DRUG EXPERIENCE
Drug Safety 7 (3): 190-199. 1992 0114-5916/92/0005-0190/$05.00/0 © Adis International Limited. All rights reserved. DRS'
Dextromethorphan
An Overview of Safety Issues
Jerzy L. Bem and Richard Peck Pharma Division, F. Hoffmann-laRoche Ltd, Basel, Switzerland
Contents J90 J9J J93 J94 J94 J95 196 J96 J96 J97 197 J97
Summary
Summary I. Spontaneous Adverse Event Reports 2. Safety in Poor Metabolisers 3. Dependence 4. Recreational Abuse 5. Overdose 5.1 Adults 5.2 Children 6. Significance of Animal Neurological Findings 7. Interaction with MAOIs 8. Risk-Benefit Considerations 9. Conclusions
Dextromethorphan is a highly effective and widely used nonopioid antitussive drug. As it has been in use for more than 30 years, a large body of clinical experience has been used to formulate a safety profile. An anthology of adverse drug events has been analysed, drawn both from published case records and a data base recording dextromethorphan-related adverse events spontaneously reported by physicians or pharmacists. The resulting safety profile indicates that adverse drug reactions are infrequent and usually not severe. The predominant symptoms are usually dose related and include neurological, cardiovascular and gastrointestinal disturbances. Particular safety concerns arise when monoamine oxidase inhibiting (MAOI) drugs and dextromethorphan are coadministered. In addition to adverse drug reactions, the safety profile of dextromethorphan is affected by episodic and sporadic abuse. In fact, abuse appeared to be the most significant hazard identified by analysis of spontaneous adverse event reporting. No evidence could be found that the well documented pharmacokinetic polymorphism observed with dextromethorphan is correlated with any clinically significant safety risk if it is used for short term treatment. In summary, the safety profile of dextromethorphan is reassuring, particularly relating to overdose in adults and children.
Dextromethorphan Safety
Dextromethorphan is the methylated dextrorotatory analogue of levorphanol. It acts on the cough centre in the medulla oblongata, raising the threshold for the cough reflex. The pharmacological properties of the drug have been extensively investigated as far back as 1953. Dextromethorphan is well absorbed from the gastrointestinal tract. It is metabolised in the liver (first-pass effect) and excreted in the urine either unchanged (see section 2 below), or as demethylated metabolites including dextrorphan, which itself has some cough-suppressing activity. Dextromethorphan does not inhibit ciliary function, thus providing a clinically significant advantage of the drug over many other cough suppressants. The superior efficacy of dextromethorphan in comparison with opioid antitussives has been demonstrated in several well controlled studies (Karttunen et al. 1987; Matthys et al. 1983). In humans dextromethorphan lacks the pharmacological characteristics of opiate alkaloids. In normal individuals at therapeutic doses dextromethorphan is devoid of analgesic, euphoriant and physical dependence-producing properties (Jasinski 1979). In animals dextromethorphan is nonaddictive, even in high experimental doses. Dextromethorphan became widely used in the 1950s and is now marketed in more than 60 countries under many different proprietary names on its own or in combination. Adverse effects with dextromethorphan have a very low, but not precisely quantifiable incidence (Reynolds 1989). Considering the widespread use of dextromethorphan, there are very few reports of toxicity at recommended concentrations. Most reports of toxicity occur at high dosages as a result of misuse, ingestion with suicidal intent, or recreational use. In addition, the evaluation of toxicity is complicated by simultaneous ingestion of other substances and by the lack of objective measurements of dextromethorphan exposure. Most of the data from clinical trials were generated in the 1950s and 1960s. The more recent trials, reviews and case reports confirm the earlier findings that adverse reactions are mild (Bickerman 1984; Matthys et al. 1983). In contrast to more
191
conventional drug safety considerations, dextromethorphan has been the subject of abuse. These episodes have occurred sporadically over the past 30 years giving rise to self-limiting localised outbreaks characterised by symptoms of overdose among a limited segment of the population. Following such abuse episodes, cases of psychological dependence have been reported, but there does not appear to be any evidence of dependence of the morphine type. Recently, the discovery of the polymorphic oxidation of debrisoquine caused a resurgence of interest in genetic factors affecting the individual response to drugs. In fact, dextromethorphan has been used as a prototype in the study of polymorphism in the metabolism of a wide variety of pharmacological substances (Perault et al. 1991). It has been suggested that dextromethorphan may provide a useful diagnostic tool to identify persons with deficiency in the debrisoquine-type of drug oxidation (Chen et al. I 990a). However, the recognition of dextromethorphan polymorphism also raises questions related to the safety of dextromethorphan in the minority of people who are poor metabolisers. In this respect, the dysphoric effects elicited by high doses (240mg) of dextromethorphan are probably manifested differently in poor metabolisers (Musacchio et al. 1989).
