- Email: [email protected]
Toxicological analysis in rats subjected to heroin and morphine overdose
Toxicology Letters 166 (2006) 11–18
Toxicological analysis in rats subjected to heroin and morphine overdose Joakim J. Strandberg a , Fredrik C. Kugelberg a , Kanar Alkass a , Anna Gustavsson a , Kolbjørn Zahlsen b , Olav Spigset b,c , Henrik Druid a,∗ a
Department of Forensic Medicine, Karolinska Institutet, SE-171 77 Stockholm, Sweden Department of Clinical Pharmacology, St. Olav University Hospital, NO-7006 Trondheim, Norway Department of Laboratory Medicine, Children’s and Women’s Disease, Norwegian University of Science and Technology, NO-7006 Trondheim, Norway b
c
Received 17 February 2006; received in revised form 10 May 2006; accepted 11 May 2006 Available online 16 May 2006
Abstract In heroin overdose deaths the blood morphine concentration varies substantially. To explore possible pharmacokinetic explanations for variable sensitivity to opiate toxicity we studied mortality and drug concentrations in male Sprague-Dawley rats. Groups of rats were injected intravenously (i.v.) with heroin, 21.5 mg/kg, or morphine, 223 mg/kg, causing a 60–80% mortality among drug-na¨ıve rats. Additional groups of rats were pre-treated with morphine for 14 days, with or without 1 week of subsequent abstinence. Brain, lung and blood samples were analyzed for 6-acetylmorphine, morphine, morphine-3-glucuronide and morphine-6-glucuronide. i.v. morphine administration to drug-na¨ıve rats resulted in both rapid and delayed deaths. The brain morphine concentration conformed to an exponential elimination curve in all samples, ruling out accumulation of morphine as an explanation for delayed deaths. This study found no support for formation of toxic concentration of morphine-6-glucuronide. Spontaneous death among both heroin and morphine rats occurred at fairly uniform brain morphine concentrations. Morphine pre-treatment significantly reduced mortality upon i.v. morphine injection, but the protective effect was less evident upon i.v. heroin challenge. The morphine pre-treatment still afforded some protection after 1 week of abstinence among rats receiving i.v. morphine, whereas rats given i.v. heroin showed similar death rate as drug-na¨ıve rats. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Glucuronide; Heroin; Opiate; Overdose; Pharmacokinetics; Tolerance
1. Introduction About 8000 acute drug-related deaths are recorded annually in the EU (http://www.emcdda.eu.int/). An evaluation of the reported deaths during several years
∗ Corresponding author. Tel.: +46 8 52 48 77 70; fax: +46 8 32 14 52. E-mail address: [email protected] (H. Druid).
in 10 countries shows that approximately 80% of these deaths are related to opiate overdose. Typically, such opiate toxicity deaths strike suddenly and unexpectedly even among experienced users. Postmortem toxicological findings in deceased heroin addicts do not support that heroin “overdose” deaths result from the intake of a large dose. In a majority of these victims the blood morphine concentration is actually found to be lower than that seen in heroin users that did not die from overdose but from other causes (Darke
0378-4274/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2006.05.007
12
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
and Zador, 1996). Further, blood morphine concentration varies substantially, and cannot be used in isolation to diagnose heroin overdose deaths (Meissner et al., 2002). By performing segmental hair analysis, we have recently found that abstinent overdose victims presented with similar blood morphine concentration as did overdose victims with evidence of continuous opioid drug use prior to death (Druid et al., 2006). In humans, heroin (3,6-diacetylmorphine) is rapidly degraded to the metabolites 6-acetylmorphine (6-AM) and morphine. Most of the morphine is further converted to morphine-3-glucuronide (M3G; ∼50%) and morphine-6-glucuronide (M6G; ∼10%) (Aderjan and Skopp, 1998). Heroin quickly enters the brain through the blood–brain barrier, however, due to its rapid metabolism, it is generally assumed that heroin itself has only minor pharmacological effects. Most of the pharmacological effects, including respiratory depression, are instead caused by morphine, or to some extent, by 6-AM and M6G (White and Irvine, 1999). Since the demonstration in the late 1960s that M6G possesses analgesic properties (Kamata et al., 1969) it has been proposed that the glucuronides play an important role for the toxicity of heroin and morphine. However, the main metabolite, M3G, does not seem to have any agonistic effect either in vivo or in cell cultures (Shimomura et al., 1971; Hemstapat et al., 2003). In contrast, there is some evidence that M3G may exert antagonistic effects and to some extent compete with morphine and M6G at binding sites (Gardmark et al., 1998). When M6G is directly administered to humans, some studies have shown that side effects like nausea, vomiting and respiratory depression are attenuated compared with the administration of analgesically equipotent doses of morphine, whereas others have found no such differences (L¨otsch, 2005). In rats, the formation of M6G is limited (Milne et al., 1996); hence, heroin and morphine administration to rats is not expected to produce toxic concentration of this metabolite. Therefore, rat studies offer good conditions for the evaluation of the inherent toxicity of morphine and heroin. Since blood morphine concentration in apparent heroin overdose deaths vary substantially, we decided to explore the possible correlation between the concentrations of several heroin metabolites and the outcome in an animal model, where rats were given high doses of either morphine or heroin. We hypothesized that a gradual accumulation of either morphine or M6G in the brain together with a drop in blood morphine concentration might explain low blood morphine concentration at death. Further, we also wanted to address the question of whether accumulation of opiate metabolites in
the brain could account for delayed deaths. In order to assess the impact of tolerance, we included groups of rats that were pre-treated with intraperitoneal (i.p.) injections of morphine, and then challenged them with an acute intravenous (i.v.) injection of heroin or morphine, with or without a preceding period of abstinence. 2. Material and methods 2.1. Animals Male Sprague-Dawley rats (Scanbur BK AB, Sollentuna, Sweden) initially weighing 216–376 g were used. Rats were allowed to acclimatize for 1 week prior to experiment. Rats were allocated in groups of four in each cage and allowed fresh water and standard rat chow ad libitum throughout the whole experimental period. Rats were kept in a 12:12 h light:dark cycle synchronous with daylight (lights on at 7.00 a.m.). All experiments were performed in strict accordance with the guidelines and the consent of the Animal Ethics Committee of Northern Stockholm (Permit no. 187/03, 413/03 and 110/05). 2.2. Drugs Heroin hydrochloride was obtained from Macfarlan Smith Ltd., Edinburgh, UK and morphine hydrochloride was purchased from the Swedish Pharmacy (Apoteket AB). Both drugs were dissolved in 0.9% NaCl and slightly heated (40–50 ◦ C) when needed to be dissolved. 2.3. Experimental procedure The experimental design of the study is illustrated in Fig. 1. Rats were randomized to seven groups, each comprising 6–14 rats treated as follows: Group 1 (a and b) i.v. injection of (a) heroin, (b) morphine; Group 2 (a and b) i.p. injections of morphine for 14 days followed by an i.v. injection of (a) heroin or (b) morphine; Group 3 (a and b) i.p. injections of morphine for 14 days followed by 1 week without any injections (i.e. abstinence) and then an i.v. injection of (a) heroin, (b) morphine; Group 4 (b) i.p. injections of (b) morphine for 14 days. i.p. injections with morphine were given once daily between 10:00 a.m. and 2:00 p.m., with a 0.6 mm × 25 mm needle for 14 consecutive days using the following dosage schedule: 20 mg/kg (days 1–3), 40 mg/kg (days 4–6), 60 mg/kg (days 7–9), 80 mg/kg (days 10–12) and 100 mg/kg (days 13–14). The volumes administered ranged from a minimum of 0.31 mL to a maximum of 0.56 mL. The up-and-down method (Dixon, 1965; Bruce, 1985) was applied to establish a dose causing 60–80% mortality. This dose was found to be 21.5 mg/kg for heroin and 223 mg/kg for morphine. Rats were given i.v. injections of these doses of morphine or heroin in the tail vein (volumes ranging from 0.42 to 0.46 mL). All i.v. injections were given in the morning (8.00–10.00 a.m.). Immediately after injection, rats were placed in separate cages and observed for a maximum of 8 h. Rats that died spontaneously during this period
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
13
Fig. 1. Experimental design. For details, see text.
