Accepted Manuscript Title: Profiles of pregabalin and gabapentin abuse by postmortem toxicology Author: Margareeta H¨akkinen Erkki Vuori Eija Kalso Merja Gergov Ilkka Ojanper¨a PII: DOI: Reference:
S0379-0738(14)00181-9 http://dx.doi.org/doi:10.1016/j.forsciint.2014.04.028 FSI 7587
To appear in:
FSI
Received date: Revised date: Accepted date:
16-2-2014 15-4-2014 27-4-2014
Please cite this article as: Margareeta H¨akkinen, Erkki Vuori, Eija Kalso, Merja Gergov, Ilkka Ojanper¨a, Profiles of pregabalin and gabapentin abuse by postmortem toxicology, Forensic Science International http://dx.doi.org/10.1016/j.forsciint.2014.04.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Profiles of pregabalin and gabapentin abuse by postmortem toxicology
ip t
Margareeta Häkkinena*, Erkki Vuoria, Eija Kalsob, Merja Gergova, Ilkka Ojanperäa Hjelt Institute, Department of Forensic Medicine, Kytösuontie 11, P.O. Box 40, FI-00014 University of Helsinki, Helsinki, Finland b Institute of Clinical Medicine, University of Helsinki, and Pain Clinic, Helsinki University Central Hospital, Helsinki, Finland
us
cr
a
* Corresponding author. Tel.: +358 414354313; fax: +358 919127518. E-mail address: [email protected] (M. Häkkinen).
an
Abstract Pregabalin (PRG) and gabapentin (GBP) are used in the treatment of neuropathic
M
pain and epilepsy, and PRG also in generalized anxiety disorder. There is increasing evidence that PRG possesses considerable abuse potential. PRG may have a higher addiction potential than GBP due to its rapid absorption and faster
d
onset of action. Our objective is to estimate the proportion of all PRG- and GBP-
Ac ce pt e
related fatalities attributable to PRG and GBP abuse. We investigated all medicolegal death cases in Finland in which PRG or GBP was found in postmortem toxicology during 2010–2011. PRG was found in 316 cases and GBP in 43 cases. Drug abuse was associated with 48.1% of the PRG and 18.6% of the GBP findings. PRG poisoning accounted for 10.1% of all PRG cases and GBP poisoning for 4.7% of all GBP cases. In the drug abuser cases, PRG poisoning represented 19.1%, and GBP poisoning 12.5%. The median blood concentration of PRG was 15 mg/L in the abuser group and 5.8 mg/L in the other cases. For GBP, these concentrations were 12 mg/L and 8.3 mg/L, respectively. In the PRG abuser group, 91.4% of cases showed concomitant opioid use, while in the rest of these cases neither alcohol nor opioids were detected, but other central nervous system acting drugs were found in each abuser case. In the GBP abuser group, 87.5% of cases showed concomitant opioid use. PRG abuse with high doses is increasingly common and can be fatal when combined with opioids.
1
Page 1 of 18
Keywords: pregabalin; gabapentin; drug abuse; postmortem toxicology; fatal poisoning; blood drug concentration
Introduction Pregabalin (PRG) is a gamma-aminobutyric acid (GABA) analog used for treating
ip t
neuropathic pain and partial seizures. In the European Union, PRG is also
approved for treating generalized anxiety disorder, and in the United States for
cr
fibromyalgia. The recommended daily dose of PRG is 150–600 mg divided into two or three smaller doses, and the defined daily dose (DDD) by the World
us
Health Organisation (WHO) is 300 mg. Gabapentin (GBP), like PRG, is a GABA analog used for neuropathic pain and epilepsy. The recommended daily dose of
an
GBP is 900–3600 mg divided into three smaller doses, and the DDD is 1800 mg.
Although in clinical trials PRG has been found to produce euphoria as an adverse
M
drug reaction, it was considered to have low potential for abuse and dependence, unlike other classes of medications used for the treatment of generalized anxiety
d
disorder [1]. However, in recent years there has been increasing concern about
Ac ce pt e
the abuse potential of PRG. Swedish researchers [2] found that reports to the national register of PRG abuse and addiction started to rise significantly in the fourth quarter of 2008, and concluded that pregabalin is likely to have abuse potential.
Similarly in Germany [3], researchers have since 2008 identified an increasing number of cases of PRG abuse or dependence reported to the national database. The mean daily PRG dosage was as high as 1424 mg, and in about one-third of patients withdrawal symptoms subsequent to discontinuation of PRG were reported. Male sex and a history of polytoxicomania were suggested as possible risk factors for the development of addictive behavior related to PRG [3]. In a German study from 2010–2012, the number of postmortem cases with a PRG finding increased from 10 to 33 within a two-year period. Among a subgroup of drug-dependent individuals, the number of PRG cases increased proportionally more, from 4 to 26 cases, within the same two-year period [4].
2
Page 2 of 18
In a driving simulation experiment, PRG was shown to cause mild effects on some components of driving performance but without serious central nervous system side effects [5]. In Finnish apprehended drivers, PRG was commonly found at concentrations higher than therapeutic together with other psycho-
ip t
active drugs, suggesting drug abuse [6]. An international team of researchers [7] analyzed both anecdotal online reports
cr
of PRG and GBP misuse and their online availability for purchase during 2008–
drug for recreational purposes to achieve specific
us
2010. Based on Internet searches, PRG was described as an ideal psychotropic
alcohol/gammahydroxybutyrate (GHB)/benzodiazepine-like effects mixed with
an
euphoria, to achieve entactogenic feelings and dextromethorphan-like disassociation, and to cope with opioid withdrawal. GBP had similar sedative,
M
dissociative and psychedelic effects but, according to one user, PRG was more desirable in recreational use than GBP [7].
d
Concern over GBP abuse has risen in Scotland, where the number of GBP
Ac ce pt e
prescriptions has increased exponentially since 2002. GBP abuse has also been noticed in the substance misuse service. Individuals abusing GBP describe experiences such as euphoria, improved sociability, relaxation and a marijuanalike “high” [8].
PRG may have a higher addiction potential than GBP due to its rapid absorption and faster onset of action. PRG has higher bioavailability than GBP [9]. Unlike GBP, PRG exhibits linear uptake without transporter saturation at therapeutic dosages, whereas the bioavailability of GBP drops as the dosage increases. In treating postherpetic neuralgia and epilepsy, PRG might be more effective than GBP [9]. PRG and GBP have no clinically relevant drug interactions. More than 90% of PRG is excreted unmetabolized. Renal failure increases PRG and GBP toxicity since both are eliminated renally [9].
3
Page 3 of 18
Despite the apparent abuse potential of PRG, no comprehensive comparative data on the profile of either PRG or GBP has been published based on findings from postmortem cases. In Finland, PRG abuse was revealed as long ago as 2007 by postmortem toxicology [10], and the phenomenon has since escalated. Our objective is to estimate the proportion of all PRG- and GBP-related fatalities attributable to PRG and GBP abuse during 2010–2011 in Finland, taking
ip t
advantage of the high medico-legal autopsy rate and the comprehensive
postmortem toxicology database generated. We further analyze the data in terms
cr
of age, gender, fatal poisonings, alcohol and opioid findings, manner of death and postmortem PRG and GBP blood concentration. Additionally, we present two
us
typical case reports concerning PRG deaths: a suicidal PRG poisoning with only small amounts of other drugs, and an accidental PRG poisoning involving
d
Data collection
M
Material and methods
an
buprenorphine and other drugs.
Ac ce pt e
Our primary data consisted of all recorded deaths in Finland in which a comprehensive postmortem toxicological analysis was performed at the Hjelt Institute, Department of Forensic Medicine during 2010–2011. All postmortem toxicology samples taken at autopsy in Finland are analyzed in our laboratory. During 2010–2011 there were 101,472 deaths in Finland, and a medico-legal autopsy was performed in 22,421 cases (22.1% of all deaths). Toxicological analyses were performed in 13,766 cases. From this material, the cases with a confirmed positive PRG or GBP finding were included in this study. All cases with a PRG or GBP finding from a postmortem sample, such as blood or urine, were recorded as PRG- or GBP-related cases. PRG and GBP concentration data were collected from postmortem femoral blood samples only.