1. Spontaneous Adverse Event Reports A relatively small number of adverse drug events (ADEs) during the past 30 years has been collected in a data base of spontaneously reported dextromethorphan-related adverse events maintained by the Drug Safety Department at Roche. ADEs are defined to include all reports affecting safety, potency, adverse reactions, clinical side effects and drug abuse problems. The drug safety data base draws upon a worldwide network of physicians and pharmacists who are actively encouraged to report upon a wide variety of prescription and over-thecounter (OTC) pharmaceutical products. A high level of cooperation is ensured by competent personnel who are assigned to evaluate the significance and severity of reports from all sources. Ad-
192
Drug Safety 7 (3) 1992
Table I. Reports of adverse drug events associated with dextromethorphan (all preparations) received by Roche Drug Safety
between 1960 and 1989. Events are grouped by body system using the WHO classification Body system
Autonomic nervous system Cardiovascular (general) Central/peripheral nervous system Endocrine Fetal Gastrointestinal system General Heart rate and rhythm Liver and biliary system Platelet, bleeding and clotting Psychiatric Red blood cell Reproductive (male) (female) Respiratory system Skin and appendages Urinary system Vascular Vision White cell and reticuloendothelial system Total
No. of events 30 8 38 7 5 41 20 21 5 161 2 20 2 5 53 5 1 25 2
452
verse events are classified into medically defined categories comprising all cases and exposures, including long-term sequelae up to 5 years. Quality control and validation of data is done by the Drug Safety department. No deliberate selection of distinct population groups is made. Analysis of these data reveals that OTC drugs generate a smaller number of spontaneous ADE reports than prescribed drugs. However, considering the very wide usage of dextromethorphan, the number of adverse events reported over a long period indicates a good safety record of dextromethorphan. 299 cases of adverse events associated with dextromethorphan (all brands and generics) have been reported worldwide up to July 1990 (table J). In addition to the spontaneous ADE reports entered directly in the database, a substantial proportion of ADEs are reported in the scientific literature, newspapers and other sources.
Psychiatric and CNS reactions (29 and 25% of ADEs, respectively) were reported most frequently. Skin (9%), gastrointestinal (7%) and cardiovascular (5%) disorders are less common. The profile of the above ADEs reported spontaneously is compatible with that described in the standard texts. Of interest is the relatively frequent occurrence of psychiatric ADEs, most being a result of drug abuse (72 reports). 35 cases were reported in the 1960s, 36 cases in the 1970s and I case in the 1980s. Other neuropsychiatric events reported are listed in table II. From these data there is no evidence of a long term or permanent effect of dextromethorphan. Dextromethorphan drug abuse has been reported from various countries over a period of at least 30 years (table III). The most recent outbreak of abuse involved 3 cases reported from Switzerlan~ in 1990. Drawn from spontaneous reports of ADEs, the range of ingested doses giving rise to adverse manifestations varied from 30 to 900mg. Approximately half of the reports involved teenagers and another third young adults. The number of publications reporting dextroTable II. Psychoneurological ADEs reported in association with dextromethorphan. Some reports describe more than one event
Adverse event
No. of reports
Abnormal thinking Aggressive reaction Agitation, anxiety, excitation Awareness disturbances Confusion, consciousness clouded Delusions, hallUCinations Depression Disinhibition Disorientation Euphoria, stimulation
3 3 9 22 27 54 42 2 22
Fear, panic reaction Lack of drive Neurosis Personality disorders Psychotic syndrome Restlessness Schizophrenic reaction Sleep disorders
36 2 28 4 7 4 12 22 2
193
Dextromethorphan Safety
Table III. Reports of drug dependence associated with dextromethorphan (all preparations) received by Roche Drug Safety between 1960 and 1989 Country Angola Australia Austria Canada Denmark Finland Germany Netherlands Philippines Switzerland Turkey UK
1960s
1970s
1980s
3 1
20
3 4 7 2
5 4 2
3
3
1 2
US Total
35
28
methorphan abuse has diminished: 16 papers were published in the 1960s, 20 in the early 1970s, I in the late 1970s and 8 in the I 980s. Although the number of incidents is too small to permit a statistical trend analysis, no indication of an increase in dextromethorphan abuse can be detected. The decrease in spontaneously reported adverse events as a function of time after initial introduction follows a pattern which has been described for prescription drugs (Guess 1989). Table IV lists a total of 12 cases of fatal events reported in patients taking dextromethorphan. All but I were known to involve overdoses, intentional or accidental. Additional drugs were taken in 6 cases and were possibly taken in another 4 cases. In the cases where the other drugs were known the contribution of these drugs to the fatal outcome is likely. In 2 cases (I from Sweden and I from Germany) death was very likely due to an extremely high overdose (up to 3g) of dextromethorphan. The UK Committee on Safety of Medicines (CSM) has received only 51 reports of ADEs between July 1963 and July 1990 and no cases of death associated with any preparation containing dextromethorphan (CSM, personal communication). The US Food and Drug Administration
(FDA), as reported in the Federal Register in 1983, released dextromethorphan safety data and concluded that because of its low toxicity, dextromethorphan is probably the safest antitussive presently available.
