Environmental Toxicology

Contents lists available at ScienceDirect

Forensic Science International

journal homepage: www.elsevier.com/locate/forsciint

Common opioids and stimulants in autopsy and DUID cases: A comparison of measured concentrations

Simon Kjær Hermansen⁎, Dorte Jensen Christoffersen

Section of Forensic Toxicology, Department of Forensic Medicine, Faculty of Health Sciences at the University of Southern Denmark, Odense, Denmark

a r t i c l e i n f o

Article history: Received 5 May 2022 Received in revised form 7 July 2022 Accepted 11 July 2022 Available online 12 July 2022

Keywords: DUID Autopsy Toxicology Intoxication Comatose-fatal levels Drugs of abuse

a b s t r a c t

Quantitative results from toxicological analyses of autopsy material are widely compared to ranges in re- ference works to determine if drug concentrations are in relevant levels for establishing intoxication. This study compares concentrations of commonly used opioids and stimulants from drug addict autopsies and driving under the influence of drugs (DUID) cases to supplement current knowledge of the possible span and overlaps of measured concentrations. The study included whole-blood results from forensic autopsies of drug addicts performed from 2015 to 2020 (n = 220) and DUID cases from 2015 to 2019 (n = 7088). The focus was on heroin/morphine, methadone, cocaine, amphetamine and MDMA concentrations because these drugs are commonly encountered in both fatal intoxications and DUID cases and the potential for abuse is well known. In the DUID group, the opioids heroin/morphine and methadone and the stimulants amphetamine and MDMA were often seen in concentrations above the reported lower comatose-fatal level whereas cocaine was almost always below. Thus, based on our data, the potential for false assessment of intoxication cases when comparing to reported comatose-fatal limits appears greatest on lower end con- centrations of heroin/morphine, methadone, amphetamine and MDMA, whereas false assessment of co- caine appears less likely because most control cases are below reported comatose-fatal levels. © 2022 Published by Elsevier B.V. This is an open access article under the CC BY license (http://creative-

commons.org/licenses/by/4.0/).

1. Introduction

Interpretation of toxicological findings from autopsy material is inherently difficult as several factors potentially influence results. These factors include age, health status, drug-intaking history, cir- cumstances of death, drug redistribution etc. Still, quantitative re- sults from toxicological analyses are widely compared to ranges in existing reference works to determine if drugs are in relevant range for proving intoxication. Several fatal intoxication reference-works exist [1–5], however, this study aims to supplement current knowledge of the possible span and potential overlaps of recently measured concentrations of commonly used opioids and stimulants in own cases and controls and discuss the consequences hereof.

We present results from fatal intoxications of drug addicts and compare them to drug addicts whose death was not drug-related and results from individuals driving under the influence of drugs (DUID), knowing that these individuals were alive during sampling. A similar case-control course of action has previously been applied for establishing widely recognized postmortem reference values

[2,6–8]. We focused on the drugs heroin/morphine, methadone, cocaine, amphetamine and MDMA and discuss obtained concentra- tions in DUID controls compared to drug addict findings. Comparing post-mortem concentrations with DUID samples will not take post- mortem drug redistribution into account. Caution should be made when comparing post- and antemortem samples.

Compared to the rest of the Nordic countries, Denmark has a high number of fatal methadone intoxications and deaths from CNS sti- mulants like cocaine, amphetamine and MDMA is increasing across the Nordic countries thus making a study like this of relevance [9]. After methadone, heroin/morphine accounts for most fatal poi- sonings.

The same blood sampling system was used on both the autopsy and the control groups facilitating optimal basis for comparison without distortion by different blood preparation techniques such as plasma versus whole blood or serum or using different conservation additives which could impact results [10,11]. Furthermore, samples were all analyzed in the same laboratory and on the same type of analytical equipment.

The forensic data originates from police districts of Southern Denmark and Funen, covering approximately 1 million inhabitants resulting in approximately a thousand DUID cases and one hundred and fifty autopsies per year. The districts of Funen and Southern Jutland has the highest prevalence of heroin/morphine in DUID cases

https://doi.org/10.1016/j.forsciint.2022.111387 0379-0738/© 2022 Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

]]]] ]]]]]]

⁎ Corresponding author. E-mail addresses: [email protected] (S.K. Hermansen),

[email protected] (D.J. Christoffersen).

Forensic Science International 338 (2022) 111387

in the country and the second highest prevalence of methadone [12], thereby creating good basis for establishing representable con- centrations.