were decapitated, whereas rats that survived were quickly anesthetized with isoflurane (Forene® ; Abbott Scandinavia AB, Solna, Sweden) and decapitated 8 h after the i.v injection. After decapitation, blood was drawn from the heart and mixed with sodium fluoride to prevent degradation of drugs. The brain was quickly removed and the occipital lobe and the cervical spinal cord were dissected free and collected for toxicological analysis. The lungs were removed and the right lung was kept for toxicological analysis. The blood, brain and lung samples were stored at −80 ◦ C until drug analysis. 2.4. Analysis of 6-AM, morphine, M3G and M6G Analysis of 6-AM, morphine, M3G and M6G in blood and tissues was performed by electrospray liquid chromatography–mass spectrometry (ESI-LC–MS). In brief, prior to extraction, tissue samples from occipital lobe, cervical spinal cord and lung were excised and weighed individually. A weighed amount ranging from 15 to 79 mg (mean 52 mg) was taken from the occipital lobe, 12–79 mg (mean 58 mg) from the cervical spinal cord and 93–110 mg (mean 102 mg) from the lung. The tissue samples were homogenized with an Ultra-Turrax homogenizer together with 1 mL of ammonium carbonate buffer containing internal standards. Blood samples (weighing 56–98 mg; mean 71 mg) were mixed with 1 mL of ammonium carbonate buffer. The homogenate and diluted blood samples were extracted according to Bogusz et al. (1997) with slight modifications, by solid phase extraction (SPE) using Chromabond C18 SPE extraction columns (Machery-Nagel GmbH). The columns
were preconditioned with 1 mL of methanol, then 1 mL of ultrapure water and finally 2 mL of ammonium carbonate buffer. The homogenized/diluted sample was loaded on column and aspirated by vacuum. The column was washed with 3 mL of ammonium carbonate buffer and then dried for 5 min. The analytes were eluted in a volume of 0.5 mL of methanol and acetic acid (9:1), evaporated and reconstituted in 50 L of ammonium acetate. The eluates were then transferred to autosampler vials for LC–MS analysis. The concentrations of analytes were determined using an Agilent 1100 mass spectrometric system consisting of a G1313A autosampler, a G1322A degassing unit, a G1311A quaternary pump, a G1316A column oven and a single quadropole G1946D mass spectrometer. The analytes were monitored in positive mode SIM at the following m/z values: Morphine 286.1, 6-AM 328.1, M3G 462.1 and M6G 462.1. Morphine-D3 (monitored at m/z 289.1) was used as an internal standard for morphine, M3G and M6G, whereas 6-AM-D3 (monitored at m/z 331.1) was used as an internal standard for 6-AM. Calibration curves were established to cover a range of concentrations from 0.1 up to 1000 g/g. Limits of quantitation for morphine and 6-AM were 0.2 g/g and for M3G and M6G 0.1 g/g with a sample size of 50 mg. The extraction recoveries ranged from 40 to 84%. 2.5. Statistics A value of p < 0.05 was considered statistically significant. Possible differences between the groups were analyzed by
14
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
Table 1 Mortality after i.v. injection of morphine or heroin, with or without morphine pre-treatment Treatment
Group 1 (i.v.) Group 2 (i.p.–i.v.) Group 3 (i.p.–abst–i.v.)
Heroin 21.5 mg/kg
Morphine 223 mg/kg
Death within 10 min
Delayed death
Sacrificed at 480 min
Time to death (min)a
Death within 10 min
Delayed death
Sacrificed at 480 min
Time to death (min)a
8 (67%) 3 (23%) 8 (62%)
0 0 0
4 (33%) 10 (77%) 5 (38%)
163 ± 67.6 370 ± 58.3 187 ± 66.9
7 (50%) 0 2 (25%)
4 (29%) 0 0
3 (21%) 7 (100%) 6 (75%)
151 ± 52.8 480 ± 0 362 ± 77.6
i.v., intravenous; i.p., intraperitoneal injections of morphine for 2 weeks; abst, 1 week of abstinence after i.p. pre-treatment. Delayed death, spontaneous death between 10 and 480 min. Rats sacrificed at 480 min were included when calculating mean time to death. a Values are mean ± S.E.M.
one-factor analysis of variance (ANOVA). When the ANOVA reached statistical significance, Fisher’s protected least significant difference (PLSD) post hoc test was applied. All statistical analyses were performed using StatView® for Windows Version 5.0 (SAS® Institute Inc., Cary, NC, USA).
3. Results i.v. doses causing approximately 70% mortality among drug-na¨ıve rats was established prior to this study (21.5 mg/kg for heroin and 223 mg/kg for morphine). In the present study these doses produced 67% mortality among drug-na¨ıve rats given i.v. heroin and 79% mortality among drug-na¨ıve rats given i.v. morphine (Table 1). In comparison, animals receiving 2 weeks of pre-treatment with i.p. morphine showed a lower mortality after i.v. heroin as well as after i.v. morphine injection. Even after 1 week of abstinence following pretreatment, the mortality was still lower than in drug-na¨ıve rats; 62% after i.v. heroin and 25% after i.v. morphine. Hence, a significant difference was found in time to death between drug-na¨ıve i.v. heroin and pre-treated i.v. heroin rats (p = 0.016); between drug-na¨ıve i.v. morphine and pre-treated i.v. morphine rats (p = 0.001), and between drug-na¨ıve i.v. morphine and pre-treated rats that were subjected to 1 week of abstinence before i.v. morphine injection (p = 0.027). The morphine concentration in all tissues varied substantially among rats that spontaneously died within 10 min after i.v. injection. Rats receiving i.v. morphine (Group 1b) had much higher concentration of morphine in blood and lung tissue than those receiving heroin (Group 1a), whereas the concentration of morphine in cervical spinal cord were similar in both groups (Table 2). In samples from the occipital lobe, i.v. heroin rats had significantly higher morphine concentration compared with i.v. morphine rats (p < 0.05). Further, i.v. heroin rats in all subgroups had higher concentration of morphine in brain and lung tissue than in blood (p-values
ranging from p < 0.01 to 0.0001). Also, the concentration of 6-AM in i.v. heroin rats was higher in brain tissue compared to concentration in blood (p-values ranging from p < 0.05 to 0.0001). However, 6-AM concentration was consistently lowest in lung tissue. The time to death for the delayed deaths (defined as death between 10 and 480 min) in Group 1b ranged from 25 to 262 min (mean 162 ± 94 min, n = 4). These rats usually showed froth in the airways as well as forced and irregular breathing. No accumulation of drug could be detected in these rats, and the drug concentrations followed an exponential elimination curve. No delayed deaths occurred among i.v. heroin rats. These rats either died quickly, within 10 min, or survived (but then showed blood-tinged froth in the mouth, and had a strained breathing until sacrificed 8 h after injection). Rats that were sacrificed 8 h after i.v. injection generally showed low concentrations of all opiate compounds (Table 2). The morphine concentration was similar across all different types of tissues, but was typically higher in lung tissue than in other tissues. When comparing the time of death with the blood and tissue concentrations of the opiate compounds, we found that the concentrations of all analytes conformed to a predicted exponential elimination (Fig. 2). 4. Discussion We developed a rat model to mimic morphine and heroin overdose death in humans. Our major findings were: (1) death occurred at different times after i.v. injection, but there was no evidence of accumulation of morphine or other opiate compounds in the brain to explain for delayed deaths; (2) pre-treatment with morphine significantly reduced mortality after i.v. morphine injection, and some protective effect was still observed after 1 week of abstinence; however, pre-treatment with morphine did not offer such protection against i.v. heroin injection;
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
15
Table 2 Drug and metabolite concentrations (g/g) following administration of heroin and morphine to rats
Group 1a: heroin (i.v.) Dead within 10 min (n = 8) 6-MAM Morphine M3G Sacrificed at 480 min (n = 4) 6-MAM Morphine M3G Group 1b: morphine (i.v.) Dead within 10 min (n = 7) Morphine M3G
Blood
Lung
Cervical spinal cord
Occipital lobe
17.5 ± 8.05 4.86 ± 1.88 0.15 (n = 2)
7.91 ± 2.38 43.5 ± 23.6 n.d.