2. Safety in Poor Metabolisers Wide interindividual variations in the measured blood concentrations of dextromethorphan have been known for a considerable time. A study on polymorphic metabolism of dextromethorphan was published by Pfaff et al. (1983). These results have been largely confirmed by numerous other authors (Kupfer et al. 1984, 1986; Schmid et al. 1985). Within the white European population individuals metabolise dextromethorphan rapidly (about 81 %), at an intermediate rate (about 6%) or slowly (about 13%) [Pfaff et al. 1983]. The so-called poor metaboliser (PM) phenotype is inherited as an autosomal-recessive trait and occurs in 5 to 10% of European, Australian Caucasian and North American populations (Chen et al. 1990a; Eichelbaum 1988; Meyer et al. 1989). The frequency of the PM phenotype was reported to be lower in I non-Caucasian population (Nakamura et al. 1985) This polymorphism arises from differences in the O-demethylation of dextromethorphan, based on an individual genetic disposition, associated with the alicyclic hydroxylation of debrisoquine. Metabolic polymorphism is now well recognised with several drugs which are metabolised via oxidation. Pharmacokinetic consequences were described for at least 25 commonly used drugs (Brosen & Gram 1989) including {1-blockers, tricyclic antidepressants, neuroleptics and codeine. Chen et al. (l990b) showed that the citric acid-induced cough threshold was doubled after dextromethorphan was given to PMs compared with extensive metabolisers (EMs). At the same time, the mean maximum plasma concentration of dextromethorphan was 20 times higher in PMs than in EMs. The authors suggested that PM and EM subjects might require different dosage regimens for cough suppression. However, these findings may also suggest that the pharmacodynamic difference between PM and EM
Drug Safety 7 (3) 1992
194
Table IV. Reports of fatal outcomes associated with dextromethorphan received by Roche Drug Safety between 1959 and 1990. All preparations and all sources of reports are included Year
Country
Dosage
Comedication
NR 1959 1966 1967 1967 1969 1975 1979 1987 1987 1990 1990
UK Germany Turkey Turkey Canada USA Austria Canada Sweden Sweden Hong Kong Hong Kong
NR 250mg High 450mg 130mg 300mg High Normal 450-3000mg High NR NR
NR None NR Opium MAOI, 2 drugs Alcohol, LSD, marijuana Opium, antihypertensives Antidepressants
Comments
2 years of chronic use
Suicide
Suicide Suicide or overdose NR Morphine
Abbreviations; NR '" not reported; MAOI '" monoamine oxidase inhibitor.
is less than predictable from the pharmacokinetic differences (2-fold difference versus 20-fold difference). These findings could explain, at least to some extent, why in clinical practice short term treatment with the standard dose of dextromethorphan appears to be effective and safe in both PM and EM populations. Further investigation should be directed towards elucidating more fully the clinical consequences of dextromethorphan polymorphism.
3. Dependence All available experimental and clinical data indicate that the dependence potential of dextromethorphan is very low. No chiral inversion of dextromethorphan (d-form) to the opioid levomethorphan (I-form) occurs in humans. Commercial dextromethorphan preparations are free from the I-isomer (Reynolds 1989). Addiction studies in animals (mice, rats, cats and monkeys) were all negative. It was not possible to demonstrate or to induce either morphine-like effects or physical dependence on dextromethorphan (Musacchio et al. 1989). Addiction studies in humans show little or no addiction potential of dextromethorphan. Jasinski (1977) surveyed the available epidemiological
studies of various morphine substitutes and calculated the availability and relative abuse incidence in relation to morphine. This incidence was zero for dextromethorphan for 1973 to 1975. The safety of antitussive drugs has been monitored by academic, governmental and industrial bodies since 1929 (Eddy 1973; Jasinski 1977; Martin 1977). The international experience was first summarised by the World Health Organization (WHO) in 1970. The WHO, FDA and the American Medical Association consider dextromethorphan to be a non-narcotic, nonsedative drug, presenting no public health problem and not requiring special regulation (Seventeenth Report of the WHO Expert Committee on Drug Dependence 1970). Dextromethorphan has been subject to limited and sporadic abuse, but there does not appear to be any evidence of dependence of the morphine type. McCarthy (1971) studied 20 dextromethorphan abusers and noticed that after a few weeks of abuse the popularity of this drug declined after the abusers started to experience a period of somnambulism, lethargy and ataxia.