The postmortem database included a forensic pathologist’s referral, laboratory analysis results, and information extracted from the death certificate issued by a forensic pathologist. The referral contained a brief description of the
4
Page 4 of 18
circumstances of death and the main autopsy findings. The laboratory data contained analysis results for PRG and GBP, alcohol and other drugs. Information from the final death certificate included the age and gender of the deceased, the cause of death with contributing factors according to the International Classification of Diseases (ICD-10), and the manner of death according to WHO. A case was considered as a PRG or GBP poisoning if a forensic pathologist had
ip t
determined the underlying cause of death to be poisoning with PRG or GBP as
cr
the main finding in the death certificate.
We went through all the final death certificates and forensic pathologists’
us
referrals individually to assess whether PRG or GBP use in each case had been abuse or medical use. A case was considered as a PRG or GBP abuser case if the
an
deceased was a known drug abuser; or if there were new injection marks or injection equipment near the body; or if amphetamines, cannabis, GHB or other
M
illicit drugs were found in the toxicological investigation; and necessarily, PRG or GBP was found in the toxicological investigation but had not been prescribed for the deceased for medical use. PRG and GBP were mostly administered orally. In
d
cases with drug injection, the pathologist was usually unable to specify which of
Ac ce pt e
the drugs found had been injected.
Laboratory analysis of PRG and GBP
Instrumentation
LC separation was carried out with a Shimadzu Prominence HPLC system consisting of two LC-20AD pumps, SIL-20AHT autosampler, DGU-20A5 vacuum degasser, and CTO-20A column oven (Shimadzu USA Manufacturing Inc., Canby, OR). A Gemini-NX column (100 mm x 2.1 mm; 3 µm) and C18 guard column (4 mm x 2 mm) were used in LC separation (Phenomenex, Torrance, CA). The MS analysis was performed using an AB Sciex 3200 Q TRAP® LC-MS/MS System instrument equipped with Turbo V™ source and TurboIonSpray® probe (AB Sciex, Concord, ON, Canada) in triple quadrupole mode. The software was Analyst 1.5.1.
5
Page 5 of 18
Extraction A whole blood sample (0.5 mL) was diluted with laboratory water (1:10). The diluted sample was pipetted into five test tubes (0.5 mL each), four of which were spiked with PRG (Pfizer Inc, New York, NY) or GBP (Panfarma Oy, Espoo, Finland) reference standard at concentrations of 0.1, 0.3, 1.0, and 3.0 mg/L.
ip t
Protein precipitation was carried out by vortex mixing with 1 mL of 20%
trichloroacetic acid for 1 min, and the sample was centrifuged at 4,330 g for 5
cr
min. Finally, 100 µL of the aqueous layer was transferred into an autosampler
us
vial and diluted with 900 µL of laboratory water during vortex mixing.
Liquid chromatography
an
The analytical column was stabilized at 40°C. The mobile phase gradient consisted of 10 mM ammonium acetate buffer (pH 3.2) and methanol (both
M
containing 0.1% formic acid) as follows: the methanolic proportion was held at 10% during the equilibrium time of 7 min and, after injection, increased to 30% in 4 min, then to 95% in 3 min, kept at 95% for 1 min to clean the column, and
d
finally decreased back to 10% in 2 min. The total flow rate through the column
Ac ce pt e
was 0.25 mL/min. The injection volume was 10 µL.
Mass spectrometry
The total flow from the LC was directed to the ion source without splitting. The needle voltage was 5.0 kV, declustering potential 30 V, and ion spray heater temperature 450 °C. The curtain gas (nitrogen) was set at 12 and collision gas (CAD, nitrogen) pressure in the collision cell at 5 in the Sciex control software. The nebulizer gas (nitrogen) was set at 40 psi (276 kPa) and turbo heater gas (nitrogen) at 60 psi (414 kPa). The target ions for PRG were m/z 160/142 and m/z 160/97, and the corresponding values for collision energy were 10 eV and 20 eV, respectively. For GBP, the target ions were m/z 172/154 and m/z 172/137, and the corresponding value for collision energy was 20 eV. A dwell time of 80 ms was used for the transitions.
Quantification
6
Page 6 of 18
To compensate for the effects of the sample matrix, quantification was based on the standard addition method, i.e. the peak areas were plotted against added drug concentrations using linear regression. The concentration of PRG or GBP was calculated from the point at which the extrapolated linear regression line crosses the concentration axis at zero peak area.
ip t
Validation
The lower limit of quantification (LOQ) for PRG and GBP was 0.1 mg/L,
cr
established as the lowest concentration fulfilling the following criteria:
inaccuracy (bias) from the calibration curve less than 20%, precision (RSD) less
us
than 20%, and symmetrical peak shape.
an
For determination of the linear range, the absolute peak area was plotted against drug concentration using the following criteria: linear regression with a
M
correlation coefficient (r2) higher than 0.997, bias from the calibration curve less than 20% for all individual calibration points, and precision better than 20%. The linear range for PRG was from 0.1 mg/L to 10 mg/L with r2 of 0.9995. For
Ac ce pt e
d
GBP, the linear range was from 0.1 mg/L to 10 mg/L with r2 of 0.9991.
Precision and bias were calculated as average values over the linear range. Spiked blood samples were prepared in two replicates, and the linear range was covered by five concentration levels. Precision was expressed as intra-day RSD (%) and inaccuracy as average bias (%) from the corresponding theoretical value at each concentration level within the linear range. Precision and bias for PRG within the linear range were 4% and 5%, respectively. For GBP, the corresponding figures were 4% and 4%, respectively.
The matrix effects were first estimated from drug standard curves from 18 authentic autopsy blood samples [11]. The variability of the standard curve slopes indicated 38% and 12% matrix effects for PRG and GBP, respectively. Secondly, six authentic blood samples were spiked in two replicates with drug concentrations of 1 and 5 mg/L. The samples were known to be negative for PRG and GBP but eventually contained other drugs. The precision of PRG and GBP
7
Page 7 of 18
concentrations within samples was 1-12% and 1-18%, respectively, calculated as the average RSD of the replicates. The precision between samples for PRG and GBP was 8-10% and 5-14%, respectively, calculated from the averaged concentrations. The similar precision within and between samples indicated no marked effect from matrix variation in terms of ionization and extraction
ip t
efficiency [12].
cr
Results
us
During the two-year study period 2010–2011, PRG was found in 316
postmortem cases (Table 1), which comprises 2.3% of all toxicologically
an
investigated medico-legal cases. A total of 48.1% of the PRG-positive cases were attributed to drug abuse. PRG poisonings, in which PRG was the main toxicological finding, represented 10.1% of all PRG cases. GBP was found in 43
M
postmortem cases, which comprises 0.31% of all toxicologically investigated medico-legal cases, and 18.6% of GBP-positive cases were attributed to drug
d
abuse. GBP poisonings, in which GBP was the main toxicological finding,
Ac ce pt e
represented 4.7% of all GBP cases. In the abuse cases the most common manner of death was accident, including accidental poisonings; in the other cases the cause was disease. (Table 1)
Figure 1 shows the number of all PRG and GBP cases, abuse cases, and fatal PRG and GBP poisonings classified according to age groups. Figure 2 illustrates the number of concomitant alcohol and opioid findings in the PRG and GBP cases. Opioids were found in 91.4% of the PRG and 87.5% of the GBP abuse cases. There were no cases with PRG or GBP alone without any other drugs. In 7.9% of the PRG abuse cases neither alcohol nor opioids were detected, but other central nervous system acting drugs such as benzodiazepines or other psycholeptics were found in every case. Alcohol and opioids or both were found in every GBP abuse case. Alcohol with or without opioids was found in 16.4% of the PRG and 37.5% of the GBP abuse cases. In other than abuse cases, alcohol was found in 22.0% of the PRG and 17.1% of the GBP cases. In PRG poisonings,
8
Page 8 of 18
alcohol was found in 15.6% of the cases. In GBP poisonings there were no alcohol findings.
Figure 3 illustrates the number of drug abuse and other cases when PRG and GBP
ip t
were found together with opioids.