39.2 ± 8.6 27.3 ± 5.24 n.d.
64.6 ± 11.8 59.4 ± 13.9 n.d.
n.d. n.d. 0.14 (n = 2)
n.d. 0.41 ± 0.26 0.12 ± 0.09
0.32 (n = 2) 0.23 (n = 1) n.d.
0.14 (n = 1) 0.16 (n = 1) n.d.
304 ± 234 4.41 ± 4.60
580 ± 295 n.a.
37.4 ± 27.8 n.a.
33.1 ± 9.92 n.a.
Delayed deaths (n = 4)
Death between 10 and 480 min. For details, see text and Fig. 2
Sacrificed at 480 min (n = 3) Morphine M3G
0.40 (n = 2) 6.29 (n = 2)
1.93 ± 0.79 n.a.
0.26 (n = 1) n.a.
0.40 (n = 2) n.a.
25.4 ± 2.6 8.91 ± 0.82 1.62 ± 1.51
16.8 ± 1.73 94.0 ± 14 1.69 ± 1.2
34.6 ± 2.14 27.0 ± 2.73 0.16 ± 0.06
67.8 ± 11.0 70.8 ± 7.94 0.11 (n = 2)
n.d. 0.21 ± 0.05 0.43 ± 0.13
0.26 (n = 1) 1.74 ± 0.9 0.43 ± 0.11
0.22 ± 0.06 0.19 (n = 2) 0.09 (n = 1)
0.27 ± 0.15 0.24 ± 0.07 n.d.
– –
– –
– –
– –
Sacrificed at 480 min (n = 7) Morphine M3G
1.25 ± 0.34 2.70 ± 2.12
5.82 ± 1.88 n.a.
0.75 (n = 2) n.a.
0.72 ± 0.24 n.a.
Group 3a: heroin (i.p.–abst–i.v.) Dead within 10 min (n = 8) 6-MAM Morphine M3G
18.3 ± 7.45 6.26 ± 2.52 n.d.
6.20 ± 2.04 65.6 ± 41.6 0.05 ± 0.01
28.0 ± 7.58 34.8 ± 14.5 n.d.
50.5 ± 12.4 79.7 ± 28.3 n.d.
Sacrificed at 480 min (n = 5) 6-MAM Morphine M3G
0.18 (n = 1) 0.19 (n = 1) 0.11 ± 0.02
n.d. 0.78 ± 0.26 0.10 ± 0.03
0.18 ± 0.06 n.d. n.d.
0.18 (n = 2) 0.22 (n = 2) n.d.
243 (n = 2) 1.89 (n = 2)
428 (n = 2) n.a.
33.5 (n = 2) n.a.
30.2 (n = 2) n.a.
0.74 ± 0.48 2.48 ± 1.81
2.74 ± 1.15 n.a.
0.39 (n = 1) n.a.
0.27 (n = 1) n.a.
Group 2a: heroin (i.p.–i.v.) Dead within 10 min (n = 3) 6-MAM Morphine M3G Sacrificed at 480 min (n = 10) 6-MAM Morphine M3G Group 2b: morphine (i.p.–i.v.) Dead within 10 min (n = 0) Morphine M3G
Group 3b: morphine (i.p.–abst–i.v.) Dead within 10 min (n = 2) Morphine M3G Sacrificed at 480 min (n = 6) Morphine M3G
16
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
Table 2 (Continued) Blood Group 4: morphine (i.p.) Sacrificed after last i.p. (n = 6) Morphine 0.11 ± 0.02 M3G 0.54 ± 0.08
Lung
Cervical spinal cord
Occipital lobe
0.51 ± 0.17 n.a.
n.d. n.a.
n.d. n.a.
Values are mean ± S.D. n.d., not detectable (below the limit of detection); n.a., not analysed; i.v., intravenous injection of either heroin 21.5 mg/kg or morphine 223 mg/kg; i.p., intraperitoneal injections of morphine for 2 weeks; abst, 1 week of abstinence after i.p. pre-treatment; 6-MAM, 6-monoacetylmorphine; M3G, morphine-3-glucuronide; M6G, morphine-6-glucuronide.
(3) the occipital lobe generally had higher opiate (6AM and morphine) concentrations than other tissues in rapid deaths after i.v. heroin injection, but were similar to cervical spinal cord concentrations after i.v. morphine injection. We aimed to use an i.v. dose that could produce spontaneous deaths among both drug-na¨ıve and morphine pre-treated rats, allowing for the evaluation of toxicological results in rats that died spontaneously as well as in rats sacrificed at defined times after drug administration. We found that heroin 21.5 mg/kg and morphine 223 mg/kg caused approximately 60–80% mortality in
na¨ıve Sprague-Dawley rats. Jackson (1952) studied the mortality after i.v. administration of heroin and morphine to na¨ıve Wistar rats and found the dose causing 50% mortality (LD50) to be 22.5 mg/kg for heroin. However, due to problems with solubility, the LD50 of morphine was concluded to “appear to” be 140 mg/kg. Siegel et al. (1982) found that i.v. administration of heroin hydrochloride 15 mg/kg to na¨ıve Wistar rats caused a mortality of 96%. Borron et al. (2002) injected morphine sulphate i.v. to Sprague-Dawley rats producing LD50 of 64 mg/kg. The difference in lethal dose in our study compared to LDs in other studies is most likely due to use
Fig. 2. Morphine concentrations in (a) blood, (b) occipital lobe, (c) cervical spinal cord and (d) lung, in rats given i.v. injections of morphine without pre-treatment (Group 1b). Correlation coefficients between morphine concentration and time to death were r = 0.566 (p = 0.0696) in blood, r = 0.799 (p = 0.0031) in occipital lobe, r = 0.435 (p = 0.2095) in cervical spinal cord and r = 0.680 (p = 0.0212) in lung.