4. Recreational Abuse Sporadic abuse of practically all cough suppressants and expectorants has been reported (McCarthy 1971). This may be in the form of a cough
Dextromethorphan Safety
suppressant alone or in combinations containing an antihistamine or pseudoephedrine. The content of alcohol in some older dextromethorphan-containing preparations, as in many other cough mixtures, is as high as 50%. In the 1960s, several cases of misuse of dextromethorphan were reported, often together with psychotomimetic drugs. Neither the number of reports nor the manner of abuse give grounds to suppose that dextromethorphan has properties leading to long-lasting addiction. In contrast, an association of abuse with psychotomimetic effects, especially of short duration, cannot be excluded (Tortella et al. 1989). The earliest reports of dextromethorphan abuse in the US date back to the 1960s. Younes (1969) described a 'kick-inducer' with a dextromethorphan-containing product among teenagers. Intentional abuse of dextromethorphan-containing preparations was studied in Utah between 1985 and 1988 by McElwee and Veltri (1990). In this large population study the authors found increased intake of dextromethorphan-containing cough syrup among teenagers, but the acute health risk appeared to be minimal. In 1990, the Pennsylvania Health Authorities held hearings on dextromethorphan abuse; a further inquiry on this subject was held by the US FDA also in 1990. Data from the National Institute on Drug Abuse consisting of reports from medical examiners and emergency rooms showed that dextromethorphan accounted for only 0.01% of all abuse accounts, ranking fortyfirst among all drugs mentioned. Data from the Drug Abuse Warning Network confirms that dextromethorphan abuse poses no acute public health dangers in the US (McElwee & Veltri 1991a,b). Until 1986 the UK Home Office Drugs Branch had not received any reports of dextromethorphan misuse (Fleming 1986). The UK CSM received no reports of abuse of dextromethorphan between July 1963 and July 1990 (CSM, personal communication). In 1986, Orrell and Campbell (1986) reported the case of a 37-year-old man with a history of abusing a wide range of drugs who had been diagnosed as schizophrenic. Over the previous year, he had abused a cough syrup containing dextro-
195
methorphan, consuming up to 2 bottles (500mg) several times a week. He was admitted after complaining of feeling anxious and aggressive. After I week on antipsychotic medication his mental state improved, but on 2 occasions he became abusive, irritable, and disruptive. A case of pure dextromethorphan hydrobromide powder abuse by a 30year-old man was reported by Fleming (1986). Euphoria and restlessness occurred about 15 min after inhaling the powder and lasted for up to 2h, followed by depression, nausea, tiredness and dizziness. There is evidence of dextromethorphan abuse in Australia before 1965 (McCarthy 1971). During 1968 and early 1969, there was a marked increase in its use among young people in and around Brisbane. The abusers reported increased perceptual awareness, altered time perception and visual hallucinations. The popularity of the drug with younger drug abusers was related mainly to its easy OTC availability. Dextromethorphan was taken mainly in mixture form, the usual dose being I bottle (100mg), often mixed to disguise the taste. Onset of the effects was within Ih, and their duration was 3 to 4h. In the 1980s, teenagers in different parts of Sweden started to abuse dextromethorphan. As a result of these reports, the Swedish National Board of Health and Welfare has enforced compulsory prescriptions by a doctor from April 1986. Dextromethorphan abuse has also been reported from several other countries including Austria, France and New Zealand (Bornstein et al. 1968; McCarthy 1971). In Switzerland, 3 cases of abuse were reported in the 1960s and 3 in the 1980s. In all of these countries, questions regarding the potential for dextromethorphan abuse have been raised by the Health Authorities. According to our data, recreational abuse remains the most important drug safety consideration associated with dextromethorphan.
5. Overdose Massive overdosage with dextromethorphan may result in respiratory depression, excitation, confusion, toxic psychosis characterised by hyper-
Drug Safety 7 (3) 1992
196
active behaviour, extreme pressure of thought and morbid auditory and visual hallucinations. However, dextromethorphan overdose is not easy to achieve: clinical safety studies with dose escalation to 330 mg administered every 6h have been tolerated with comparatively minor and rapidly reversible neurological symptoms (Steinberg & Bell 1991 ). 5.1 Adults Several cases of overdose have been reported from Sweden. The inappropriate use of dextromethorphan-containing preparations was usually combined with alcohol consumption. The accompanying symptoms, some of them severe, were not life-threatening in any patient of this group. The first 2 cases of fatal overdose with dextromethorphan were reported in 1987 in Sweden (Rammer et al. 1988). The autopsy showed no gross pathology in I of thel'e cases, and signs of pulmonary aspiration of gastric content in the second. Study of tissues and blood showed concentrations of dextromethorphan to be 2- to 3-fold higher than the concentration of its metabolite dextrorphan. The authors commented on the existence of individual variations in the metabolic rate of dextromethorphan, but they did not suggest that any of these cases was a poor metaboliser. On the basis of the blood and tissue dextromethorphan concentrations, this conclusion would appear to be appropriate. McElwee and Veltri (1990) analysed 265 cases of dextromethorphan exposure reported to the Intermountain Regional Poison Control Center (lRPCC) in Utah. In 1989, IRPCC had a catchment population of about 1.7 million people. Most exposures were accidental. Intentional exposure accounted for about 22% of cases. In the cases with known outcome, most resulted in either no effect or minor effect for the patients. None of the patients died or developed life-threatening symptoms or permanent disability from exposure. The authors concluded that there is only minor abuse of dextromethorphan, and the risk to health from those exposures is minimal.
5.2 Children A few serious dextromethorphan overdoses have been reported in children. The clinical effect of such overdoses was predominantly neurological. The cases were characterised by central excitation rather than sedation. All patients treated with naloxone responded dramatically (Boeckx 1987; Henreting et al. 1988; Katona et al. 1986; Shaul et al. 1977). The FDA reviewed all reported overdose cases (Federal Register 1983) and found no fatalities even with doses exceeding 100 times the normal dose. Up to 1983, despite widespread use of preparations containing dextromethorphan, only 33 cases of accidental overdose requiring medical intervention were known. Five of these cases have been the subject of publications. The age of the children concerned was between 10 months and 10 years. The amount taken varied from 15 to 600mg. A fatal outcome following accidental overdose with dextromethorphan alone is very unlikely. Naloxone is available as an effective antidote, and improvement in the child-proof packaging has further increased safety for children.