2. Material and methods

2.1. Sample material

The present study involved whole-blood results from forensic autopsies of drug addicts performed from 2015 to 2020 (n = 220) and DUID cases from 2015 to 2019 (n = 7088).

Included drug addict cases were deaths, where it has been con- cluded that the subjects at or before time of death were misusers of illegal drugs of abuse, i.e. drugs included in the Danish Ministry of Health announcement number 698 of 31. august 1993 or later revi- sions concerning illegal drugs of abuse (excluding cannabis and khat). The information of drug abuse might be obtained from police reports, autopsy reports, forensic toxicology reports, information from witnesses or from other institutions.

The cause of death was reviewed on a case-to-case basis and agreed upon by examining final forensic reports establishing cause of death and categorizing them for subsequent analysis. Drug addicts were grouped into fatal mono-intoxications (A), fatal combination- intoxications (mixed cases including ethanol) (B) and death by other causes than intoxication (C). Death by other causes could be hanging, drowning, stabbing etc. Cases involving pronounced putrefaction, use of different matrices (heart blood, liver blood, muscle tissue, urine etc.) or history of hospital treatment were excluded from the study. Average time between time of death and laboratory sample receival was 8 days (range 2–19 days). In cases where time of death remained unknown, the finding date of the body was used instead. DUID cases (D) consisted of routine cases found positive for one or more drugs above the Danish legal limit (limits listed in [13]). Group D can in some rare cases consist of injured drivers, but no post- mortem samples were in the group. To enable comparison between groups, DUID concentrations were in this case unadjusted for ana- lytical uncertainty (not minimum values). In Denmark, DUID cases are normally subtracted by a 33% safety margin to compensate for analytical uncertainty (reporting of minimum values).

2.2. Analytical methods

Samples were screened using liquid chromatography coupled with a quadropole time-of-flight mass spectrometer (LC-QTOF MS). The LC was a Dionex UltiMate 3000 Rapid Separation ultra-high- performance LC (UHPLC) (Thermo Fisher Scientific, MA, USA) with a VanGuard Pre-Column C18, 1.7 µm,2.1 × 5 mm (Waters Corporation, MA, USA) and an Acclaim RSLC 120 C18. 2.2 µm, 120 Å,2.1 × 100 mm column (Thermo Fisher Scientific, MA, USA). The MS was an Impact II or HD QTOF (Bruker, Billerica, MA, USA).

Screening was performed in both positive and negative ioniza- tion mode. For positive ionization mode mobile phases were H2O/ MeOH 9:1 (v/v) with 5 mM ammonium formate and 0,01% formic acid (mobile phase A) and MeOH with 5 mM ammonium formate and 0,01% formic acid (mobile phase B). For negative ionization mode mobile phases were H2O/MeOH 9:1 (v/v) with 5 mM ammo- nium acetate (mobile phase C) and MeOH with 5 mM ammonium acetate (mobile phase D). All reagents were LC-MS grade.

Quantitative analysis was performed using an LC coupled with a triple quadropole mass spectrometer (LC-MS/MS). The LC was a Dionex UltiMate 3000 Rapid Separation ultra-high-performance LC (UHPLC) (Thermo Fisher Scientific, MA, USA) with a VanGuard Pre- Column C18, 1.7 µm, 2.1 × 5 mm (Waters Corporation, MA, USA) and an Acquity UPLC BEH C18 Column, 130 Å, 1.7 µm, 2.1 mm × 50 mm (Waters Corporation, MA, USA). The MS was a TSQ Endura Triple

Quadropole Mass spectrometer (Thermo Fisher Scientific, MA, USA). Mobile phase A and B (defined above) was used.

The laboratory is accredited under DS/EN ISO/IEC 17025:2017 and participates in proficiency scheme LGC “Forensic Blood Toxicology” (QUARTZ).

2.3. Standards

Drug identification was confirmed and controlled using certified analytical standards from accredited suppliers. Mixed standard so- lutions were prepared and stored at − 20 °C prior to analysis. Deuterated internal standards (IS) were amphetamin-D11, bupre- norphin-D4, cocain-D3, diazepam-D5, LSD-D3, MDMA-D5, me- thadon-D9, morphin-D6 and pentobarbital-D5.