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
of different kind of rat strains (Sprague-Dawley vs Wistar) and to different drug preparations (hydrochloride vs sulphate). The observation of delayed death among approximately 20–50% of the human heroin addicts (Nakamura, 1978; Manning et al., 1983; Darke et al., 2000) also applies to rats. Borron et al. (2002) showed that the median time to death was approximately 2.5 h (range 0.8–24 h) after i.v administration of morphine sulphate 64 mg/kg (LD50) to Sprague-Dawley rats. The time to death increased dramatically when rats were pre-treated with flunitrazepam, producing a delayed death about 13.5 h after the morphine injection. In our study we observed delayed deaths among opiate-na¨ıve rats given i.v. morphine, whereas no delayed deaths were seen among i.v. heroin rats. The reason for these delayed deaths remains unclear, but they could not be explained by pharmacokinetic mechanisms, since there was no evidence of accumulation of drugs in the brain or other tissues. Surprisingly, rats given i.v. heroin either died rapidly or survived the whole surveillance period. The absence of delayed death among the heroin injected rats might be explained by a too short observation time (8 h). Surviving rats generally showed irregular and decreased breathing and slightly blood-tinged froth in the airways, so it is possible that some of them might have died later if the surveillance period had been longer. It is also possible that a slightly lower dose might have produced some delayed deaths. In humans, abstinence, e.g. due to imprisonment or enrolment in treatment programs, is one factor that is frequently claimed to increase the risk of opiate overdose, even by drug addicts themselves (Bennett and Higgins, 1999). However, interviewed drug addicts rank abstinence as less important than the intake of an excessive amount of drug or mixing of different drugs (Bennett and Higgins, 1999). Siegel et al. (1982) investigated the impact of tolerance in an overdose paradigm using Wistar rats. Animals were pre-treated with increasing doses of heroin (i.v.) and thereafter subjected to an i.v. dose of heroin, killing approximately 96% of the na¨ıve rats and only 32% of the tolerant rats. In our study, pre-treatment for 14 days with morphine i.p. reduced the mortality from 79 to 0% after administration of i.v. morphine. This protective effect of morphine pre-treatment was less evident in rats given i.v. heroin; mortality was reduced from 67 to 23%. Still, after 1 week of abstinence, the mortality following i.v. morphine was reduced, whereas rats given i.v. heroin died at the same rate as did opiate-na¨ıve rats. The reason for this difference in impact of pre-treatment on the outcome after morphine and heroin administration remains unsolved, but one possibility is that it is
17
related to the more rapid delivery of morphine to key areas in the brain after heroin injection. If mu-opioid receptors in the respiratory centers are exposed to a very rapid and high concentration of morphine, there might not be enough time for the acute adaptations to come into action, regardless of long-term adaptations in the cells. Another possibility is that 6-AM contributes to the toxicity via signaling not affected by morphine pre-treatment. In the present study, we also found differences in the incorporation of drugs and metabolites into specific tissues. The occipital lobe generally had a higher concentration of 6-AM and morphine compared to the cervical spinal cord after i.v. heroin. Bullock et al. (1977) examined the regional distribution of morphine and its metabolites (not specified) after subcutaneous injection with radioactively labeled morphine. Morphine concentration in na¨ıve rats was found to be significantly higher in cerebellum compared to midbrain, hypothalamus, cortex and striatum. The morphine concentration was also slightly higher in cerebellum than in medulla oblongata. Further, Bullock et al. (1977) demonstrated that the distribution of morphine uptake was similar between na¨ıve and morphine-dependent rats, but the amount of morphine uptake was significantly decreased in dependent rats. The differences in tissue disposition might have several explanations, such as variations in fat content or in the density of opioid receptors. In very rapid deaths, differences in regional blood flow might also influence the concentration in various brain regions. Several groups have shown that M6G have greater antinociceptive effect compared to morphine when administered directly to rats (Abbott and Palmour, 1988; Sullivan et al., 1989). In rats, the formation of M6G is limited. Antonilli et al. (2003) showed that chronic heroin administration, but not chronic morphine administration resulted in low concentration of M6G. In the present study, we observed a minimal amount of M6G formation in blood (0.07–0.24 g/g) in a few rats after a very high dose of i.v. morphine (223 mg/kg) but not after i.v. heroin injection. Since the formation of M6G was very petite, it seems unlikely that this metabolite contributed to any significant degree to the mortality in this study. The study design aimed at mimicking human opiate overdose. There are certain species differences regarding metabolism rate, and metabolic pathways between humans and rodents that may render the results not completely applicable to humans. However, the observation of respiratory arrest (before cardiac arrest) and the occurrence of delayed deaths lend support to such applicability. Since toxicokinetic studies are ethically impossible
18
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
to perform on human subjects, we think that the results in this study contribute to the understanding of the toxic effects that opioids confer to human overdose victims. In conclusion, the results of the present study corroborate the impact of tolerance in acute opioid toxicity, although pre-treatment with morphine was more protective in rats administered i.v. morphine than in rats given i.v. heroin injection. The reason for the delayed deaths observed was seemingly not related to pharmacokinetic factors and warrants further investigation. Acknowledgements We thank Wenche Rødseth Brede, Department of Clinical Pharmacology, St. Olav University Hospital, Trondheim, Norway, for technical assistance with the drug analysis. This study was funded by grants from the Swedish National Drug Policy Coordinator, the Swedish National Board of Forensic Medicine, the Swedish Medical Society, Lennanders Foundation and the Magn. Bergvall Foundation. References Abbott, F.V., Palmour, R.M., 1988. Morphine-6-glucuronide: analgesic effects and receptor binding profile in rats. Life Sci. 43, 1685–1695. Aderjan, R.E., Skopp, G., 1998. Formation and clearance of active and inactive metabolites of opiates in humans. Ther. Drug Monit. 20, 561–569. Antonilli, L., Suriano, C., Paolone, G., Badiani, A., Nencini, P., 2003. Repeated exposures to heroin and/or cadmium alter the rate of formation of morphine glucuronides in the rat. J. Pharmacol. Exp. Ther. 307, 651–660. Bennett, G.A., Higgins, D.S., 1999. Accidental overdose among injecting drug users in Dorset, UK. Addiction 94, 1179–1189. Bogusz, M.J., Maier, R.D., Erkens, M., Driessen, S., 1997. Determination of morphine and its 3- and 6-glucuronides, codeine, codeine-glucuronide and 6-monoacetylmorphine in body fluids by liquid chromatography atmospheric pressure chemical ionization mass spectrometry. J. Chromatogr. B: Biomed. Sci. Appl. 5, 115– 127. Borron, S.W., Monier, C., Ris`ede, P., Baud, F.J., 2002. Flunitrazepam variably alters morphine, buprenorphine, and methadone lethality in the rat. Hum. Exp. Toxicol. 21, 599–605. Bruce, R.D., 1985. An up-and-down procedure for acute toxicity testing. Fund. Appl. Toxicol. 5, 151–157. Bullock, P., Spanner, S., Ansell, G.B., 1977. Distribution of morphine and morphine metabolites in rat brain. Biochem. Soc. Trans. 5, 166–168.
Darke, S., Ross, J., Zador, D., Sunjic, S., 2000. Heroin-related deaths in New South Wales, Australia, 1992–1996. Drug Alcohol Depend. 60, 141–150. Darke, S., Zador, D., 1996. Fatal heroin ‘overdose’: a review. Addiction 91, 1765–1772. Dixon, W.J., 1965. The up-and-down method for small samples. J. Am. Stat. Assoc. 60, 967–978. Druid, H., Strandberg, J.J., Alkass, K., Nystrom, I., Kugelberg, F.C., Kronstrand, R., 2006. Evaluation of the role of abstinence in heroin overdose deaths using segmental hair analysis. Forens. Sci. Int., Mar 22 [Epub ahead of print]. Gardmark, M., Karlsson, M.O., Jonsson, F., Hammarlund-Udenaes, M., 1998. Morphine-3-glucuronide has a minor effect on morphine antinociception. Pharmacodynamic modeling. J. Pharm. Sci. 87, 813–820. Hemstapat, K., Smith, S.A., Monteith, G.R., Smith, M.T., 2003. The neuroexcitatory morphine metabolite, morphine-3-glucuronide (M3G), is not neurotoxic in primary cultures of either hippocampal or cerebellar granule neurones. Pharmacol. Toxicol. 93, 197–200. Jackson, H., 1952. The evaluation of analgesic potency of drugs using thermal stimulation in the rat. Br. J. Pharmacol. Chemother. 7, 196–203. Kamata, O., Watanabe, S., Ishii, S., 1969. Analgesic effect of morphine glucuronides. In: Proceedings of the 89th Meeting Pharmacology Society, Japan, p. 443. L¨otsch, J., 2005. Opioid metabolites. J. Pain Symptom Manage. 29, 10–24. Manning, F.J., Ingraham, L.H., DeRouin, E.M., Vaughn, M.S., Kukura, F.C., St Michel, G.R., 1983. Drug “overdoses” among U.S. soldiers in Europe, 1978–1979. II. autopsies following deaths and neardeaths. Int. J. Addict. 18, 153–166. Meissner, C., Recker, S., Reiter, A., Friedrich, H.J., Oehmichen, M., 2002. Fatal versus non-fatal heroin “overdose”: blood morphine concentrations with fatal outcome in comparison to those of intoxicated drivers. Forens. Sci. Int. 130, 49–54. Milne, R.W., Nation, R.L., Somogyi, A.A., 1996. The disposition of morphine and its 3- and 6-glucuronide metabolites in humans and animals, and the importance of the metabolites to the pharmacological effects of morphine. Drug Metab. Rev. 28, 345–472. Nakamura, G.R., 1978. Toxicologic assessments in acute heroin fatalities. Clin. Toxicol. 13, 75–87. Shimomura, K., Kamata, O., Ueki, S., Ida, S., Oguri, K., 1971. Analgesic effect of morphine glucuronides. Tohoku J. Exp. Med. 105, 45–52. Siegel, S., Hinson, R.E., Krank, M.D., McCully, J., 1982. Heroin “overdose” death: contribution of drug-associated environmental cues. Science 216, 436–437. Sullivan, A.F., McQuay, H.J., Bailey, D., Dickenson, A.H., 1989. The spinal antinociceptive actions of morphine metabolites morphine6-glucuronide and normorphine in the rat. Brain Res. 482, 219–224. White, J.M., Irvine, R.J., 1999. Mechanisms of fatal opioid overdose. Addiction 94, 961–972.