6. Significance of Animal Neurological Findings In various animal and in vitro models, high doses of dextromethorphan appear to have both neuroprotective and neurotoxic properties. The neuroprotective activity of the drug has been reported amongst others by Choi (1987) who demonstrated that dextromethorphan and dextrorphan can substantially reduce the neurotoxicity of glutamate in vitro. He further suggested that dextromethorphan or related compounds may eventually have clinical therapeutic utility in a variety of neurological disease states. Prince and Feeser (1988) demonstrated a protective action of dextromethorphan against cerebral infarction in a rat model of hypoxia-ischaemia. Engbar and Chase (1988) propose that dextromethorphan might prevent neurotoxicity due to long term exposure to low levels of excitotoxins, as may occur in neurological dis-
197
Dextromethorphan Safety
orders such as Huntington's disease and Alzheimer's disease. Weisand et al. (1990) reported that dextromethorphan injections produced swollen neuronal somata in the retrosplenal cortex of male rats. Evidence suggesting that other glutamate receptor antagonists may have neurotoxic effects in rats was provided by Olney et al. (1989). They demonstrated vacuolar changes in specific populations of brain neurons when the noncompetitive N-methylD-aspartate (NMDA) antagonists MK 801, phencyclidine, tiletamine and ketamine were administered subcutaneously to adult rats in relatively low doses. Anticholinergic compounds were reported to protect neurons from vacuolisation without interfering with neuroprotection from MK 801 (01!:ley et al. 1989). No clinical evidence that the histological findings in animals are relevant to human therapy with dextromethorphan exists at present. The issues of dose, duration, route of administration and metabolic differences between man and rat make the extrapolation of the animal data to humans extremely difficult. The significance of vacuolisation observed in rat brains after subcutaneous injection requires more research.
7. Interaction with MAOIs An interaction between dextromethorphan and MAOIs is well documented in the literature (Harrison et al. 1989; Rivers & Homer 1970; Shamsie & Barriga 1971). The interaction manifests itselfas a serotonergic syndrome: high blood pressure, hyperpyrexia, arrhythmia (Browne & Linter 1987) or myoclonus. The interaction may be fatal (Shamsie & Barriga 1971). The pharmacological mechanism for the interaction of dextromethorphan with MAOIs may be that of dextromethorphan blocking the neuronal reuptake of serotonin (Sinclair 1973). Lethal doses of dextromethorphan in combination with an MAOI in rabbits were very high (Sinclair 1973). Many reported cases of dextromethorphan interaction with MAOIs involve OTe preparations which also contain other potentially hazardous
components. The spontaneous report data base has obtained 1 fatality report of a case where a patient took overdoses of both dextromethorphan and an MAO!. The clinical relevance of the interaction between dextromethorphan and MAOIs remains controversial. Administration of relatively high levels of the MAOI toloxatone failed to significantly alter the metabolic ratio of dextromethorphan in humans (Perault et al 1991). Similarly, the use of dextromethorphan and MAOIs in clinical practice has been reported without incident (Bazire 1987). Nevertheless, in consideration of the possible risk, dextromethorphan-containing formulations should not be used under ordinary circumstances in conjunction with MAOI therapy.
8. Risk-Benefit Considerations Unproductive, persistent cough is not uncommon and often distressing, but the symptoms are rarely serious enough to justify the use of other than very safe drugs. In comparative clinical trials dextromethorphan has been shown to be superior or equal to other antitussives (Matthys et al. 1983). All the efficacious antitussives cause adverse effects. For comparison, other commonly used antitussives have their own safety considerations. Opiates in general have more central depression and addiction potential than dextromethorphan. Antihistamines and anticholinergic agents can, in larger doses, inhibit the secretion of the bronchial glands and also will cause central depression. Ammonium chloride can occasionally cause clinically significant acidosis. Iodide is not safe in certain patients and can cause allergic reactions. Sympathomimetics are associated with a number of side effects. Lastly, benzocaine can cause allergic reactions or sensitise a patient.
9. Conclusions Dextromethorphan has proven itself in long term clinical use to be a highly effective nonopioid antitussive drug. From an analysis of adverse drug events drawn from published case records and
198
spontaneously reported events, dextromethorphan has a wide margin of safety. Dextromethorphanrelated adverse drug reactions are infrequent and usually not severe. Overdose is rare both in adults and children. However, serious concerns surround the probable interaction between MAO! drugs and dextromethorphan. In addition to adverse drug reactions, the safety profile of dextromethorphan is affected by outbreaks of abuse: this constituted the most significant hazard identified by analysis of spontaneous reporting. Pharmacokinetic polymorphism does not appear to present a clinically significant safety risk if dextromethorphan is used for short term treatment. Given the very large volume of dextromethorphan use and the comparatively small number of serious ADEs reviewed here, the risk is low. In conclusion, the safety profile of dextromethorphan is extremely favourable.