Case reports
cr
“Only PRG”
A 31-year-old male with suicidal intentions had taken a variety of medications
us
during the night. He had no valid PRG prescription. He was found dead the next day. The following were found in his postmortem blood: PRG 110 mg/L,
an
oxazepam 0.30 mg/L, temazepam 0.15 mg/L, traces of quetiapine and levomepromazine, and ethanol 0.24‰. The concentration of PRG was more than 13 times above the upper limit of the therapeutic range, whereas the
M
concentrations of oxazepam and temazepam were within therapeutic ranges. The underlying cause of death was classified as PRG poisoning. Coronary disease
Ac ce pt e
d
was considered a contributory factor. The manner of death was suicide.
“PRG + OPI”
A 26-year-old male, a known abuser of buprenorphine and amphetamine, had taken large amounts of various medicinal substances during the night. He lacked a PRG prescription. He was found dead the next day. The following substances were found in his postmortem blood: PRG 48 mg/L, buprenorphine 0.86 g/L, norbuprenorphine 1.2 g/L, temazepam 0.75 mg/L, nordazepam 0.63 mg/L, diazepam 0.17 mg/L, oxazepam 0.17 mg/L, and traces of tetrahydrocannabinol (THC). The concentrations of buprenorphine and benzodiazepines were within the therapeutic ranges but the concentration of PRG was nearly 6 times above the upper limit of the therapeutic range. Findings in postmortem urine were buprenorphine 120 g/L, norbuprenorphine 56 g/L and THC 46 g/L. The underlying cause of death was classified as PRG, buprenorphine and temazepam poisoning with PRG as the main finding. The manner of death was accident.
9
Page 9 of 18
Discussion We have shown that 48% of the PRG findings in toxicologically investigated medico-legal autopsy cases in Finland were associated with PRG abuse. For GBP, however, only 19% was abuse. Despite the high rate of postmortem toxicological
ip t
assessments in Finland, these figures cannot be directly extrapolated to the general population, as drug abuse is more commonly found in subjects ending up in medico-legal investigations. Still, the extent of PRG abuse in our material
us
cr
reflects the situation among the living.
PRG consumption in Finland during 2010–2011 was on average 4.6 DDD/1000 inhabitants/day, and GBP consumption 1.0 DDD/1000 inhabitants/day [13],
an
which indicates that PRG was used 4.6 times more frequently than GBP. In our data, the ratio of PRG to GBP findings in postmortem cases was 7.3, indicating
living users.
M
that PRG was proportionally more frequent in postmortem cases than among
d
This study showed that 91% of the deceased PRG abusers and 88% of the
Ac ce pt e
deceased GBP abusers had combined PRG with opioids. In a therapeutic context, PRG or GBP and opioids can be administered concomitantly in severe pain conditions. The abuse potential of PRG has previously been shown to be particularly high among opioid addicts [4,14,15] and other substance-dependent patients [2,16,17]. On the grounds of its benzodiazepine-like effects, its potential to cause euphoria, and its booster effect with opioids, PRG has been classified as a Schedule V controlled substance [18]. However, our study is the first to reveal the frequency of fatal outcome following simultaneous PRG or GBP and opioid abuse. GBP abuse has also been associated with opioid abuse. In a previous postmortem study from Scotland, in 75% of the GBP cases morphine or methadone were detected, which, according to the researchers, implied possible opioid dependence and concomitant abuse of GBP [8].
PRG and GBP abusers were significantly younger (median 30 years) than other users of these substances (median 58 years) (Figure 1). All the fatal PRG
10
Page 10 of 18
poisonings occurred in age groups under 50 years. A similar age structure is typical of fatal opioid poisonings in Finland [19-21].
When the maximum recommended dose (600 mg/day) is not exceeded, steady state PRG concentrations range from 2.8 to 8.3 mg/L in plasma [22]. In our study, the median PRG concentration in the postmortem blood found in the
ip t
abuse cases (15 mg/L) was over the therapeutic range, while the median
concentration in other PRG cases (5.8 mg/L) was within the range. In a previous
cr
postmortem study, a median PRG concentration of 5.18 mg/L in all PRG-positive cases was reported [4]. In a case report describing an attempted PRG self-
us
poisoning with recovery, the measured PRG serum concentrations were 66.5 mg/L at 3 hours post-injection, 51.2 mg/L at 7 h, and 15.2 mg/L at 27 h [23]. In a
an
case report on attempted self-poisoning with PRG and lamotrigin, the measured serum PRG was 60 mg/L during the first day and 15 mg/L on the second day
M
[24]. In another suicide attempt, the serum PRG was 20.8 mg/l [25]. One PRGdependent patient had a PRG level as high as 29 mg/L on arrival at a withdrawal clinic [16]. It is not clear how many of these patients would have survived
d
without intensive treatment. In apprehended drivers, the median PRG serum
Ac ce pt e
concentration was 6.2 mg/L (range 0.2-111.6 mg/L), and in nearly 50% of these cases the serum concentration was above the typical therapeutic range [6]. GBP plasma concentrations with therapeutic doses generally range from 2.0 to 6.0 mg/L in plasma, although higher concentrations can be tolerated [26]. The postmortem median blood concentration of GBP was above the threshold of 6 mg/L in both abuse (12 mg/L) and other cases (8.3 mg/L). A fatal GBP intoxication with a postmortem concentration of 88 mg/L has been reported [27], and a metaxalone and GBP intoxication with 24 mg/L GBP [28]. In a case series of 20 survived GBP mono-intoxications there was a median GBP concentration of 18.0 mg/L (range 0.9-83 mg/L) in the group of eight asymptomatic patients, and a median of 19.5 mg/L (range 1-48 mg/L) in the group of 12 symptomatic patients who survived without medical care [29].
11
Page 11 of 18
In our material, the median PRG and GBP concentrations in postmortem blood were generally lower than the previously published serum concentrations in patients who had survived poisoning. However, there were no PRG or GBP mono-intoxications in our material; all cases included other CNS acting drugs. In Finland, drug abusers typically use several drugs, and mono-intoxications are therefore very rare [30]. According to pathologists’ referrals and death
ip t
certificates, deaths from PRG and GBP poisonings are usually delayed: for
example, the patient had gone to sleep and was found dead the next day, hence
cr
our postmortem concentrations are usually not peak concentrations. As to the mechanism of toxicity, profound CNS depression with possible respiratory
us
failure could cause overdose-related deaths with PRG and GBP, particularly when coadministration with opioids occurs. Cardiac complications and serotonin
an
syndrome with opioids have also been reported [31,32].
M
PRG has been studied in the treatment of alcohol-dependent patients [33]. In our study, concomitant alcohol and PRG use in PRG abuse cases (16%) and in fatal PRG poisonings (16%) was less frequent than in other PRG cases (22%). This
d
observation suggests that PRG is not widely abused with alcohol, and
Ac ce pt e
consequently this may have implications for therapy.
Conclusions
In our postmortem material, PRG was a more frequent finding than GBP, the difference being greater than could be predicted from their sales. Nearly half of the deceased with a positive PRG finding were drug abusers, whereas less than a fifth of GBP use was attributed to abuse. In all PRG- and GBP-related cases other drugs were also found. PRG was most commonly abused together with opioids. Even though some patients can survive with treatment after very high PRG concentrations, our study shows that PRG abuse can be fatal, especially when combined with opioids. Despite appropriate therapeutic indications, there is no longer any doubt that PRG, and to a lesser extent GBP, should be considered and classified as a benzodiazepine-like drug with considerable abuse potential. 12
Page 12 of 18
Acknowledgements Margareeta Häkkinen received a research grant for this study from the Finnish
ip t
Foundation for Alcohol Studies.