Toxicological analysis in rats subjected to heroin and morphine overdose Joakim J. Strandberg a , Fredrik C. Kugelberg a , Kanar Alkass a , Anna Gustavsson a , Kolbjørn Zahlsen b , Olav Spigset b,c , Henrik Druid a,∗ a
Department of Forensic Medicine, Karolinska Institutet, SE-171 77 Stockholm, Sweden Department of Clinical Pharmacology, St. Olav University Hospital, NO-7006 Trondheim, Norway Department of Laboratory Medicine, Children’s and Women’s Disease, Norwegian University of Science and Technology, NO-7006 Trondheim, Norway b
c
Received 17 February 2006; received in revised form 10 May 2006; accepted 11 May 2006 Available online 16 May 2006
Abstract In heroin overdose deaths the blood morphine concentration varies substantially. To explore possible pharmacokinetic explanations for variable sensitivity to opiate toxicity we studied mortality and drug concentrations in male Sprague-Dawley rats. Groups of rats were injected intravenously (i.v.) with heroin, 21.5 mg/kg, or morphine, 223 mg/kg, causing a 60–80% mortality among drug-na¨ıve rats. Additional groups of rats were pre-treated with morphine for 14 days, with or without 1 week of subsequent abstinence. Brain, lung and blood samples were analyzed for 6-acetylmorphine, morphine, morphine-3-glucuronide and morphine-6-glucuronide. i.v. morphine administration to drug-na¨ıve rats resulted in both rapid and delayed deaths. The brain morphine concentration conformed to an exponential elimination curve in all samples, ruling out accumulation of morphine as an explanation for delayed deaths. This study found no support for formation of toxic concentration of morphine-6-glucuronide. Spontaneous death among both heroin and morphine rats occurred at fairly uniform brain morphine concentrations. Morphine pre-treatment significantly reduced mortality upon i.v. morphine injection, but the protective effect was less evident upon i.v. heroin challenge. The morphine pre-treatment still afforded some protection after 1 week of abstinence among rats receiving i.v. morphine, whereas rats given i.v. heroin showed similar death rate as drug-na¨ıve rats. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Glucuronide; Heroin; Opiate; Overdose; Pharmacokinetics; Tolerance
1. Introduction About 8000 acute drug-related deaths are recorded annually in the EU (http://www.emcdda.eu.int/). An evaluation of the reported deaths during several years
∗ Corresponding author. Tel.: +46 8 52 48 77 70; fax: +46 8 32 14 52. E-mail address: [email protected] (H. Druid).
in 10 countries shows that approximately 80% of these deaths are related to opiate overdose. Typically, such opiate toxicity deaths strike suddenly and unexpectedly even among experienced users. Postmortem toxicological findings in deceased heroin addicts do not support that heroin “overdose” deaths result from the intake of a large dose. In a majority of these victims the blood morphine concentration is actually found to be lower than that seen in heroin users that did not die from overdose but from other causes (Darke
0378-4274/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2006.05.007
12
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
and Zador, 1996). Further, blood morphine concentration varies substantially, and cannot be used in isolation to diagnose heroin overdose deaths (Meissner et al., 2002). By performing segmental hair analysis, we have recently found that abstinent overdose victims presented with similar blood morphine concentration as did overdose victims with evidence of continuous opioid drug use prior to death (Druid et al., 2006). In humans, heroin (3,6-diacetylmorphine) is rapidly degraded to the metabolites 6-acetylmorphine (6-AM) and morphine. Most of the morphine is further converted to morphine-3-glucuronide (M3G; ∼50%) and morphine-6-glucuronide (M6G; ∼10%) (Aderjan and Skopp, 1998). Heroin quickly enters the brain through the blood–brain barrier, however, due to its rapid metabolism, it is generally assumed that heroin itself has only minor pharmacological effects. Most of the pharmacological effects, including respiratory depression, are instead caused by morphine, or to some extent, by 6-AM and M6G (White and Irvine, 1999). Since the demonstration in the late 1960s that M6G possesses analgesic properties (Kamata et al., 1969) it has been proposed that the glucuronides play an important role for the toxicity of heroin and morphine. However, the main metabolite, M3G, does not seem to have any agonistic effect either in vivo or in cell cultures (Shimomura et al., 1971; Hemstapat et al., 2003). In contrast, there is some evidence that M3G may exert antagonistic effects and to some extent compete with morphine and M6G at binding sites (Gardmark et al., 1998). When M6G is directly administered to humans, some studies have shown that side effects like nausea, vomiting and respiratory depression are attenuated compared with the administration of analgesically equipotent doses of morphine, whereas others have found no such differences (L¨otsch, 2005). In rats, the formation of M6G is limited (Milne et al., 1996); hence, heroin and morphine administration to rats is not expected to produce toxic concentration of this metabolite. Therefore, rat studies offer good conditions for the evaluation of the inherent toxicity of morphine and heroin. Since blood morphine concentration in apparent heroin overdose deaths vary substantially, we decided to explore the possible correlation between the concentrations of several heroin metabolites and the outcome in an animal model, where rats were given high doses of either morphine or heroin. We hypothesized that a gradual accumulation of either morphine or M6G in the brain together with a drop in blood morphine concentration might explain low blood morphine concentration at death. Further, we also wanted to address the question of whether accumulation of opiate metabolites in
the brain could account for delayed deaths. In order to assess the impact of tolerance, we included groups of rats that were pre-treated with intraperitoneal (i.p.) injections of morphine, and then challenged them with an acute intravenous (i.v.) injection of heroin or morphine, with or without a preceding period of abstinence. 2. Material and methods 2.1. Animals Male Sprague-Dawley rats (Scanbur BK AB, Sollentuna, Sweden) initially weighing 216–376 g were used. Rats were allowed to acclimatize for 1 week prior to experiment. Rats were allocated in groups of four in each cage and allowed fresh water and standard rat chow ad libitum throughout the whole experimental period. Rats were kept in a 12:12 h light:dark cycle synchronous with daylight (lights on at 7.00 a.m.). All experiments were performed in strict accordance with the guidelines and the consent of the Animal Ethics Committee of Northern Stockholm (Permit no. 187/03, 413/03 and 110/05). 2.2. Drugs Heroin hydrochloride was obtained from Macfarlan Smith Ltd., Edinburgh, UK and morphine hydrochloride was purchased from the Swedish Pharmacy (Apoteket AB). Both drugs were dissolved in 0.9% NaCl and slightly heated (40–50 ◦ C) when needed to be dissolved. 2.3. Experimental procedure The experimental design of the study is illustrated in Fig. 1. Rats were randomized to seven groups, each comprising 6–14 rats treated as follows: Group 1 (a and b) i.v. injection of (a) heroin, (b) morphine; Group 2 (a and b) i.p. injections of morphine for 14 days followed by an i.v. injection of (a) heroin or (b) morphine; Group 3 (a and b) i.p. injections of morphine for 14 days followed by 1 week without any injections (i.e. abstinence) and then an i.v. injection of (a) heroin, (b) morphine; Group 4 (b) i.p. injections of (b) morphine for 14 days. i.p. injections with morphine were given once daily between 10:00 a.m. and 2:00 p.m., with a 0.6 mm × 25 mm needle for 14 consecutive days using the following dosage schedule: 20 mg/kg (days 1–3), 40 mg/kg (days 4–6), 60 mg/kg (days 7–9), 80 mg/kg (days 10–12) and 100 mg/kg (days 13–14). The volumes administered ranged from a minimum of 0.31 mL to a maximum of 0.56 mL. The up-and-down method (Dixon, 1965; Bruce, 1985) was applied to establish a dose causing 60–80% mortality. This dose was found to be 21.5 mg/kg for heroin and 223 mg/kg for morphine. Rats were given i.v. injections of these doses of morphine or heroin in the tail vein (volumes ranging from 0.42 to 0.46 mL). All i.v. injections were given in the morning (8.00–10.00 a.m.). Immediately after injection, rats were placed in separate cages and observed for a maximum of 8 h. Rats that died spontaneously during this period
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
13
Fig. 1. Experimental design. For details, see text.