2.4. Sample preparation

Whole-blood was sampled by medical staff into Vacuette sodium fluoride and potassium oxalate (FX) tubes (Greiner Bio-One GmbH, Kremsmünster, Austria) and delivered to Section of Forensic Toxicology. For analysis, 0.25 g of whole-blood was aliquoted. Ten µl IS mix was added followed by protein precipitation carried out by adding 500 µl acetonitrile to each sample tube. Samples were then mixed using a whirl-mixer for 1 min and cold-centrifugated for 10 min at 15,000 rpm. Supernatant was extracted and diluted 1:1 by mixing 200 µl sample with 200 µl type 1 ultrapure water (Milli-Q® IQ 7000 Ultrapure Water System, Merck KGaA, Darmstadt, Germany).

2.5. Data processing and graphical representation

Statistical tests were performed using GraphPad Prism 9 for Windows (LLC Inc.). Groups did not pass normality tests; thus, nonparametric tests were used for groups comparisons with a 0.05 significance level. Nonparametric tests were Kruskal-Wallis test with Dunns’ test (uncorrected) for multiple comparison. Box and whisker plots were used for graphical representation of data. Boxes are 25th to 75th percentiles and the line in the middle represents the median. Whiskers are 10th to 90th percentiles but for groups where n < = 3 range was plotted. Obtained concentrations were compared to the widely recognized comatose-fatal (from) reference concentrations in plasma reported in [14]. In case of methadone, the comatose-fatal plasma concentration was converted to whole blood using a blood/ plasma (b/p) ratio of 0.7 [15]. In case of MDMA a b/p of 1.2 was used [15]. In Danish forensics, quantitative results for psychoactive sub- stances are reported in mg/kg instead of mg/L (or ng/mL) as they are based on weighed samples. Reported concentrations in mg/kg can be considered equivalent to mg/L under the assumption that the den- sity of the sampled blood is ≈ 1 g/mL.

3. Results

Total eligible cases are listed in Table 1. Morphine levels in the case and control groups were illustrated

in Fig. 1. Median morphine levels were significantly higher in both the fatal mono-intoxications and the fatal combination-intoxications

Table 1 Eligible cases in the groups.

Cases (n) Controls (n)

Drug Group A Group B Group C Group D Total

Morphine 12 48 8 192 260 Methadone 8 60 9 189 266 Cocaine 2 18 4 682 706 Amphetamine 2 8 6 1529 1545 MDMA 1 2 0 180 183

S.K. Hermansen and D.J. Christoffersen Forensic Science International 338 (2022) 111387

2

cases compared to control groups. The 10th percentile in group A and B corresponded to the recognized comatose-fatal lower limit listed in [14]. Notably, 39% of individuals in the DUID control group exhibited whole blood concentrations above the comatose-fatal lower limit with 5 individuals having extreme concentrations above 2.0 mg/kg.

In Fig. 1 methadone levels were significantly higher in the in- toxication groups A and B compared to DUID cases (group D). The 10th percentile in group A corresponded with the comatose-fatal lower limit as do median levels in DUID cases.

Cocaine in Fig. 1 showed no statistical difference between groups and overlaps between cases and controls were smaller than for

morphine and methadone. Very few fatal mono-intoxications with cocaine existed and the large majority of DUID cases had con- centrations below the comatose-fatal lower limit.

In the case of amphetamine (Fig. 1) few fatal intoxications existed but 101 (7%) of the DUID cases had concentrations in the comatose- fatal lower area.

MDMA in Fig. 1 few or none eligible cases existed in groups A, B and C, and comparison between groups should not be made. How- ever, more than 20% of the DUID cases had concentrations in the comatose-fatal area.

Fig. 1. 1; Morphine, 2; Methadone, 3; Cocaine, 4; Amphetamine, 5; MDMA (* p < 0.05; ** p < 0.01; *** p < 0.001; **** p < 0.0001). Red dotted lines indicate recognized comatose- fatal (from) reference concentrations in plasma [14]. Methadone and MDMA plasma concentrations were adjusted to whole blood concentrations using blood/plasma ratios from [15]. Gray area used when reference works indicated ranges.