References
Ac ce pt e
d
M
an
us
cr
[1] B. Bandelow, D. Wedekind, T. Leon, Pregabalin for the treatment of generalized anxiety disorder: a novel pharmacologic intervention , Expert Rev Neurother. 7 (2007) 769-781. [2] S. Schwan, A. Sundström, E. Stjernberg, E. Hallberg, P. Hallberg, A signal for an abuse liability for pregabalin ─ results from the Swedish spontaneous adverse drug reaction reporting system, Eur J Clin Pharmacol. 66 (2010) 947-953. [3] M. Gahr, R.W. Freudenmann, C. Hiemke, M.A. Kölle, C. Schönfeldt-Lecuona, Pregabalin abuse and dependence in Germany: results from a database query, Eur J Clin Pharmacol. 69 (2013) 1335-1342. [4] S. Lottner-Nau, B. Övgüer, L.D. Paul, M. Graw, H. Sachs, G. Roider, Abuse of pregabalin ─ results of the postmortem toxicology from 2010 to 2012, Toxichem Krimtech. 80 (2013) 339-342. [5] M. Nomoto, T. Tsujii, W. Kyaw, N. Nishikawa, H. Iwaki, M. Nagai, PP033 ─ Effects of pregabalin on driving, Clin Ther. 35 (2013) e28. [6] P. Kriikku, J. Rintatalo, L. Wilhelm, J. Hurme, J. Kramer, I. Ojanperä, Pregabalin serum levels in apprehended drivers, Forensic Sci Int. (In press). [7] F. Schifano, S. D'Offizi, M. Piccione, O. Corazza, P. Deluca, Z. Davey, et al., Is there a recreational misuse potential for pregabalin? Analysis of anecdotal online reports in comparison with related gabapentin and clonazepam data. Psychother Psychosom. 80 (2011) 118-122. [8] B. Smith, C. Higgins, A. Baldacchino, B. Kidd, J. Bannister, Substance misuse of gabapentin, Brit J Gen Pract. 62 (2012) 406-407. [9] H.N. Bockbrader, D. Wesche, R. Miller, S. Chapel, N. Janiczek, P. Burger, A comparison of the pharmacokinetics and pharmacodynamics of pregabalin and gabapentin, Clin Pharmacokinet. 49 (2010) 661-669. [10] E. Vuori, I. Ojanperä, J. Nokua, R.L. Ojansivu, Fatal poisonings in Finland during 2005–2007 (In Finnish; English summary) , Finnish Medical Journal. 64 (2009) 3187-3195. [11] B.K. Matuszewski, Standard line slopes as a measure of a relative matrix effect in quantitative HPLC-MS bioanalysis, J Chromatogr B Analyt Technol Biomed Life Sci. 830 (2006) 293-300. [12] P. Heikman, M. Häkkinen, M. Gergov, I. Ojanperä, Urine naloxone concentration at different phases of buprenorphine maintenance treatment, Drug Test Anal. 6 (2014) 220-225. [13] Finnish Statistics on Medicines 2011, Finnish Medicines Agency FIMEA and Social Insurance Institution, Helsinki, 2012. [14] M. Grosshans, T. Lemenager, C. Vollmert, N. Kämmerer, R. Screiner, J. Mutschler, et al., Pregabalin abuse among opiate addicted patients, Eur J Clin Pharmacol. 69 (2013) 2021-2025.
13
Page 13 of 18
Ac ce pt e
d
M
an
us
cr
ip t
[15] F.A. Filipetto, C.P. Zipp, J.S. Coren, Potential for pregabalin abuse or diversion after past drug-seeking behavior, J Am Osteopath Assoc. 110 (2010) 605-607. [16] M. Grosshans, J. Mutschler, D. Hermann, O. Klein, H. Dressing, F. Kiefer, et al., Pregabalin abuse, dependence, and withdrawal: a case report, Am J Psychiatry. 167 (2010) 869. [17] M. Gahr, B. Franke, R.W. Freudenmann, M.A. Kölle, C. Schönfeldt-Lecuona, Concerns about pregabalin: further experience with its potential of causing addictive behaviors, J Addict Med. 7 (2013) 147-149. [18] M.L. Blommel, A.L. Blommel, Pregabalin: An antiepileptic agent useful for neuropathic pain, Am J Health-Syst Pharm. 64 (2007) 1475-1482. [19] M. Häkkinen, T. Launiainen, E. Vuori, I. Ojanperä, Benzodiazepines and alcohol are associated with cases of fatal buprenorphine poisoning, Eur J Clin Pharmacol. 68 (2012) 301-309. [20] M. Häkkinen, P. Heikman, I. Ojanperä, Parenteral buprenorphine-naloxone abuse is a major cause of fatal buprenorphine-related poisoning, Forensic Sci Int. 232 (2013) 11-15. [21] K.W. Simonsen, P.T. Normann, G. Ceder, E. Vuori, S. Thordardottir, G. Thelander, et al., Fatal poisoning in drug addicts in the Nordic countries in 2007, Forensic Sci Int. 207 (2011) 170-176. [22] D. Berry, C. Millington, Analysis of pregabalin at the therapeutic concentrations in human plasma/serum by reversed-phase HPLC, Ther Drug Monit. 27 (2005) 451-456. [23] D.M. Wood, D.J. Berry, G. Glover, J. Eastwood, P.I. Dargan, Significant pregabalin toxicity managed with supportive care alone, J Med Toxicol. 6 (2010) 435-437. [24] A.J. Braga, K. Chidley, Self-poisoning with lamotrigine and pregabalin, Anaesthesia. 62 (2007) 524-527. [25] C. Miljevic, C. Crnobaric, S. Nikolic, D. Lecic-Tosevski, A case of pregabalin intoxication, Psychiatrike. 23 (2012) 162-165. [26] M. Schulz, S. Iwersen-Bergmann, H. Andresen, A. Schmoldt, Therapeutic and toxic blood concentrations of nearly 1,000 drugs and other xenobiotics, Crit Care. 16 (2012) R136. [27] O. Middleton, Suicide by gabapentin overdose, J Forensic Sci. 56 (2011) 1373-1375. [28] K.A. Moore, B. Levine, D. Fowler, A fatality involving metaxalone, Forensic Sci Int. 149 (2005) 249-251. [29] W. Klein-Schwartz, G. Shepherd, S. Gorman, B. Dahl, Characterization of gabapentin overdose using a poison center case series, J Toxicol Clin Toxicol. 41 (2003) 11-15. [30] I. Onyeka, H. Uosukainen, M.J. Korhonen, C. Beynon, J.S. Bell, K. Ronkainen, et al., Sociodemographic characteristics and drug abuse patterns of treatmentseeking illicit drug abusers in Finland, 1997-2008: the Huuti study, J Addict Dis. 31 (2012) 350-362. [31] G. Erdoğan, D. Ceyhan, S. Güleç, Possible heart failure associated with pregabalin use: case report, Agri. 2 (2011) 80-83. [32] H.K. Song, Serotonin syndrome with perioperative oxycodone and pregabalin, Pain Physician. 16 (2013) E632-E633.
14
Page 14 of 18
[33] R. Guglielmo, G. Martinotti, M. Clerici, L. Janiri, Pregabalin for alcohol dependence: a critical review of the literature, Adv Ther. 29 (2012) 947-957. Table Table 1. Characteristics of postmortem cases positive for pregabalin (PRG) or gabapentin (GBP) in the abuse and other use groups.
a
152 (48.1) 8 (18.6)
29 (19.1) 1 (12.5)
113 (74.3) 5 (62.5)
30 (18-66) 30 (24-47)
164 (51.9) 35 (81.4)
3 (1.8) 1 (2.9)
36 (22.0) 5 (14.3)
58 (19-92) 58 (24-94)
Men, N (%)
Concentrationb, median (range), mg/L
ip t
Age, median (range)
cr
Accidents, N (%)
127 (83.6) 6 (75.0)
us
Poisoningsa, N (%)
95 (57.9) 24 (68.6)
15 (0.43-110) 12 (0.62-45) 5.8 (0.28-110) 8.3 (2.7-93)
an
Abuse PRG GBP Other use PRG GBP
Cases, N (%)
PRG or GBP as the main finding in fatal poisoning concentration in femoral blood
Ac ce pt e
d
Figures
M
b Postmortem
15
Page 15 of 18
ip t cr us an
M
Figure 1. Number of all pregabalin (PRG) and gabapentin (GBP) positive cases,
Ac ce pt e
Finland in 2010–2011.
d
abuse cases and fatal poisonings in different age groups in postmortem cases in
16
Page 16 of 18
ip t cr us
Figure 2. Number of PRG and GBP positive cases with an alcohol (ALC) and opioid (OPI) finding in the abuser group and in the other cases group. All the
an
“Only PRG” or “Only GBP” cases also included other CNS acting drugs besides
Ac ce pt e
d
M
PRG or GBP.
17
Page 17 of 18
ip t cr us
Figure 3. Number of PRG- and GBP-positive cases with an opioid finding in the
Ac ce pt e
d
M
an
abuser group compared with the other PRG and GBP cases.