were decapitated, whereas rats that survived were quickly anesthetized with isoflurane (Forene® ; Abbott Scandinavia AB, Solna, Sweden) and decapitated 8 h after the i.v injection. After decapitation, blood was drawn from the heart and mixed with sodium fluoride to prevent degradation of drugs. The brain was quickly removed and the occipital lobe and the cervical spinal cord were dissected free and collected for toxicological analysis. The lungs were removed and the right lung was kept for toxicological analysis. The blood, brain and lung samples were stored at −80 ◦ C until drug analysis. 2.4. Analysis of 6-AM, morphine, M3G and M6G Analysis of 6-AM, morphine, M3G and M6G in blood and tissues was performed by electrospray liquid chromatography–mass spectrometry (ESI-LC–MS). In brief, prior to extraction, tissue samples from occipital lobe, cervical spinal cord and lung were excised and weighed individually. A weighed amount ranging from 15 to 79 mg (mean 52 mg) was taken from the occipital lobe, 12–79 mg (mean 58 mg) from the cervical spinal cord and 93–110 mg (mean 102 mg) from the lung. The tissue samples were homogenized with an Ultra-Turrax homogenizer together with 1 mL of ammonium carbonate buffer containing internal standards. Blood samples (weighing 56–98 mg; mean 71 mg) were mixed with 1 mL of ammonium carbonate buffer. The homogenate and diluted blood samples were extracted according to Bogusz et al. (1997) with slight modifications, by solid phase extraction (SPE) using Chromabond C18 SPE extraction columns (Machery-Nagel GmbH). The columns
were preconditioned with 1 mL of methanol, then 1 mL of ultrapure water and finally 2 mL of ammonium carbonate buffer. The homogenized/diluted sample was loaded on column and aspirated by vacuum. The column was washed with 3 mL of ammonium carbonate buffer and then dried for 5 min. The analytes were eluted in a volume of 0.5 mL of methanol and acetic acid (9:1), evaporated and reconstituted in 50 L of ammonium acetate. The eluates were then transferred to autosampler vials for LC–MS analysis. The concentrations of analytes were determined using an Agilent 1100 mass spectrometric system consisting of a G1313A autosampler, a G1322A degassing unit, a G1311A quaternary pump, a G1316A column oven and a single quadropole G1946D mass spectrometer. The analytes were monitored in positive mode SIM at the following m/z values: Morphine 286.1, 6-AM 328.1, M3G 462.1 and M6G 462.1. Morphine-D3 (monitored at m/z 289.1) was used as an internal standard for morphine, M3G and M6G, whereas 6-AM-D3 (monitored at m/z 331.1) was used as an internal standard for 6-AM. Calibration curves were established to cover a range of concentrations from 0.1 up to 1000 g/g. Limits of quantitation for morphine and 6-AM were 0.2 g/g and for M3G and M6G 0.1 g/g with a sample size of 50 mg. The extraction recoveries ranged from 40 to 84%. 2.5. Statistics A value of p < 0.05 was considered statistically significant. Possible differences between the groups were analyzed by
14
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
Table 1 Mortality after i.v. injection of morphine or heroin, with or without morphine pre-treatment Treatment
Group 1 (i.v.) Group 2 (i.p.–i.v.) Group 3 (i.p.–abst–i.v.)
Heroin 21.5 mg/kg
Morphine 223 mg/kg
Death within 10 min
Delayed death
Sacrificed at 480 min
Time to death (min)a
Death within 10 min
Delayed death
Sacrificed at 480 min
Time to death (min)a
8 (67%) 3 (23%) 8 (62%)
0 0 0
4 (33%) 10 (77%) 5 (38%)
163 ± 67.6 370 ± 58.3 187 ± 66.9
7 (50%) 0 2 (25%)
4 (29%) 0 0
3 (21%) 7 (100%) 6 (75%)
151 ± 52.8 480 ± 0 362 ± 77.6
i.v., intravenous; i.p., intraperitoneal injections of morphine for 2 weeks; abst, 1 week of abstinence after i.p. pre-treatment. Delayed death, spontaneous death between 10 and 480 min. Rats sacrificed at 480 min were included when calculating mean time to death. a Values are mean ± S.E.M.
one-factor analysis of variance (ANOVA). When the ANOVA reached statistical significance, Fisher’s protected least significant difference (PLSD) post hoc test was applied. All statistical analyses were performed using StatView® for Windows Version 5.0 (SAS® Institute Inc., Cary, NC, USA).
3. Results i.v. doses causing approximately 70% mortality among drug-na¨ıve rats was established prior to this study (21.5 mg/kg for heroin and 223 mg/kg for morphine). In the present study these doses produced 67% mortality among drug-na¨ıve rats given i.v. heroin and 79% mortality among drug-na¨ıve rats given i.v. morphine (Table 1). In comparison, animals receiving 2 weeks of pre-treatment with i.p. morphine showed a lower mortality after i.v. heroin as well as after i.v. morphine injection. Even after 1 week of abstinence following pretreatment, the mortality was still lower than in drug-na¨ıve rats; 62% after i.v. heroin and 25% after i.v. morphine. Hence, a significant difference was found in time to death between drug-na¨ıve i.v. heroin and pre-treated i.v. heroin rats (p = 0.016); between drug-na¨ıve i.v. morphine and pre-treated i.v. morphine rats (p = 0.001), and between drug-na¨ıve i.v. morphine and pre-treated rats that were subjected to 1 week of abstinence before i.v. morphine injection (p = 0.027). The morphine concentration in all tissues varied substantially among rats that spontaneously died within 10 min after i.v. injection. Rats receiving i.v. morphine (Group 1b) had much higher concentration of morphine in blood and lung tissue than those receiving heroin (Group 1a), whereas the concentration of morphine in cervical spinal cord were similar in both groups (Table 2). In samples from the occipital lobe, i.v. heroin rats had significantly higher morphine concentration compared with i.v. morphine rats (p < 0.05). Further, i.v. heroin rats in all subgroups had higher concentration of morphine in brain and lung tissue than in blood (p-values
ranging from p < 0.01 to 0.0001). Also, the concentration of 6-AM in i.v. heroin rats was higher in brain tissue compared to concentration in blood (p-values ranging from p < 0.05 to 0.0001). However, 6-AM concentration was consistently lowest in lung tissue. The time to death for the delayed deaths (defined as death between 10 and 480 min) in Group 1b ranged from 25 to 262 min (mean 162 ± 94 min, n = 4). These rats usually showed froth in the airways as well as forced and irregular breathing. No accumulation of drug could be detected in these rats, and the drug concentrations followed an exponential elimination curve. No delayed deaths occurred among i.v. heroin rats. These rats either died quickly, within 10 min, or survived (but then showed blood-tinged froth in the mouth, and had a strained breathing until sacrificed 8 h after injection). Rats that were sacrificed 8 h after i.v. injection generally showed low concentrations of all opiate compounds (Table 2). The morphine concentration was similar across all different types of tissues, but was typically higher in lung tissue than in other tissues. When comparing the time of death with the blood and tissue concentrations of the opiate compounds, we found that the concentrations of all analytes conformed to a predicted exponential elimination (Fig. 2). 4. Discussion We developed a rat model to mimic morphine and heroin overdose death in humans. Our major findings were: (1) death occurred at different times after i.v. injection, but there was no evidence of accumulation of morphine or other opiate compounds in the brain to explain for delayed deaths; (2) pre-treatment with morphine significantly reduced mortality after i.v. morphine injection, and some protective effect was still observed after 1 week of abstinence; however, pre-treatment with morphine did not offer such protection against i.v. heroin injection;
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
15
Table 2 Drug and metabolite concentrations (g/g) following administration of heroin and morphine to rats
Group 1a: heroin (i.v.) Dead within 10 min (n = 8) 6-MAM Morphine M3G Sacrificed at 480 min (n = 4) 6-MAM Morphine M3G Group 1b: morphine (i.v.) Dead within 10 min (n = 7) Morphine M3G
Blood
Lung
Cervical spinal cord
Occipital lobe
17.5 ± 8.05 4.86 ± 1.88 0.15 (n = 2)
7.91 ± 2.38 43.5 ± 23.6 n.d.
39.2 ± 8.6 27.3 ± 5.24 n.d.