References Bazire s. MAOls and narcotic analgesics. British Journal of Psychiatry 151 : 701-701, 1987 Bickerman HA. Antitussive drugs. In Modell W (Ed.) Drugs of Choice 1984-1985, pp. 483-499, CV Mosby Company, St Louis, 1984 Boeckx RL. False positive EMIT® D.A.V. phencyclidine assay as a result of an overdose of dextromethorphan. Clinical Chemistry 33: 974-975, 1987 Bornstein S, Czermak M, Postel J. A propos a case of voluntary drug poisoning with dextromethorphan hydrobromide. Annals of Medical Psychology I: 447-451,1968 Brosen K, Gram LF. Clinical significance of the sparteine/debrisoquine oxidation polymorphism. European Journal of Clinical Pharmacology 36: 537-547, 1989 Browne B, Linter S. Monoamine oxidase inhibitors and narcotic analgesics. British Journal of Psychiatry 151 : 210-212, 1987 Chen ZR, Somogyi A, Bochner F. Dextromethorphan: pharmacogenetics, and a pilot study to determine its disposition and antitussive effect in poor and extensive metabolisers. European Journal of Clinical Pharmacology 183: 1573-1574, 1990a Chen ZR, Somogyi A, Bochner F. Simultaneous determination of dextromethorphan and three metabolites in plasma and urine using high-performance liquid chromatography with application to their disposition in man. Therapeutic Drug Monitoring 12: 97- 104, 1990b Choi DW. Dextrorphan and dextromethorphan attenuate glutamate neurotoxicity. Brain Research 403: 333-336, 1987 Eddy NB. The National Research Council involvement in the opiate problem: 1928-1971. National Academy of Sciences, Washington, D.C 1973 Eichelbaum M. Genetic polymorphism of sparteine-
Drug Safety 7 (3): 190-199. 1992 0114-5916/92/0005-0190/$05.00/0 © Adis International Limited. All rights reserved. DRS'
Dextromethorphan
An Overview of Safety Issues
Jerzy L. Bem and Richard Peck Pharma Division, F. Hoffmann-laRoche Ltd, Basel, Switzerland
Contents J90 J9J J93 J94 J94 J95 196 J96 J96 J97 197 J97
Summary
Summary I. Spontaneous Adverse Event Reports 2. Safety in Poor Metabolisers 3. Dependence 4. Recreational Abuse 5. Overdose 5.1 Adults 5.2 Children 6. Significance of Animal Neurological Findings 7. Interaction with MAOIs 8. Risk-Benefit Considerations 9. Conclusions
Dextromethorphan is a highly effective and widely used nonopioid antitussive drug. As it has been in use for more than 30 years, a large body of clinical experience has been used to formulate a safety profile. An anthology of adverse drug events has been analysed, drawn both from published case records and a data base recording dextromethorphan-related adverse events spontaneously reported by physicians or pharmacists. The resulting safety profile indicates that adverse drug reactions are infrequent and usually not severe. The predominant symptoms are usually dose related and include neurological, cardiovascular and gastrointestinal disturbances. Particular safety concerns arise when monoamine oxidase inhibiting (MAOI) drugs and dextromethorphan are coadministered. In addition to adverse drug reactions, the safety profile of dextromethorphan is affected by episodic and sporadic abuse. In fact, abuse appeared to be the most significant hazard identified by analysis of spontaneous adverse event reporting. No evidence could be found that the well documented pharmacokinetic polymorphism observed with dextromethorphan is correlated with any clinically significant safety risk if it is used for short term treatment. In summary, the safety profile of dextromethorphan is reassuring, particularly relating to overdose in adults and children.