18
Page 18 of 18
S0379-0738(14)00181-9 http://dx.doi.org/doi:10.1016/j.forsciint.2014.04.028 FSI 7587
To appear in:
FSI
Received date: Revised date: Accepted date:
16-2-2014 15-4-2014 27-4-2014
Please cite this article as: Margareeta H¨akkinen, Erkki Vuori, Eija Kalso, Merja Gergov, Ilkka Ojanper¨a, Profiles of pregabalin and gabapentin abuse by postmortem toxicology, Forensic Science International http://dx.doi.org/10.1016/j.forsciint.2014.04.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Profiles of pregabalin and gabapentin abuse by postmortem toxicology
ip t
Margareeta Häkkinena*, Erkki Vuoria, Eija Kalsob, Merja Gergova, Ilkka Ojanperäa Hjelt Institute, Department of Forensic Medicine, Kytösuontie 11, P.O. Box 40, FI-00014 University of Helsinki, Helsinki, Finland b Institute of Clinical Medicine, University of Helsinki, and Pain Clinic, Helsinki University Central Hospital, Helsinki, Finland
us
cr
a
* Corresponding author. Tel.: +358 414354313; fax: +358 919127518. E-mail address: [email protected] (M. Häkkinen).
an
Abstract Pregabalin (PRG) and gabapentin (GBP) are used in the treatment of neuropathic
M
pain and epilepsy, and PRG also in generalized anxiety disorder. There is increasing evidence that PRG possesses considerable abuse potential. PRG may have a higher addiction potential than GBP due to its rapid absorption and faster
d
onset of action. Our objective is to estimate the proportion of all PRG- and GBP-
Ac ce pt e
related fatalities attributable to PRG and GBP abuse. We investigated all medicolegal death cases in Finland in which PRG or GBP was found in postmortem toxicology during 2010–2011. PRG was found in 316 cases and GBP in 43 cases. Drug abuse was associated with 48.1% of the PRG and 18.6% of the GBP findings. PRG poisoning accounted for 10.1% of all PRG cases and GBP poisoning for 4.7% of all GBP cases. In the drug abuser cases, PRG poisoning represented 19.1%, and GBP poisoning 12.5%. The median blood concentration of PRG was 15 mg/L in the abuser group and 5.8 mg/L in the other cases. For GBP, these concentrations were 12 mg/L and 8.3 mg/L, respectively. In the PRG abuser group, 91.4% of cases showed concomitant opioid use, while in the rest of these cases neither alcohol nor opioids were detected, but other central nervous system acting drugs were found in each abuser case. In the GBP abuser group, 87.5% of cases showed concomitant opioid use. PRG abuse with high doses is increasingly common and can be fatal when combined with opioids.
1
Page 1 of 18
Keywords: pregabalin; gabapentin; drug abuse; postmortem toxicology; fatal poisoning; blood drug concentration
Introduction Pregabalin (PRG) is a gamma-aminobutyric acid (GABA) analog used for treating
ip t
neuropathic pain and partial seizures. In the European Union, PRG is also
approved for treating generalized anxiety disorder, and in the United States for
cr
fibromyalgia. The recommended daily dose of PRG is 150–600 mg divided into two or three smaller doses, and the defined daily dose (DDD) by the World
us
Health Organisation (WHO) is 300 mg. Gabapentin (GBP), like PRG, is a GABA analog used for neuropathic pain and epilepsy. The recommended daily dose of
an
GBP is 900–3600 mg divided into three smaller doses, and the DDD is 1800 mg.
Although in clinical trials PRG has been found to produce euphoria as an adverse
M
drug reaction, it was considered to have low potential for abuse and dependence, unlike other classes of medications used for the treatment of generalized anxiety
d
disorder [1]. However, in recent years there has been increasing concern about
Ac ce pt e
the abuse potential of PRG. Swedish researchers [2] found that reports to the national register of PRG abuse and addiction started to rise significantly in the fourth quarter of 2008, and concluded that pregabalin is likely to have abuse potential.
Similarly in Germany [3], researchers have since 2008 identified an increasing number of cases of PRG abuse or dependence reported to the national database. The mean daily PRG dosage was as high as 1424 mg, and in about one-third of patients withdrawal symptoms subsequent to discontinuation of PRG were reported. Male sex and a history of polytoxicomania were suggested as possible risk factors for the development of addictive behavior related to PRG [3]. In a German study from 2010–2012, the number of postmortem cases with a PRG finding increased from 10 to 33 within a two-year period. Among a subgroup of drug-dependent individuals, the number of PRG cases increased proportionally more, from 4 to 26 cases, within the same two-year period [4].
2
Page 2 of 18
In a driving simulation experiment, PRG was shown to cause mild effects on some components of driving performance but without serious central nervous system side effects [5]. In Finnish apprehended drivers, PRG was commonly found at concentrations higher than therapeutic together with other psycho-
ip t
active drugs, suggesting drug abuse [6]. An international team of researchers [7] analyzed both anecdotal online reports
cr
of PRG and GBP misuse and their online availability for purchase during 2008–
drug for recreational purposes to achieve specific
us
2010. Based on Internet searches, PRG was described as an ideal psychotropic
alcohol/gammahydroxybutyrate (GHB)/benzodiazepine-like effects mixed with
an
euphoria, to achieve entactogenic feelings and dextromethorphan-like disassociation, and to cope with opioid withdrawal. GBP had similar sedative,
M
dissociative and psychedelic effects but, according to one user, PRG was more desirable in recreational use than GBP [7].
d
Concern over GBP abuse has risen in Scotland, where the number of GBP
Ac ce pt e
prescriptions has increased exponentially since 2002. GBP abuse has also been noticed in the substance misuse service. Individuals abusing GBP describe experiences such as euphoria, improved sociability, relaxation and a marijuanalike “high” [8].
PRG may have a higher addiction potential than GBP due to its rapid absorption and faster onset of action. PRG has higher bioavailability than GBP [9]. Unlike GBP, PRG exhibits linear uptake without transporter saturation at therapeutic dosages, whereas the bioavailability of GBP drops as the dosage increases. In treating postherpetic neuralgia and epilepsy, PRG might be more effective than GBP [9]. PRG and GBP have no clinically relevant drug interactions. More than 90% of PRG is excreted unmetabolized. Renal failure increases PRG and GBP toxicity since both are eliminated renally [9].
3
Page 3 of 18
Despite the apparent abuse potential of PRG, no comprehensive comparative data on the profile of either PRG or GBP has been published based on findings from postmortem cases. In Finland, PRG abuse was revealed as long ago as 2007 by postmortem toxicology [10], and the phenomenon has since escalated. Our objective is to estimate the proportion of all PRG- and GBP-related fatalities attributable to PRG and GBP abuse during 2010–2011 in Finland, taking
ip t
advantage of the high medico-legal autopsy rate and the comprehensive
postmortem toxicology database generated. We further analyze the data in terms
cr
of age, gender, fatal poisonings, alcohol and opioid findings, manner of death and postmortem PRG and GBP blood concentration. Additionally, we present two
us
typical case reports concerning PRG deaths: a suicidal PRG poisoning with only small amounts of other drugs, and an accidental PRG poisoning involving
d
Data collection
M
Material and methods
an
buprenorphine and other drugs.
Ac ce pt e
Our primary data consisted of all recorded deaths in Finland in which a comprehensive postmortem toxicological analysis was performed at the Hjelt Institute, Department of Forensic Medicine during 2010–2011. All postmortem toxicology samples taken at autopsy in Finland are analyzed in our laboratory. During 2010–2011 there were 101,472 deaths in Finland, and a medico-legal autopsy was performed in 22,421 cases (22.1% of all deaths). Toxicological analyses were performed in 13,766 cases. From this material, the cases with a confirmed positive PRG or GBP finding were included in this study. All cases with a PRG or GBP finding from a postmortem sample, such as blood or urine, were recorded as PRG- or GBP-related cases. PRG and GBP concentration data were collected from postmortem femoral blood samples only.