S.K. Hermansen and D.J. Christoffersen Forensic Science International 338 (2022) 111387

3

4. Discussion

4.1. DUIDs are often above the lower limit of fatal intoxication

The present study focused on heroin/morphine, methadone, co- caine, amphetamine and MDMA concentrations in drugs addicts and DUID cases. These drugs are commonly encountered in both fatal intoxications and DUID cases and the potential for abuse is well known. DUID cases did in many instances exhibit whole blood concentrations well above the established lower limit of fatal in- toxication. A considerable overlap was to be expected due to toler- ance of especially the opioids. However, in case of heroin/morphine, we did not expect to see 39% of cases above the reported comatose- fatal lower limit. Furthermore, the examples above 2.0 mg/kg are considerable above concentrations seen in most of our fatal mono- intoxications. It should be noted that we do not know the exact circumstances of these cases, but individuals in the DUID group were involved in driving a vehicle at time of arrest without immediate subsequent autopsy. The 5 individuals in the DUID group having extreme heroin/morphine concentrations well above 2.0 mg/kg are all suspected to be tolerant users due to concomitant methadone and benzodiazepines in the blood samples. Methadone is an in- dicator of active treatment for opioid dependence and the combi- nation with benzodiazepines is commonly seen in drug addicts. Also, a single individual is represented in two of the extreme cases, thus displaying repeat offenses. Even though median levels are still sig- nificantly different between cases and controls these findings un- derline the importance of case history when evaluating fatal intoxications among drug addicts and tolerant individuals. A case compilation has listed large morphine concentrations in surgical patients from 0.8 to 2.6 mg/L; all patients required assisted venti- lation when subjected to these concentrations [15]. The same re- ference work listed concentrations from 0.2 to 2.3 mg/L in fatalities. Here we report that these ranges are also possible in living in- dividuals driving under the influence, underlining that it is practi- cally impossible to evaluate cases on the heroin/morphine concentrations alone. Similar observations were made for metha- done and amphetamine. For methadone, DUID controls grouped evenly around the lower comatose-fatal limit from [14], meaning that about half the individuals were driving with comatose-fatal levels. Amphetamine is often compared to cocaine in terms of ef- fects, user groups and side effects. In our data, amphetamine in DUID controls had a markedly bigger cross-over into comatose-fatal levels than cocaine. This was also true for MDMA.

The DUID group would in all probability consist of a mixture of both naïve and tolerant users, pain patients, drug addicts and re- present various routes of administration (nasal, inhalation, in- travenous etc.) making it useful for distribution analysis – especially towards the higher end of the scale, due to the high number of cases. It should be noted that only cases above legal limits are represented which is expected to shift median levels slightly upward compared to all apprehended drivers. Morphine and methadone 10th percen- tiles of fatal mono-intoxications corresponds to the lower comatose- fatal limit. This is expected when stringently applying reference concentrations for assessing intoxication. Thus, these data should not be used for corroborating existing limits, as they themselves are based hereupon.

In case of morphine, the lower comatose-fatal limit reported in [14] seem to be based on a single observation in [16]. The case is an intravenous multiple drug overdose of a 45-year-old woman. Others have put ranges for morphine fatalities from 0.2 – 2.3 mg/L and 0.3–0.43 mg/L for heroin fatalities [15]. Thus, the 0.1 mg/L lower comatose fatal limit is probably an extreme low, where intoxication potentially was aided by combining other drugs.

To summarize, in the DUID group, the opioids heroin/morphine and methadone and the stimulants amphetamine and MDMA were

often seen in concentrations above the reported lower comatose- fatal levels whereas cocaine was almost always below. It is not the study’s objective to question the limits, merely to supplement cur- rent knowledge about the possible span and potential overlaps of measured concentrations together with a data-based evaluation of which drugs are most prone to false assessment regarding fatal in- toxications. Based on our data, the potential for false assessment of intoxication cases when comparing to reported comatose-fatal limits appear greatest on lower end concentrations of heroin/mor- phine, methadone, amphetamine and MDMA, whereas false as- sessment of cocaine appears less likely because almost all control cases are below reported comatose-fatal levels.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

CRediT authorship contribution statement

Simon Kjær Hermansen: Conceptualization, Methodology, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Visualization. Dorte Jensen Christoffersen: Conceptualization, Data curation, Writing – review & editing, Supervision.

Acknowledgments

A special thanks goes to Tina Ravn Pedersen and Lone Lindal for proof reading the manuscript and our laboratory personnel for car- rying out the analyzes.

Declarations of interest

None.