64.6 ± 11.8 59.4 ± 13.9 n.d.
n.d. n.d. 0.14 (n = 2)
n.d. 0.41 ± 0.26 0.12 ± 0.09
0.32 (n = 2) 0.23 (n = 1) n.d.
0.14 (n = 1) 0.16 (n = 1) n.d.
304 ± 234 4.41 ± 4.60
580 ± 295 n.a.
37.4 ± 27.8 n.a.
33.1 ± 9.92 n.a.
Delayed deaths (n = 4)
Death between 10 and 480 min. For details, see text and Fig. 2
Sacrificed at 480 min (n = 3) Morphine M3G
0.40 (n = 2) 6.29 (n = 2)
1.93 ± 0.79 n.a.
0.26 (n = 1) n.a.
0.40 (n = 2) n.a.
25.4 ± 2.6 8.91 ± 0.82 1.62 ± 1.51
16.8 ± 1.73 94.0 ± 14 1.69 ± 1.2
34.6 ± 2.14 27.0 ± 2.73 0.16 ± 0.06
67.8 ± 11.0 70.8 ± 7.94 0.11 (n = 2)
n.d. 0.21 ± 0.05 0.43 ± 0.13
0.26 (n = 1) 1.74 ± 0.9 0.43 ± 0.11
0.22 ± 0.06 0.19 (n = 2) 0.09 (n = 1)
0.27 ± 0.15 0.24 ± 0.07 n.d.
– –
– –
– –
– –
Sacrificed at 480 min (n = 7) Morphine M3G
1.25 ± 0.34 2.70 ± 2.12
5.82 ± 1.88 n.a.
0.75 (n = 2) n.a.
0.72 ± 0.24 n.a.
Group 3a: heroin (i.p.–abst–i.v.) Dead within 10 min (n = 8) 6-MAM Morphine M3G
18.3 ± 7.45 6.26 ± 2.52 n.d.
6.20 ± 2.04 65.6 ± 41.6 0.05 ± 0.01
28.0 ± 7.58 34.8 ± 14.5 n.d.
50.5 ± 12.4 79.7 ± 28.3 n.d.
Sacrificed at 480 min (n = 5) 6-MAM Morphine M3G
0.18 (n = 1) 0.19 (n = 1) 0.11 ± 0.02
n.d. 0.78 ± 0.26 0.10 ± 0.03
0.18 ± 0.06 n.d. n.d.
0.18 (n = 2) 0.22 (n = 2) n.d.
243 (n = 2) 1.89 (n = 2)
428 (n = 2) n.a.
33.5 (n = 2) n.a.
30.2 (n = 2) n.a.
0.74 ± 0.48 2.48 ± 1.81
2.74 ± 1.15 n.a.
0.39 (n = 1) n.a.
0.27 (n = 1) n.a.
Group 2a: heroin (i.p.–i.v.) Dead within 10 min (n = 3) 6-MAM Morphine M3G Sacrificed at 480 min (n = 10) 6-MAM Morphine M3G Group 2b: morphine (i.p.–i.v.) Dead within 10 min (n = 0) Morphine M3G
Group 3b: morphine (i.p.–abst–i.v.) Dead within 10 min (n = 2) Morphine M3G Sacrificed at 480 min (n = 6) Morphine M3G
16
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
Table 2 (Continued) Blood Group 4: morphine (i.p.) Sacrificed after last i.p. (n = 6) Morphine 0.11 ± 0.02 M3G 0.54 ± 0.08
Lung
Cervical spinal cord
Occipital lobe
0.51 ± 0.17 n.a.
n.d. n.a.
n.d. n.a.
Values are mean ± S.D. n.d., not detectable (below the limit of detection); n.a., not analysed; i.v., intravenous injection of either heroin 21.5 mg/kg or morphine 223 mg/kg; i.p., intraperitoneal injections of morphine for 2 weeks; abst, 1 week of abstinence after i.p. pre-treatment; 6-MAM, 6-monoacetylmorphine; M3G, morphine-3-glucuronide; M6G, morphine-6-glucuronide.
(3) the occipital lobe generally had higher opiate (6AM and morphine) concentrations than other tissues in rapid deaths after i.v. heroin injection, but were similar to cervical spinal cord concentrations after i.v. morphine injection. We aimed to use an i.v. dose that could produce spontaneous deaths among both drug-na¨ıve and morphine pre-treated rats, allowing for the evaluation of toxicological results in rats that died spontaneously as well as in rats sacrificed at defined times after drug administration. We found that heroin 21.5 mg/kg and morphine 223 mg/kg caused approximately 60–80% mortality in
na¨ıve Sprague-Dawley rats. Jackson (1952) studied the mortality after i.v. administration of heroin and morphine to na¨ıve Wistar rats and found the dose causing 50% mortality (LD50) to be 22.5 mg/kg for heroin. However, due to problems with solubility, the LD50 of morphine was concluded to “appear to” be 140 mg/kg. Siegel et al. (1982) found that i.v. administration of heroin hydrochloride 15 mg/kg to na¨ıve Wistar rats caused a mortality of 96%. Borron et al. (2002) injected morphine sulphate i.v. to Sprague-Dawley rats producing LD50 of 64 mg/kg. The difference in lethal dose in our study compared to LDs in other studies is most likely due to use
Fig. 2. Morphine concentrations in (a) blood, (b) occipital lobe, (c) cervical spinal cord and (d) lung, in rats given i.v. injections of morphine without pre-treatment (Group 1b). Correlation coefficients between morphine concentration and time to death were r = 0.566 (p = 0.0696) in blood, r = 0.799 (p = 0.0031) in occipital lobe, r = 0.435 (p = 0.2095) in cervical spinal cord and r = 0.680 (p = 0.0212) in lung.
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
of different kind of rat strains (Sprague-Dawley vs Wistar) and to different drug preparations (hydrochloride vs sulphate). The observation of delayed death among approximately 20–50% of the human heroin addicts (Nakamura, 1978; Manning et al., 1983; Darke et al., 2000) also applies to rats. Borron et al. (2002) showed that the median time to death was approximately 2.5 h (range 0.8–24 h) after i.v administration of morphine sulphate 64 mg/kg (LD50) to Sprague-Dawley rats. The time to death increased dramatically when rats were pre-treated with flunitrazepam, producing a delayed death about 13.5 h after the morphine injection. In our study we observed delayed deaths among opiate-na¨ıve rats given i.v. morphine, whereas no delayed deaths were seen among i.v. heroin rats. The reason for these delayed deaths remains unclear, but they could not be explained by pharmacokinetic mechanisms, since there was no evidence of accumulation of drugs in the brain or other tissues. Surprisingly, rats given i.v. heroin either died rapidly or survived the whole surveillance period. The absence of delayed death among the heroin injected rats might be explained by a too short observation time (8 h). Surviving rats generally showed irregular and decreased breathing and slightly blood-tinged froth in the airways, so it is possible that some of them might have died later if the surveillance period had been longer. It is also possible that a slightly lower dose might have produced some delayed deaths. In humans, abstinence, e.g. due to imprisonment or enrolment in treatment programs, is one factor that is frequently claimed to increase the risk of opiate overdose, even by drug addicts themselves (Bennett and Higgins, 1999). However, interviewed drug addicts rank abstinence as less important than the intake of an excessive amount of drug or mixing of different drugs (Bennett and Higgins, 1999). Siegel et al. (1982) investigated the impact of tolerance in an overdose paradigm using Wistar rats. Animals were pre-treated with increasing doses of heroin (i.v.) and thereafter subjected to an i.v. dose of heroin, killing approximately 96% of the na¨ıve rats and only 32% of the tolerant rats. In our study, pre-treatment for 14 days with morphine i.p. reduced the mortality from 79 to 0% after administration of i.v. morphine. This protective effect of morphine pre-treatment was less evident in rats given i.v. heroin; mortality was reduced from 67 to 23%. Still, after 1 week of abstinence, the mortality following i.v. morphine was reduced, whereas rats given i.v. heroin died at the same rate as did opiate-na¨ıve rats. The reason for this difference in impact of pre-treatment on the outcome after morphine and heroin administration remains unsolved, but one possibility is that it is
17
