Abuse liability of centrally acting non-opioid analgesics and muscle relaxants – a brief update based on a comparison of pharmacovigilance data and evidence from the literature

Abstract

There is a lack of data regarding the abuse liability of centrally acting non-opioid analgesics (NOA) and muscle relaxants (MR). A comparison of data retrieved from a German pharmacovigilance database (BfArM; accessed May 2013) and data from the literature concerning the abuse liability of NOA and MR approved in Germany was performed. The BfArM-database demonstrated cases of abuse only for clonidine and paracetamol, whereas the literature suggests evidence for an abuse potential of baclofen, clonidine, ketamine, metamizole, methocarbamol, orphenadrine, paracetamol, propyphenazone, and tizanidine. The low number of detected cases in the BfArM-database could be a result of under-reporting.

Addiction, diversion, drug safety, pain management, pharmacovigilance

Dear editor

Opioid analgesics feature a relevant potential for abuse that must be considered in pain management. Concerning the various other substance classes, such as centrally acting non-opioid analgesics and centrally acting muscle relaxants that are also frequently used in pain management there is a lack of data regarding their abuse liability. Taking into account the high prevalence of chronic pain syndromes usually necessitating pharmacologic treatment and the enormous costs and impairments induced by substance use disorders, it is particularly important to assess the abuse liability of drugs used in pain management. Indeed, there is limited evidence arising from experimental studies in human subjects and/or animal models [e.g. methocarbamol (Preston et al., 1989; Sannerud et al., 1991] and anecdotal reports [e.g. clonidine (Schaut and Schnoll, 1983) or acetaminophen (Blakley and Schilling, 2008)] for a potential for abuse of several agents frequently used in pain management. However, in this regard, the literature does not provide systematic data from a post-marketing setting, thus impeding a proper safety assessment of particular agents from a ‘naturalistic’ perspective. Here, databases of national pharmacovigilance institutions may provide valuable insights into adverse drug reactions (ADR) such as the abuse and dependence potential related to particular pharmaceutical agents. Bringing together evidence from the literature and pharmacovigilance data in a combined approach might be most effective in the assessment of the abuse liability of particular agents.

Pharmacovigilance institutions such as the German Federal Institute for Drugs and Medical Devices [Bundesinstitut für Arzneimittel und Medizinprodukte (BfArM]) Germany] record, cumulate and evaluate spontaneous reports of any ADR related to pharmaceuticals in a standardised manner (electronic documentation of reported ADR has been compiled by the BfArM since 1993). Sources of reports at the BfArM-database are clinicians, pharmaceutical companies, the Drug Commission of the German Medical Association and systematic clinical trials.

In May 2013 a query of the entire BfArM database for ADR was performed using the standardised Medical Dictionary for Regulatory Activities (MeDRA) query (SMQ; applied SMQ-term was ‘drug abuse’) in order to retrieve insights into the abuse of non-opiod analgesics (ketamine, metamizole/dipyrone, orphenadrine, paracetamol/acetaminophen, phenazone, propyphenazone, ziconotide) and centrally acting muscle relaxants (baclofen, methocarbamol, pridinol, tolperisone, tizanidine), which are approved in Germany. Further databases, such as the database of the AMSP (‘Arzneimittelsicherheit in der Psychiatrie e.V.’; www.amsp.de/doku.php), which specifically address ADR related to psychotropic pharmaceuticals, were not used due to their insufficient structures in regard to addictive disorders. Furthermore, a literature search in the databases of MEDLINE, Embase and Scopus was performed by using the search term ‘abuse’ AND ‘respective agent’ – as listed above. Articles (case reports or clinical trials/ experimental studies) that provided evidence for an abuse potential of the respective agent were included (by means of plausible case descriptions of addictive behaviours related to the respective agent or experimental studies in humans providing hints for an abuse potential.