Dextromethorphan Safety
Dextromethorphan is the methylated dextrorotatory analogue of levorphanol. It acts on the cough centre in the medulla oblongata, raising the threshold for the cough reflex. The pharmacological properties of the drug have been extensively investigated as far back as 1953. Dextromethorphan is well absorbed from the gastrointestinal tract. It is metabolised in the liver (first-pass effect) and excreted in the urine either unchanged (see section 2 below), or as demethylated metabolites including dextrorphan, which itself has some cough-suppressing activity. Dextromethorphan does not inhibit ciliary function, thus providing a clinically significant advantage of the drug over many other cough suppressants. The superior efficacy of dextromethorphan in comparison with opioid antitussives has been demonstrated in several well controlled studies (Karttunen et al. 1987; Matthys et al. 1983). In humans dextromethorphan lacks the pharmacological characteristics of opiate alkaloids. In normal individuals at therapeutic doses dextromethorphan is devoid of analgesic, euphoriant and physical dependence-producing properties (Jasinski 1979). In animals dextromethorphan is nonaddictive, even in high experimental doses. Dextromethorphan became widely used in the 1950s and is now marketed in more than 60 countries under many different proprietary names on its own or in combination. Adverse effects with dextromethorphan have a very low, but not precisely quantifiable incidence (Reynolds 1989). Considering the widespread use of dextromethorphan, there are very few reports of toxicity at recommended concentrations. Most reports of toxicity occur at high dosages as a result of misuse, ingestion with suicidal intent, or recreational use. In addition, the evaluation of toxicity is complicated by simultaneous ingestion of other substances and by the lack of objective measurements of dextromethorphan exposure. Most of the data from clinical trials were generated in the 1950s and 1960s. The more recent trials, reviews and case reports confirm the earlier findings that adverse reactions are mild (Bickerman 1984; Matthys et al. 1983). In contrast to more
191
conventional drug safety considerations, dextromethorphan has been the subject of abuse. These episodes have occurred sporadically over the past 30 years giving rise to self-limiting localised outbreaks characterised by symptoms of overdose among a limited segment of the population. Following such abuse episodes, cases of psychological dependence have been reported, but there does not appear to be any evidence of dependence of the morphine type. Recently, the discovery of the polymorphic oxidation of debrisoquine caused a resurgence of interest in genetic factors affecting the individual response to drugs. In fact, dextromethorphan has been used as a prototype in the study of polymorphism in the metabolism of a wide variety of pharmacological substances (Perault et al. 1991). It has been suggested that dextromethorphan may provide a useful diagnostic tool to identify persons with deficiency in the debrisoquine-type of drug oxidation (Chen et al. I 990a). However, the recognition of dextromethorphan polymorphism also raises questions related to the safety of dextromethorphan in the minority of people who are poor metabolisers. In this respect, the dysphoric effects elicited by high doses (240mg) of dextromethorphan are probably manifested differently in poor metabolisers (Musacchio et al. 1989).
1. Spontaneous Adverse Event Reports A relatively small number of adverse drug events (ADEs) during the past 30 years has been collected in a data base of spontaneously reported dextromethorphan-related adverse events maintained by the Drug Safety Department at Roche. ADEs are defined to include all reports affecting safety, potency, adverse reactions, clinical side effects and drug abuse problems. The drug safety data base draws upon a worldwide network of physicians and pharmacists who are actively encouraged to report upon a wide variety of prescription and over-thecounter (OTC) pharmaceutical products. A high level of cooperation is ensured by competent personnel who are assigned to evaluate the significance and severity of reports from all sources. Ad-
192
Drug Safety 7 (3) 1992
Table I. Reports of adverse drug events associated with dextromethorphan (all preparations) received by Roche Drug Safety
between 1960 and 1989. Events are grouped by body system using the WHO classification Body system
Autonomic nervous system Cardiovascular (general) Central/peripheral nervous system Endocrine Fetal Gastrointestinal system General Heart rate and rhythm Liver and biliary system Platelet, bleeding and clotting Psychiatric Red blood cell Reproductive (male) (female) Respiratory system Skin and appendages Urinary system Vascular Vision White cell and reticuloendothelial system Total
No. of events 30 8 38 7 5 41 20 21 5 161 2 20 2 5 53 5 1 25 2
452
verse events are classified into medically defined categories comprising all cases and exposures, including long-term sequelae up to 5 years. Quality control and validation of data is done by the Drug Safety department. No deliberate selection of distinct population groups is made. Analysis of these data reveals that OTC drugs generate a smaller number of spontaneous ADE reports than prescribed drugs. However, considering the very wide usage of dextromethorphan, the number of adverse events reported over a long period indicates a good safety record of dextromethorphan. 299 cases of adverse events associated with dextromethorphan (all brands and generics) have been reported worldwide up to July 1990 (table J). In addition to the spontaneous ADE reports entered directly in the database, a substantial proportion of ADEs are reported in the scientific literature, newspapers and other sources.
