, sealed it immediately, and added xenon gas as the internal standard. HS-GC–MS analysis used freshly prepared aqueous calibrators at 0.1, 0.2, 1.0 and 10 mL/L, with quality controls before and after the case samples. Specimens were tested in duplicate.
Using the m/z 30 fragment for nitrous oxide reduced interference from carbon dioxide. This is a method-specific analytical choice; it does not establish that every mass-spectrometric nitrous oxide procedure has equivalent selectivity or sensitivity.
Calculated concentrations and qualitative reports were different
The positive samples had calculated concentrations reported as 0.1–48 mL nitrous oxide per L of blood. The laboratory defined a positive result as a duplicate mean above 0.1 mL/L and issued the result as “detected.” The rounded lower endpoint should not erase the stated above-cutoff rule.
The threshold produced a response approximately ten times the atmospheric background response in this method. It was a reporting choice for evidence of exposure, not an experimentally determined concentration at which driving performance becomes unsafe.
Although calibration produced numerical values, the authors deliberately limited the case reporting. Biological elimination, storage and opening the original tubes for transfer could all reduce the amount measured. The magnitude of their combined contribution remained unknown.
Why the numbers cannot be back-calculated
A fixed multiplier cannot recover the concentration at police apprehension from these results. The study did not establish individual elimination curves or quantify each specimen’s loss. A low measured value consequently does not measure the intensity of an earlier effect, and a negative result cannot precisely date or exclude every earlier exposure.
After death, the sampling context changes further. The living-driver dataset contains ongoing physiological elimination before collection, while autopsy specimens involve different sites, intervals and handling. The authors’ discussion of previously published deaths was not a paired living-to-postmortem experiment within these 62 cases.
The broader specimen lesson also appears in hospital ethanol specimens used in death investigations: retain the original matrix and timing rather than silently treating unlike results as interchangeable.
Co-detected drugs did not establish simultaneous intake
Only eight cases underwent blood-alcohol testing; three were positive. Fifty-two underwent testing for other traffic-impairing drugs, and 26 had findings above the study’s stated legal-limit criteria. These denominators are different from the 52 nitrous-oxide-positive samples. The study’s historical Danish classification should not be presented as a legal rule for another jurisdiction.
Neither the co-findings nor the nitrous oxide measurement alone established a specific drug combination’s contribution to driving behavior, a crash or a death. The useful conclusion is that targeted detection was possible in ordinary whole-blood samples, with substantial limits on quantitative reconstruction and postmortem extrapolation.
Primary source
Lindholm AØ, Nielsen MKK, Kristensen M, Rasmussen BS. Driving under the influence of nitrous oxide – A retrospective study of HS-GC-MS analysis in whole blood. Forensic Science International. 2024;354:111904. doi:10.1016/j.forsciint.2023.111904.
Leave a Reply