related to the more rapid delivery of morphine to key areas in the brain after heroin injection. If mu-opioid receptors in the respiratory centers are exposed to a very rapid and high concentration of morphine, there might not be enough time for the acute adaptations to come into action, regardless of long-term adaptations in the cells. Another possibility is that 6-AM contributes to the toxicity via signaling not affected by morphine pre-treatment. In the present study, we also found differences in the incorporation of drugs and metabolites into specific tissues. The occipital lobe generally had a higher concentration of 6-AM and morphine compared to the cervical spinal cord after i.v. heroin. Bullock et al. (1977) examined the regional distribution of morphine and its metabolites (not specified) after subcutaneous injection with radioactively labeled morphine. Morphine concentration in na¨ıve rats was found to be significantly higher in cerebellum compared to midbrain, hypothalamus, cortex and striatum. The morphine concentration was also slightly higher in cerebellum than in medulla oblongata. Further, Bullock et al. (1977) demonstrated that the distribution of morphine uptake was similar between na¨ıve and morphine-dependent rats, but the amount of morphine uptake was significantly decreased in dependent rats. The differences in tissue disposition might have several explanations, such as variations in fat content or in the density of opioid receptors. In very rapid deaths, differences in regional blood flow might also influence the concentration in various brain regions. Several groups have shown that M6G have greater antinociceptive effect compared to morphine when administered directly to rats (Abbott and Palmour, 1988; Sullivan et al., 1989). In rats, the formation of M6G is limited. Antonilli et al. (2003) showed that chronic heroin administration, but not chronic morphine administration resulted in low concentration of M6G. In the present study, we observed a minimal amount of M6G formation in blood (0.07–0.24 g/g) in a few rats after a very high dose of i.v. morphine (223 mg/kg) but not after i.v. heroin injection. Since the formation of M6G was very petite, it seems unlikely that this metabolite contributed to any significant degree to the mortality in this study. The study design aimed at mimicking human opiate overdose. There are certain species differences regarding metabolism rate, and metabolic pathways between humans and rodents that may render the results not completely applicable to humans. However, the observation of respiratory arrest (before cardiac arrest) and the occurrence of delayed deaths lend support to such applicability. Since toxicokinetic studies are ethically impossible
18
J.J. Strandberg et al. / Toxicology Letters 166 (2006) 11–18
to perform on human subjects, we think that the results in this study contribute to the understanding of the toxic effects that opioids confer to human overdose victims. In conclusion, the results of the present study corroborate the impact of tolerance in acute opioid toxicity, although pre-treatment with morphine was more protective in rats administered i.v. morphine than in rats given i.v. heroin injection. The reason for the delayed deaths observed was seemingly not related to pharmacokinetic factors and warrants further investigation. Acknowledgements We thank Wenche Rødseth Brede, Department of Clinical Pharmacology, St. Olav University Hospital, Trondheim, Norway, for technical assistance with the drug analysis. This study was funded by grants from the Swedish National Drug Policy Coordinator, the Swedish National Board of Forensic Medicine, the Swedish Medical Society, Lennanders Foundation and the Magn. Bergvall Foundation. References Abbott, F.V., Palmour, R.M., 1988. Morphine-6-glucuronide: analgesic effects and receptor binding profile in rats. Life Sci. 43, 1685–1695. Aderjan, R.E., Skopp, G., 1998. Formation and clearance of active and inactive metabolites of opiates in humans. Ther. Drug Monit. 20, 561–569. Antonilli, L., Suriano, C., Paolone, G., Badiani, A., Nencini, P., 2003. Repeated exposures to heroin and/or cadmium alter the rate of formation of morphine glucuronides in the rat. J. Pharmacol. Exp. Ther. 307, 651–660. Bennett, G.A., Higgins, D.S., 1999. Accidental overdose among injecting drug users in Dorset, UK. Addiction 94, 1179–1189. Bogusz, M.J., Maier, R.D., Erkens, M., Driessen, S., 1997. Determination of morphine and its 3- and 6-glucuronides, codeine, codeine-glucuronide and 6-monoacetylmorphine in body fluids by liquid chromatography atmospheric pressure chemical ionization mass spectrometry. J. Chromatogr. B: Biomed. Sci. Appl. 5, 115– 127. Borron, S.W., Monier, C., Ris`ede, P., Baud, F.J., 2002. Flunitrazepam variably alters morphine, buprenorphine, and methadone lethality in the rat. Hum. Exp. Toxicol. 21, 599–605. Bruce, R.D., 1985. An up-and-down procedure for acute toxicity testing. Fund. Appl. Toxicol. 5, 151–157. Bullock, P., Spanner, S., Ansell, G.B., 1977. Distribution of morphine and morphine metabolites in rat brain. Biochem. Soc. Trans. 5, 166–168.
Darke, S., Ross, J., Zador, D., Sunjic, S., 2000. Heroin-related deaths in New South Wales, Australia, 1992–1996. Drug Alcohol Depend. 60, 141–150. Darke, S., Zador, D., 1996. Fatal heroin ‘overdose’: a review. Addiction 91, 1765–1772. Dixon, W.J., 1965. The up-and-down method for small samples. J. Am. Stat. Assoc. 60, 967–978. Druid, H., Strandberg, J.J., Alkass, K., Nystrom, I., Kugelberg, F.C., Kronstrand, R., 2006. Evaluation of the role of abstinence in heroin overdose deaths using segmental hair analysis. Forens. Sci. Int., Mar 22 [Epub ahead of print]. Gardmark, M., Karlsson, M.O., Jonsson, F., Hammarlund-Udenaes, M., 1998. Morphine-3-glucuronide has a minor effect on morphine antinociception. Pharmacodynamic modeling. J. Pharm. Sci. 87, 813–820. Hemstapat, K., Smith, S.A., Monteith, G.R., Smith, M.T., 2003. The neuroexcitatory morphine metabolite, morphine-3-glucuronide (M3G), is not neurotoxic in primary cultures of either hippocampal or cerebellar granule neurones. Pharmacol. Toxicol. 93, 197–200. Jackson, H., 1952. The evaluation of analgesic potency of drugs using thermal stimulation in the rat. Br. J. Pharmacol. Chemother. 7, 196–203. Kamata, O., Watanabe, S., Ishii, S., 1969. Analgesic effect of morphine glucuronides. In: Proceedings of the 89th Meeting Pharmacology Society, Japan, p. 443. L¨otsch, J., 2005. Opioid metabolites. J. Pain Symptom Manage. 29, 10–24. Manning, F.J., Ingraham, L.H., DeRouin, E.M., Vaughn, M.S., Kukura, F.C., St Michel, G.R., 1983. Drug “overdoses” among U.S. soldiers in Europe, 1978–1979. II. autopsies following deaths and neardeaths. Int. J. Addict. 18, 153–166. Meissner, C., Recker, S., Reiter, A., Friedrich, H.J., Oehmichen, M., 2002. Fatal versus non-fatal heroin “overdose”: blood morphine concentrations with fatal outcome in comparison to those of intoxicated drivers. Forens. Sci. Int. 130, 49–54. Milne, R.W., Nation, R.L., Somogyi, A.A., 1996. The disposition of morphine and its 3- and 6-glucuronide metabolites in humans and animals, and the importance of the metabolites to the pharmacological effects of morphine. Drug Metab. Rev. 28, 345–472. Nakamura, G.R., 1978. Toxicologic assessments in acute heroin fatalities. Clin. Toxicol. 13, 75–87. Shimomura, K., Kamata, O., Ueki, S., Ida, S., Oguri, K., 1971. Analgesic effect of morphine glucuronides. Tohoku J. Exp. Med. 105, 45–52. Siegel, S., Hinson, R.E., Krank, M.D., McCully, J., 1982. Heroin “overdose” death: contribution of drug-associated environmental cues. Science 216, 436–437. Sullivan, A.F., McQuay, H.J., Bailey, D., Dickenson, A.H., 1989. The spinal antinociceptive actions of morphine metabolites morphine6-glucuronide and normorphine in the rat. Brain Res. 482, 219–224. White, J.M., Irvine, R.J., 1999. Mechanisms of fatal opioid overdose. Addiction 94, 961–972.