In contrast to the literature that provides some evidence for an abuse potential of baclofen (Nasti and Brakoulias, 2011), clonidine (Schaut and Schnoll, 1983; Dy and Yates, 1996; Dennison, 2001; Schindler et al., 2013), ketamine (Ahmed and Petchkovsky, 1980; Critchlow, 2006), metamizole (dipyrone) (Abbott and Hellemans, 2000), methocarbamol (Hayes and Balster, 1989; Preston et al., 1989, 1992; Sannerud et al., 1991), orphenadrine (Shariatmadari, 1975; Mugglestone, 1985; Schifano et al., 1988), paracetamol (acetaminophen) (Nakra et al., 1973; Barker et al., 1977; Abbott and Hellemans, 2000; Colás Chacartequi et al., 2005; Blakley and Schilling, 2008), propyphenazone (Colás Chacartequi et al., 2005), and tizanidine (Zullino et al., 2003), ADR-reports from the BfArM-database suggesting an abuse potential were only identified for clonidine and paracetamol (see Table 1 for details).

Table 1

Number and characteristics of spontaneous reports of abuse of centrally acting non-opioid analgesics and muscle relaxants in Germany (BfArM data) in comparison to corresponding evidence from the literature

Centrally acting non-opioid analgesics
Agent	Number of identified reports in the BfArM-database	Characteristics of cases from the BfArM-database	Previous evidence for an abuse potential from the literature
			Evidence	Reference(s)	Number of reported cases
Clonidine	1	52-yr-old female; daily dose between 3.6 and 7.5 mg; polytoxicomania; report from 1996	+	(Schaut and Schnoll, 1983; Dy and Yates, 1996; Dennison, 2001; Schindler et al., 2013)	16
Ketamine	0	NA	+	(Ahmed and Petchkovsky, 1980; Critchlow, 2006)	2
Metamizole (dipyrone)	0	NA	+	(Abbott and Hellemans, 2000)	Not quantified
Orphenadrine	0	NA	+	(Shariatmadari, 1975; Mugglestone, 1985; Schifano et al., 1988)	3
Paracetamol (acetaminophen)	10	Mean age 44.4 ± 21.6 yr (median 44; range 14–73 yr)*; Females n = 7/70%; Mean daily dose 5243.8 ± 4092.3 mg (median 4350; range 1500–14 000)**; current abuse of other agents with abuse potential in n = 2/20%; first report 1993, last 2013	+	(Nakra et al., 1973; Barker et al., 1977; Abbott and Hellemans, 2000; Colás Chacartequi et al., 2005; Blakley and Schilling, 2008)	>2622 (including results of a CSS)
Phenazone	0	NA	−	–	NA
Propyphenazone	0	NA	+	(Colás Chacartequi et al., 2005)	Not quantified
Ziconotide	0	NA	−	–	NA
Centrally acting muscle relaxants
Baclofen	0	NA	+	(Nasti and Brakoulias, 2011)	1
Methocarbamol	0	NA	+	(Hayes and Balster, 1989; Preston et al., 1989, 1992; Sannerud et al., 1991)	14 (ES)
Pridinol	0	NA	−	–	NA
Tolperisone	0	NA	−	–	NA
Tizanidine	0	NA	+	(Zullino et al., 2003)	1

Legend: CSS = cross-sectional study; ES = experimental study; NA = not applicable; yr = year(s); + = present, − = not present; ^*= missing data on n = 1/10%; ^**= missing data on n = 3/30%.

Under-reporting of recognized ADR is a basic problem affecting pharmacovigilance data, which are based on spontaneous reports of ADR and thus, by analysing these data, generally no conclusions can be drawn regarding the true incidence and prevalence rates of a particular ADR. Nevertheless, the BfArM-database did not indicate any new safety signals concerning the abuse liability of the evaluated agents, meaning that the assessed pharmacovigilance data (originating from a post-marketing setting) are neither in contrast to nor do they exceed the evidence of the available literature or suggest that abuse of the analysed agents had become an issue of major clinical concern.

In conclusion, our combined approach using data from a German pharmacovigilance database and data gained from corresponding medical literature provides preliminary evidence that – principally based on the absolute number of identified cases in the literature (for details see Table 1) – some centrally acting non-opioid analgesics (especially acetaminophen, clonidine, ketamine and orphenadrine) and centrally acting MR (especially baclofen, methocarbamol and tizanidine) may feature a clinically relevant abuse potential that has not become a major issue in clinical practice until now.

Acknowledgments

None.

Statements of Interest

None.