The postmortem database included a forensic pathologist’s referral, laboratory analysis results, and information extracted from the death certificate issued by a forensic pathologist. The referral contained a brief description of the
4
Page 4 of 18
circumstances of death and the main autopsy findings. The laboratory data contained analysis results for PRG and GBP, alcohol and other drugs. Information from the final death certificate included the age and gender of the deceased, the cause of death with contributing factors according to the International Classification of Diseases (ICD-10), and the manner of death according to WHO. A case was considered as a PRG or GBP poisoning if a forensic pathologist had
ip t
determined the underlying cause of death to be poisoning with PRG or GBP as
cr
the main finding in the death certificate.
We went through all the final death certificates and forensic pathologists’
us
referrals individually to assess whether PRG or GBP use in each case had been abuse or medical use. A case was considered as a PRG or GBP abuser case if the
an
deceased was a known drug abuser; or if there were new injection marks or injection equipment near the body; or if amphetamines, cannabis, GHB or other
M
illicit drugs were found in the toxicological investigation; and necessarily, PRG or GBP was found in the toxicological investigation but had not been prescribed for the deceased for medical use. PRG and GBP were mostly administered orally. In
d
cases with drug injection, the pathologist was usually unable to specify which of
Ac ce pt e
the drugs found had been injected.
Laboratory analysis of PRG and GBP
Instrumentation
LC separation was carried out with a Shimadzu Prominence HPLC system consisting of two LC-20AD pumps, SIL-20AHT autosampler, DGU-20A5 vacuum degasser, and CTO-20A column oven (Shimadzu USA Manufacturing Inc., Canby, OR). A Gemini-NX column (100 mm x 2.1 mm; 3 µm) and C18 guard column (4 mm x 2 mm) were used in LC separation (Phenomenex, Torrance, CA). The MS analysis was performed using an AB Sciex 3200 Q TRAP® LC-MS/MS System instrument equipped with Turbo V™ source and TurboIonSpray® probe (AB Sciex, Concord, ON, Canada) in triple quadrupole mode. The software was Analyst 1.5.1.
5
Page 5 of 18
Extraction A whole blood sample (0.5 mL) was diluted with laboratory water (1:10). The diluted sample was pipetted into five test tubes (0.5 mL each), four of which were spiked with PRG (Pfizer Inc, New York, NY) or GBP (Panfarma Oy, Espoo, Finland) reference standard at concentrations of 0.1, 0.3, 1.0, and 3.0 mg/L.
ip t
Protein precipitation was carried out by vortex mixing with 1 mL of 20%
trichloroacetic acid for 1 min, and the sample was centrifuged at 4,330 g for 5
cr
min. Finally, 100 µL of the aqueous layer was transferred into an autosampler
us
vial and diluted with 900 µL of laboratory water during vortex mixing.
Liquid chromatography
an
The analytical column was stabilized at 40°C. The mobile phase gradient consisted of 10 mM ammonium acetate buffer (pH 3.2) and methanol (both
M
containing 0.1% formic acid) as follows: the methanolic proportion was held at 10% during the equilibrium time of 7 min and, after injection, increased to 30% in 4 min, then to 95% in 3 min, kept at 95% for 1 min to clean the column, and
d
finally decreased back to 10% in 2 min. The total flow rate through the column
Ac ce pt e
was 0.25 mL/min. The injection volume was 10 µL.
Mass spectrometry
The total flow from the LC was directed to the ion source without splitting. The needle voltage was 5.0 kV, declustering potential 30 V, and ion spray heater temperature 450 °C. The curtain gas (nitrogen) was set at 12 and collision gas (CAD, nitrogen) pressure in the collision cell at 5 in the Sciex control software. The nebulizer gas (nitrogen) was set at 40 psi (276 kPa) and turbo heater gas (nitrogen) at 60 psi (414 kPa). The target ions for PRG were m/z 160/142 and m/z 160/97, and the corresponding values for collision energy were 10 eV and 20 eV, respectively. For GBP, the target ions were m/z 172/154 and m/z 172/137, and the corresponding value for collision energy was 20 eV. A dwell time of 80 ms was used for the transitions.
Quantification
6
Page 6 of 18
To compensate for the effects of the sample matrix, quantification was based on the standard addition method, i.e. the peak areas were plotted against added drug concentrations using linear regression. The concentration of PRG or GBP was calculated from the point at which the extrapolated linear regression line crosses the concentration axis at zero peak area.
ip t
Validation
The lower limit of quantification (LOQ) for PRG and GBP was 0.1 mg/L,
cr
established as the lowest concentration fulfilling the following criteria:
inaccuracy (bias) from the calibration curve less than 20%, precision (RSD) less
us
than 20%, and symmetrical peak shape.
an
For determination of the linear range, the absolute peak area was plotted against drug concentration using the following criteria: linear regression with a
M
correlation coefficient (r2) higher than 0.997, bias from the calibration curve less than 20% for all individual calibration points, and precision better than 20%. The linear range for PRG was from 0.1 mg/L to 10 mg/L with r2 of 0.9995. For
Ac ce pt e
d
GBP, the linear range was from 0.1 mg/L to 10 mg/L with r2 of 0.9991.
Precision and bias were calculated as average values over the linear range. Spiked blood samples were prepared in two replicates, and the linear range was covered by five concentration levels. Precision was expressed as intra-day RSD (%) and inaccuracy as average bias (%) from the corresponding theoretical value at each concentration level within the linear range. Precision and bias for PRG within the linear range were 4% and 5%, respectively. For GBP, the corresponding figures were 4% and 4%, respectively.
The matrix effects were first estimated from drug standard curves from 18 authentic autopsy blood samples [11]. The variability of the standard curve slopes indicated 38% and 12% matrix effects for PRG and GBP, respectively. Secondly, six authentic blood samples were spiked in two replicates with drug concentrations of 1 and 5 mg/L. The samples were known to be negative for PRG and GBP but eventually contained other drugs. The precision of PRG and GBP
7
Page 7 of 18
concentrations within samples was 1-12% and 1-18%, respectively, calculated as the average RSD of the replicates. The precision between samples for PRG and GBP was 8-10% and 5-14%, respectively, calculated from the averaged concentrations. The similar precision within and between samples indicated no marked effect from matrix variation in terms of ionization and extraction
ip t
efficiency [12].
cr
Results
us
During the two-year study period 2010–2011, PRG was found in 316
postmortem cases (Table 1), which comprises 2.3% of all toxicologically
an
investigated medico-legal cases. A total of 48.1% of the PRG-positive cases were attributed to drug abuse. PRG poisonings, in which PRG was the main toxicological finding, represented 10.1% of all PRG cases. GBP was found in 43
M
postmortem cases, which comprises 0.31% of all toxicologically investigated medico-legal cases, and 18.6% of GBP-positive cases were attributed to drug
d
abuse. GBP poisonings, in which GBP was the main toxicological finding,
Ac ce pt e
represented 4.7% of all GBP cases. In the abuse cases the most common manner of death was accident, including accidental poisonings; in the other cases the cause was disease. (Table 1)
Figure 1 shows the number of all PRG and GBP cases, abuse cases, and fatal PRG and GBP poisonings classified according to age groups. Figure 2 illustrates the number of concomitant alcohol and opioid findings in the PRG and GBP cases. Opioids were found in 91.4% of the PRG and 87.5% of the GBP abuse cases. There were no cases with PRG or GBP alone without any other drugs. In 7.9% of the PRG abuse cases neither alcohol nor opioids were detected, but other central nervous system acting drugs such as benzodiazepines or other psycholeptics were found in every case. Alcohol and opioids or both were found in every GBP abuse case. Alcohol with or without opioids was found in 16.4% of the PRG and 37.5% of the GBP abuse cases. In other than abuse cases, alcohol was found in 22.0% of the PRG and 17.1% of the GBP cases. In PRG poisonings,
8
Page 8 of 18
alcohol was found in 15.6% of the cases. In GBP poisonings there were no alcohol findings.
Figure 3 illustrates the number of drug abuse and other cases when PRG and GBP
ip t
were found together with opioids.
