A fentanyl result reported as “greater than 40 ng/mL” supplies a lower bound, not an exact concentration of 40 ng/mL. A large Ontario death-investigation series illustrates why the distinction matters: most results above the method’s upper quantitation limit were not converted into exact measurements by dilution and reanalysis.
The selected population
Adamo, Marshall and Woodall reviewed 4,395 investigations submitted to Ontario’s Centre of Forensic Sciences during 2020–2022 in which fentanyl was quantified above the study’s 1.3-ng/mL inclusion threshold. The laboratory received 21,809 medicolegal investigations over that period. The 4,395 cases therefore represent a selected fentanyl-positive death-investigation dataset, not all deaths or all people exposed to fentanyl.
Most analyzed specimens were femoral blood: 3,898, or 89%. The remainder included heart blood, hospital samples and specimens with an unspecified source. Inclusion allowed autopsy blood from an intact vessel or an antemortem hospital specimen. There was no uniform death-to-collection interval for the full series.
The upper limit and the dilution workflow
The fentanyl calibration series extended to 40 ng/mL. Above-range specimens could be diluted and reanalyzed, or reported as greater than the upper quantitation limit, with a dilution factor applied where appropriate. Dilution integrity had been validated up to a factor of 20.
There were 614 cases with results above 40 ng/mL, accounting for 14% of the dataset. Seventy were diluted and given quantitative results from 41 to 483 ng/mL. The other 544 were reported semiquantitatively: 540 as greater than 40 ng/mL, and four with higher reported bounds after dilution.
Those four included one result greater than 160 ng/mL, two greater than 200 ng/mL and one greater than 2,000 ng/mL. The last used a dilution factor of 50, outside the study method’s validated dilution factor of 20. The authors retained it as an estimated concentration and acknowledged that limitation.
A bound cannot be treated as a point estimate
Two results reported as greater than 40 ng/mL cannot be ranked by magnitude from that wording alone. One could be modestly above the limit and another substantially higher. Substituting 40 for each result in a calculation would impose an assumption that the report did not establish.
The same caution applies to a result greater than 2,000 ng/mL. It should not be rewritten as exactly 2,000 ng/mL, and its dilution-verification limits must remain visible. The report’s inequality and qualification are part of the scientific result.
The study’s mix of exact and bounded values also matters when comparing averages, concentration distributions or apparent changes over time. A larger extreme value is not necessarily evidence that the typical concentration rose by a corresponding amount. Full specimen-level data and a stated treatment of bounded values would be needed for a new statistical analysis.
Analytical magnitude does not determine cause of death
The case with the estimated result above 2,000 ng/mL also had cocaine at 7.2 mg/L, benzoylecgonine above 20 mg/L and methamphetamine at 0.064 mg/L. Several additional screening findings were unconfirmed. Its cause of death was classified as mixed drug intoxication.
A different case had femoral fentanyl of 400 ng/mL, with methamphetamine, amphetamine and etizolam also detected, but the cause of death was a gunshot wound. Neither case creates a concentration threshold separating fentanyl-caused death from incidental detection. The forensic pathologist used toxicology together with the circumstances and other evidence.
What to retain when reviewing the report
The useful record includes the specimen source, units, calibration limits, actual dilution factor, validation scope and whether the result is exact, estimated or bounded. Screening identification, confirmation and numerical quantitation are separate parts of that record.
These distinctions complement method-specific analytical validation and specimen-specific fentanyl interpretation. A concentration, even when accurately quantified, does not independently establish impairment or causation.
Primary source
Adamo A, Marshall K, Woodall KL. Fentanyl-related deaths in Ontario, Canada: toxicological findings and circumstances of death in 4395 cases (2020–22). Journal of Analytical Toxicology. 2024;48:598–605. https://doi.org/10.1093/jat/bkae061.
Leave a Reply