References

[1] H.E. Edvardsen, T. Tverborgvik, J. Frost, S. Rogde, I. Morild, H. Waal, et al., Differences in combinations and concentrations of drugs of abuse in fatal in- toxication and driving under the influence cases, Forensic Sci. Int. 281 (2017) 127–133, https://doi.org/10.1016/j.forsciint.2017.10.045

[2] A.K. Jönsson, C. Söderberg, K.A. Espnes, J. Ahlner, A. Eriksson, M. Reis, et al., Sedative and hypnotic drugs–fatal and non-fatal reference blood concentrations, Forensic Sci. Int. 236 (2014) 138–145, https://doi.org/10.1016/j.forsciint.2014.01. 005

[3] T. Launiainen, I. Ojanperä, Drug concentrations in post-mortem femoral blood compared with therapeutic concentrations in plasma, Drug Test. Anal. 6 (4) (2014) 308–316, https://doi.org/10.1002/dta.1507

[4] A.B. Bech, T. Clausen, H. Waal, V. Vindenes, H.E. Edvardsen, J. Frost, et al., Post- mortem toxicological analyses of blood samples from 107 patients receiving opioid agonist treatment: substances detected and pooled opioid and benzo- diazepine concentrations, Addiction 116 (2020) 845–855, https://doi.org/10. 1111/add.15211

[5] R.A. Ketola, I. Ojanperä, Summary statistics for drug concentrations in post- mortem femoral blood representing all causes of death, Drug Test. Anal. 11 (9) (2019) 1326–1337, https://doi.org/10.1002/dta.2655

[6] C. Söderberg, A. Tillmar, A. Johansson, E. Wernvik, A.K. Jönsson, H. Druid, The importance of sample size with regard to the robustness of postmortem re- ference values, Forensic Sci. Int. 311 (2020) 110292, https://doi.org/10.1016/j. forsciint.2020.110292

[7] C. Söderberg, E. Wernvik, A. Tillmar, O. Spigset, R. Kronstrand, M. Reis, et al., Antipsychotics - postmortem fatal and non-fatal reference concentrations, Forensic Sci. Int. 266 (2016) 91–101, https://doi.org/10.1016/j.forsciint.2016.05. 002

[8] H. Druid, P. Holmgren, A compilation of fatal and control concentrations of drugs in postmortem femoral blood, J. Forensic Sci. 42 (1) (1997) 79–87, https://doi. org/10.1520/JFS14071J

[9] K.W. Simonsen, P. Kriikku, G. Thelander, H.M.E. Edvardsen, S. Thordardottir, C.U. Andersen, et al., Fatal poisoning in drug addicts in the Nordic countries in 2017, Forensic Sci. Int. 313 (2020) 110343, https://doi.org/10.1016/j.forsciint. 2020.110343

[10] D. Montenarh, M.P. Wernet, M. Hopf, H.H. Maurer, P.H. Schmidt, A.H. Ewald, Quantification of 33 antidepressants by LC-MS/MS—comparative validation in

S.K. Hermansen and D.J. Christoffersen Forensic Science International 338 (2022) 111387

4

whole blood, plasma, and serum, Anal. Bioanal. Chem. 406 (24) (2014) 5939–5953, https://doi.org/10.1007/s00216-014-8019-x

[11] C. Wiedfeld, J. Krueger, G. Skopp, F. Musshoff, Comparison of concentrations of drugs between blood samples with and without fluoride additive—important findings for Δ9-tetrahydrocannabinol and amphetamine, Int. J. Leg. Med. 133 (1) (2019) 109–116, https://doi.org/10.1007/s00414-018-1797-5

[12] K.W. Simonsen, J.B. Hasselstrøm, S.K. Hermansen, B.S. Rasmussen, M.F. Andreasen, D.J. Christoffersen, et al., The incidence of psychoactive sub- stances and alcohol among impaired drivers in Denmark in 2015–2019, Forensic Sci. Int. 333 (2022) 111207, https://doi.org/10.1016/j.forsciint.2022.111207

[13] Bekendtgørelse om ændring af bekendtgørelse om klassificering af færdselssikkerhedsfarlige stoffer (retsinformation.dk), 2011. ⟨https://www. retsinformation.dk/eli/lta/2011/901⟩ (in Danish).

[14] M. Schulz, A. Schmoldt, H. Andresen-Streichert, S. Iwersen-Bergmann, Revisited: therapeutic and toxic blood concentrations of more than 1100 drugs and other xenobiotics, Crit. Care 24 (1) (2020) 195, https://doi.org/10.1186/s13054-020- 02915-5

[15] R.C. Baselt, Disposition of Toxic Drugs and Chemicals in Man, eleventh ed., Biomedical Publications, 2017.

[16] F. Musshoff, S. Padosch, S. Steinborn, B. Madea, Fatal blood and tissue con- centrations of more than 200 drugs, Forensic Sci. Int. 142 (2–3) (2004) 161–210, https://doi.org/10.1016/j.forsciint.2004.02.017

S.K. Hermansen and D.J. Christoffersen Forensic Science International 338 (2022) 111387

5