Case reports
cr
“Only PRG”
A 31-year-old male with suicidal intentions had taken a variety of medications
us
during the night. He had no valid PRG prescription. He was found dead the next day. The following were found in his postmortem blood: PRG 110 mg/L,
an
oxazepam 0.30 mg/L, temazepam 0.15 mg/L, traces of quetiapine and levomepromazine, and ethanol 0.24‰. The concentration of PRG was more than 13 times above the upper limit of the therapeutic range, whereas the
M
concentrations of oxazepam and temazepam were within therapeutic ranges. The underlying cause of death was classified as PRG poisoning. Coronary disease
Ac ce pt e
d
was considered a contributory factor. The manner of death was suicide.
“PRG + OPI”
A 26-year-old male, a known abuser of buprenorphine and amphetamine, had taken large amounts of various medicinal substances during the night. He lacked a PRG prescription. He was found dead the next day. The following substances were found in his postmortem blood: PRG 48 mg/L, buprenorphine 0.86 g/L, norbuprenorphine 1.2 g/L, temazepam 0.75 mg/L, nordazepam 0.63 mg/L, diazepam 0.17 mg/L, oxazepam 0.17 mg/L, and traces of tetrahydrocannabinol (THC). The concentrations of buprenorphine and benzodiazepines were within the therapeutic ranges but the concentration of PRG was nearly 6 times above the upper limit of the therapeutic range. Findings in postmortem urine were buprenorphine 120 g/L, norbuprenorphine 56 g/L and THC 46 g/L. The underlying cause of death was classified as PRG, buprenorphine and temazepam poisoning with PRG as the main finding. The manner of death was accident.
9
Page 9 of 18
Discussion We have shown that 48% of the PRG findings in toxicologically investigated medico-legal autopsy cases in Finland were associated with PRG abuse. For GBP, however, only 19% was abuse. Despite the high rate of postmortem toxicological
ip t
assessments in Finland, these figures cannot be directly extrapolated to the general population, as drug abuse is more commonly found in subjects ending up in medico-legal investigations. Still, the extent of PRG abuse in our material
us
cr
reflects the situation among the living.
PRG consumption in Finland during 2010–2011 was on average 4.6 DDD/1000 inhabitants/day, and GBP consumption 1.0 DDD/1000 inhabitants/day [13],
an
which indicates that PRG was used 4.6 times more frequently than GBP. In our data, the ratio of PRG to GBP findings in postmortem cases was 7.3, indicating
living users.
M
that PRG was proportionally more frequent in postmortem cases than among
d
This study showed that 91% of the deceased PRG abusers and 88% of the
Ac ce pt e
deceased GBP abusers had combined PRG with opioids. In a therapeutic context, PRG or GBP and opioids can be administered concomitantly in severe pain conditions. The abuse potential of PRG has previously been shown to be particularly high among opioid addicts [4,14,15] and other substance-dependent patients [2,16,17]. On the grounds of its benzodiazepine-like effects, its potential to cause euphoria, and its booster effect with opioids, PRG has been classified as a Schedule V controlled substance [18]. However, our study is the first to reveal the frequency of fatal outcome following simultaneous PRG or GBP and opioid abuse. GBP abuse has also been associated with opioid abuse. In a previous postmortem study from Scotland, in 75% of the GBP cases morphine or methadone were detected, which, according to the researchers, implied possible opioid dependence and concomitant abuse of GBP [8].
PRG and GBP abusers were significantly younger (median 30 years) than other users of these substances (median 58 years) (Figure 1). All the fatal PRG
10
Page 10 of 18
poisonings occurred in age groups under 50 years. A similar age structure is typical of fatal opioid poisonings in Finland [19-21].
When the maximum recommended dose (600 mg/day) is not exceeded, steady state PRG concentrations range from 2.8 to 8.3 mg/L in plasma [22]. In our study, the median PRG concentration in the postmortem blood found in the
ip t
abuse cases (15 mg/L) was over the therapeutic range, while the median
concentration in other PRG cases (5.8 mg/L) was within the range. In a previous
cr
postmortem study, a median PRG concentration of 5.18 mg/L in all PRG-positive cases was reported [4]. In a case report describing an attempted PRG self-
us
poisoning with recovery, the measured PRG serum concentrations were 66.5 mg/L at 3 hours post-injection, 51.2 mg/L at 7 h, and 15.2 mg/L at 27 h [23]. In a
an
case report on attempted self-poisoning with PRG and lamotrigin, the measured serum PRG was 60 mg/L during the first day and 15 mg/L on the second day
M
[24]. In another suicide attempt, the serum PRG was 20.8 mg/l [25]. One PRGdependent patient had a PRG level as high as 29 mg/L on arrival at a withdrawal clinic [16]. It is not clear how many of these patients would have survived
d
without intensive treatment. In apprehended drivers, the median PRG serum
Ac ce pt e
concentration was 6.2 mg/L (range 0.2-111.6 mg/L), and in nearly 50% of these cases the serum concentration was above the typical therapeutic range [6]. GBP plasma concentrations with therapeutic doses generally range from 2.0 to 6.0 mg/L in plasma, although higher concentrations can be tolerated [26]. The postmortem median blood concentration of GBP was above the threshold of 6 mg/L in both abuse (12 mg/L) and other cases (8.3 mg/L). A fatal GBP intoxication with a postmortem concentration of 88 mg/L has been reported [27], and a metaxalone and GBP intoxication with 24 mg/L GBP [28]. In a case series of 20 survived GBP mono-intoxications there was a median GBP concentration of 18.0 mg/L (range 0.9-83 mg/L) in the group of eight asymptomatic patients, and a median of 19.5 mg/L (range 1-48 mg/L) in the group of 12 symptomatic patients who survived without medical care [29].
11
Page 11 of 18
In our material, the median PRG and GBP concentrations in postmortem blood were generally lower than the previously published serum concentrations in patients who had survived poisoning. However, there were no PRG or GBP mono-intoxications in our material; all cases included other CNS acting drugs. In Finland, drug abusers typically use several drugs, and mono-intoxications are therefore very rare [30]. According to pathologists’ referrals and death
ip t
certificates, deaths from PRG and GBP poisonings are usually delayed: for
example, the patient had gone to sleep and was found dead the next day, hence
cr
our postmortem concentrations are usually not peak concentrations. As to the mechanism of toxicity, profound CNS depression with possible respiratory
us
failure could cause overdose-related deaths with PRG and GBP, particularly when coadministration with opioids occurs. Cardiac complications and serotonin
an
syndrome with opioids have also been reported [31,32].
M
PRG has been studied in the treatment of alcohol-dependent patients [33]. In our study, concomitant alcohol and PRG use in PRG abuse cases (16%) and in fatal PRG poisonings (16%) was less frequent than in other PRG cases (22%). This
d
observation suggests that PRG is not widely abused with alcohol, and
Ac ce pt e
consequently this may have implications for therapy.
Conclusions
In our postmortem material, PRG was a more frequent finding than GBP, the difference being greater than could be predicted from their sales. Nearly half of the deceased with a positive PRG finding were drug abusers, whereas less than a fifth of GBP use was attributed to abuse. In all PRG- and GBP-related cases other drugs were also found. PRG was most commonly abused together with opioids. Even though some patients can survive with treatment after very high PRG concentrations, our study shows that PRG abuse can be fatal, especially when combined with opioids. Despite appropriate therapeutic indications, there is no longer any doubt that PRG, and to a lesser extent GBP, should be considered and classified as a benzodiazepine-like drug with considerable abuse potential. 12
Page 12 of 18
Acknowledgements Margareeta Häkkinen received a research grant for this study from the Finnish
ip t
Foundation for Alcohol Studies.
References
Ac ce pt e
d
M
an
us
cr
[1] B. Bandelow, D. Wedekind, T. Leon, Pregabalin for the treatment of generalized anxiety disorder: a novel pharmacologic intervention , Expert Rev Neurother. 7 (2007) 769-781. [2] S. Schwan, A. Sundström, E. Stjernberg, E. Hallberg, P. Hallberg, A signal for an abuse liability for pregabalin ─ results from the Swedish spontaneous adverse drug reaction reporting system, Eur J Clin Pharmacol. 66 (2010) 947-953. [3] M. Gahr, R.W. Freudenmann, C. Hiemke, M.A. Kölle, C. Schönfeldt-Lecuona, Pregabalin abuse and dependence in Germany: results from a database query, Eur J Clin Pharmacol. 69 (2013) 1335-1342. [4] S. Lottner-Nau, B. Övgüer, L.D. Paul, M. Graw, H. Sachs, G. Roider, Abuse of pregabalin ─ results of the postmortem toxicology from 2010 to 2012, Toxichem Krimtech. 80 (2013) 339-342. [5] M. Nomoto, T. Tsujii, W. Kyaw, N. Nishikawa, H. Iwaki, M. Nagai, PP033 ─ Effects of pregabalin on driving, Clin Ther. 35 (2013) e28. [6] P. Kriikku, J. Rintatalo, L. Wilhelm, J. Hurme, J. Kramer, I. Ojanperä, Pregabalin serum levels in apprehended drivers, Forensic Sci Int. (In press). [7] F. Schifano, S. D'Offizi, M. Piccione, O. Corazza, P. Deluca, Z. Davey, et al., Is there a recreational misuse potential for pregabalin? Analysis of anecdotal online reports in comparison with related gabapentin and clonazepam data. Psychother Psychosom. 80 (2011) 118-122. [8] B. Smith, C. Higgins, A. Baldacchino, B. Kidd, J. Bannister, Substance misuse of gabapentin, Brit J Gen Pract. 62 (2012) 406-407. [9] H.N. Bockbrader, D. Wesche, R. Miller, S. Chapel, N. Janiczek, P. Burger, A comparison of the pharmacokinetics and pharmacodynamics of pregabalin and gabapentin, Clin Pharmacokinet. 49 (2010) 661-669. [10] E. Vuori, I. Ojanperä, J. Nokua, R.L. Ojansivu, Fatal poisonings in Finland during 2005–2007 (In Finnish; English summary) , Finnish Medical Journal. 64 (2009) 3187-3195. [11] B.K. Matuszewski, Standard line slopes as a measure of a relative matrix effect in quantitative HPLC-MS bioanalysis, J Chromatogr B Analyt Technol Biomed Life Sci. 830 (2006) 293-300. [12] P. Heikman, M. Häkkinen, M. Gergov, I. Ojanperä, Urine naloxone concentration at different phases of buprenorphine maintenance treatment, Drug Test Anal. 6 (2014) 220-225. [13] Finnish Statistics on Medicines 2011, Finnish Medicines Agency FIMEA and Social Insurance Institution, Helsinki, 2012. [14] M. Grosshans, T. Lemenager, C. Vollmert, N. Kämmerer, R. Screiner, J. Mutschler, et al., Pregabalin abuse among opiate addicted patients, Eur J Clin Pharmacol. 69 (2013) 2021-2025.
13
Page 13 of 18
Ac ce pt e
d
M
an
us
cr
ip t
[15] F.A. Filipetto, C.P. Zipp, J.S. Coren, Potential for pregabalin abuse or diversion after past drug-seeking behavior, J Am Osteopath Assoc. 110 (2010) 605-607. [16] M. Grosshans, J. Mutschler, D. Hermann, O. Klein, H. Dressing, F. Kiefer, et al., Pregabalin abuse, dependence, and withdrawal: a case report, Am J Psychiatry. 167 (2010) 869. [17] M. Gahr, B. Franke, R.W. Freudenmann, M.A. Kölle, C. Schönfeldt-Lecuona, Concerns about pregabalin: further experience with its potential of causing addictive behaviors, J Addict Med. 7 (2013) 147-149. [18] M.L. Blommel, A.L. Blommel, Pregabalin: An antiepileptic agent useful for neuropathic pain, Am J Health-Syst Pharm. 64 (2007) 1475-1482. [19] M. Häkkinen, T. Launiainen, E. Vuori, I. Ojanperä, Benzodiazepines and alcohol are associated with cases of fatal buprenorphine poisoning, Eur J Clin Pharmacol. 68 (2012) 301-309. [20] M. Häkkinen, P. Heikman, I. Ojanperä, Parenteral buprenorphine-naloxone abuse is a major cause of fatal buprenorphine-related poisoning, Forensic Sci Int. 232 (2013) 11-15. [21] K.W. Simonsen, P.T. Normann, G. Ceder, E. Vuori, S. Thordardottir, G. Thelander, et al., Fatal poisoning in drug addicts in the Nordic countries in 2007, Forensic Sci Int. 207 (2011) 170-176. [22] D. Berry, C. Millington, Analysis of pregabalin at the therapeutic concentrations in human plasma/serum by reversed-phase HPLC, Ther Drug Monit. 27 (2005) 451-456. [23] D.M. Wood, D.J. Berry, G. Glover, J. Eastwood, P.I. Dargan, Significant pregabalin toxicity managed with supportive care alone, J Med Toxicol. 6 (2010) 435-437. [24] A.J. Braga, K. Chidley, Self-poisoning with lamotrigine and pregabalin, Anaesthesia. 62 (2007) 524-527. [25] C. Miljevic, C. Crnobaric, S. Nikolic, D. Lecic-Tosevski, A case of pregabalin intoxication, Psychiatrike. 23 (2012) 162-165. [26] M. Schulz, S. Iwersen-Bergmann, H. Andresen, A. Schmoldt, Therapeutic and toxic blood concentrations of nearly 1,000 drugs and other xenobiotics, Crit Care. 16 (2012) R136. [27] O. Middleton, Suicide by gabapentin overdose, J Forensic Sci. 56 (2011) 1373-1375. [28] K.A. Moore, B. Levine, D. Fowler, A fatality involving metaxalone, Forensic Sci Int. 149 (2005) 249-251. [29] W. Klein-Schwartz, G. Shepherd, S. Gorman, B. Dahl, Characterization of gabapentin overdose using a poison center case series, J Toxicol Clin Toxicol. 41 (2003) 11-15. [30] I. Onyeka, H. Uosukainen, M.J. Korhonen, C. Beynon, J.S. Bell, K. Ronkainen, et al., Sociodemographic characteristics and drug abuse patterns of treatmentseeking illicit drug abusers in Finland, 1997-2008: the Huuti study, J Addict Dis. 31 (2012) 350-362. [31] G. Erdoğan, D. Ceyhan, S. Güleç, Possible heart failure associated with pregabalin use: case report, Agri. 2 (2011) 80-83. [32] H.K. Song, Serotonin syndrome with perioperative oxycodone and pregabalin, Pain Physician. 16 (2013) E632-E633.
14
Page 14 of 18
[33] R. Guglielmo, G. Martinotti, M. Clerici, L. Janiri, Pregabalin for alcohol dependence: a critical review of the literature, Adv Ther. 29 (2012) 947-957. Table Table 1. Characteristics of postmortem cases positive for pregabalin (PRG) or gabapentin (GBP) in the abuse and other use groups.
a
152 (48.1) 8 (18.6)
29 (19.1) 1 (12.5)
113 (74.3) 5 (62.5)
30 (18-66) 30 (24-47)
164 (51.9) 35 (81.4)
3 (1.8) 1 (2.9)
36 (22.0) 5 (14.3)
58 (19-92) 58 (24-94)
Men, N (%)
Concentrationb, median (range), mg/L
ip t
Age, median (range)
cr
Accidents, N (%)
127 (83.6) 6 (75.0)
us
Poisoningsa, N (%)
95 (57.9) 24 (68.6)
15 (0.43-110) 12 (0.62-45) 5.8 (0.28-110) 8.3 (2.7-93)
an
Abuse PRG GBP Other use PRG GBP
Cases, N (%)
PRG or GBP as the main finding in fatal poisoning concentration in femoral blood
Ac ce pt e
d
Figures
M
b Postmortem
15
Page 15 of 18
ip t cr us an
M
Figure 1. Number of all pregabalin (PRG) and gabapentin (GBP) positive cases,
Ac ce pt e
Finland in 2010–2011.
d
abuse cases and fatal poisonings in different age groups in postmortem cases in
16
Page 16 of 18
ip t cr us
Figure 2. Number of PRG and GBP positive cases with an alcohol (ALC) and opioid (OPI) finding in the abuser group and in the other cases group. All the
an
“Only PRG” or “Only GBP” cases also included other CNS acting drugs besides
Ac ce pt e
d
M
PRG or GBP.
17
Page 17 of 18
ip t cr us
Figure 3. Number of PRG- and GBP-positive cases with an opioid finding in the
Ac ce pt e
d
M
an
abuser group compared with the other PRG and GBP cases.
18
Page 18 of 18