Psychiatric and CNS reactions (29 and 25% of ADEs, respectively) were reported most frequently. Skin (9%), gastrointestinal (7%) and cardiovascular (5%) disorders are less common. The profile of the above ADEs reported spontaneously is compatible with that described in the standard texts. Of interest is the relatively frequent occurrence of psychiatric ADEs, most being a result of drug abuse (72 reports). 35 cases were reported in the 1960s, 36 cases in the 1970s and I case in the 1980s. Other neuropsychiatric events reported are listed in table II. From these data there is no evidence of a long term or permanent effect of dextromethorphan. Dextromethorphan drug abuse has been reported from various countries over a period of at least 30 years (table III). The most recent outbreak of abuse involved 3 cases reported from Switzerlan~ in 1990. Drawn from spontaneous reports of ADEs, the range of ingested doses giving rise to adverse manifestations varied from 30 to 900mg. Approximately half of the reports involved teenagers and another third young adults. The number of publications reporting dextroTable II. Psychoneurological ADEs reported in association with dextromethorphan. Some reports describe more than one event
Adverse event
No. of reports
Abnormal thinking Aggressive reaction Agitation, anxiety, excitation Awareness disturbances Confusion, consciousness clouded Delusions, hallUCinations Depression Disinhibition Disorientation Euphoria, stimulation
3 3 9 22 27 54 42 2 22
Fear, panic reaction Lack of drive Neurosis Personality disorders Psychotic syndrome Restlessness Schizophrenic reaction Sleep disorders
36 2 28 4 7 4 12 22 2
193
Dextromethorphan Safety
Table III. Reports of drug dependence associated with dextromethorphan (all preparations) received by Roche Drug Safety between 1960 and 1989 Country Angola Australia Austria Canada Denmark Finland Germany Netherlands Philippines Switzerland Turkey UK
1960s
1970s
1980s
3 1
20
3 4 7 2
5 4 2
3
3
1 2
US Total
35
28
methorphan abuse has diminished: 16 papers were published in the 1960s, 20 in the early 1970s, I in the late 1970s and 8 in the I 980s. Although the number of incidents is too small to permit a statistical trend analysis, no indication of an increase in dextromethorphan abuse can be detected. The decrease in spontaneously reported adverse events as a function of time after initial introduction follows a pattern which has been described for prescription drugs (Guess 1989). Table IV lists a total of 12 cases of fatal events reported in patients taking dextromethorphan. All but I were known to involve overdoses, intentional or accidental. Additional drugs were taken in 6 cases and were possibly taken in another 4 cases. In the cases where the other drugs were known the contribution of these drugs to the fatal outcome is likely. In 2 cases (I from Sweden and I from Germany) death was very likely due to an extremely high overdose (up to 3g) of dextromethorphan. The UK Committee on Safety of Medicines (CSM) has received only 51 reports of ADEs between July 1963 and July 1990 and no cases of death associated with any preparation containing dextromethorphan (CSM, personal communication). The US Food and Drug Administration
(FDA), as reported in the Federal Register in 1983, released dextromethorphan safety data and concluded that because of its low toxicity, dextromethorphan is probably the safest antitussive presently available.
2. Safety in Poor Metabolisers Wide interindividual variations in the measured blood concentrations of dextromethorphan have been known for a considerable time. A study on polymorphic metabolism of dextromethorphan was published by Pfaff et al. (1983). These results have been largely confirmed by numerous other authors (Kupfer et al. 1984, 1986; Schmid et al. 1985). Within the white European population individuals metabolise dextromethorphan rapidly (about 81 %), at an intermediate rate (about 6%) or slowly (about 13%) [Pfaff et al. 1983]. The so-called poor metaboliser (PM) phenotype is inherited as an autosomal-recessive trait and occurs in 5 to 10% of European, Australian Caucasian and North American populations (Chen et al. 1990a; Eichelbaum 1988; Meyer et al. 1989). The frequency of the PM phenotype was reported to be lower in I non-Caucasian population (Nakamura et al. 1985) This polymorphism arises from differences in the O-demethylation of dextromethorphan, based on an individual genetic disposition, associated with the alicyclic hydroxylation of debrisoquine. Metabolic polymorphism is now well recognised with several drugs which are metabolised via oxidation. Pharmacokinetic consequences were described for at least 25 commonly used drugs (Brosen & Gram 1989) including {1-blockers, tricyclic antidepressants, neuroleptics and codeine. Chen et al. (l990b) showed that the citric acid-induced cough threshold was doubled after dextromethorphan was given to PMs compared with extensive metabolisers (EMs). At the same time, the mean maximum plasma concentration of dextromethorphan was 20 times higher in PMs than in EMs. The authors suggested that PM and EM subjects might require different dosage regimens for cough suppression. However, these findings may also suggest that the pharmacodynamic difference between PM and EM
Drug Safety 7 (3) 1992
194
Table IV. Reports of fatal outcomes associated with dextromethorphan received by Roche Drug Safety between 1959 and 1990. All preparations and all sources of reports are included Year
Country
Dosage
Comedication
NR 1959 1966 1967 1967 1969 1975 1979 1987 1987 1990 1990
UK Germany Turkey Turkey Canada USA Austria Canada Sweden Sweden Hong Kong Hong Kong
NR 250mg High 450mg 130mg 300mg High Normal 450-3000mg High NR NR
NR None NR Opium MAOI, 2 drugs Alcohol, LSD, marijuana Opium, antihypertensives Antidepressants
Comments
2 years of chronic use
Suicide
Suicide Suicide or overdose NR Morphine
Abbreviations; NR '" not reported; MAOI '" monoamine oxidase inhibitor.
is less than predictable from the pharmacokinetic differences (2-fold difference versus 20-fold difference). These findings could explain, at least to some extent, why in clinical practice short term treatment with the standard dose of dextromethorphan appears to be effective and safe in both PM and EM populations. Further investigation should be directed towards elucidating more fully the clinical consequences of dextromethorphan polymorphism.
3. Dependence All available experimental and clinical data indicate that the dependence potential of dextromethorphan is very low. No chiral inversion of dextromethorphan (d-form) to the opioid levomethorphan (I-form) occurs in humans. Commercial dextromethorphan preparations are free from the I-isomer (Reynolds 1989). Addiction studies in animals (mice, rats, cats and monkeys) were all negative. It was not possible to demonstrate or to induce either morphine-like effects or physical dependence on dextromethorphan (Musacchio et al. 1989). Addiction studies in humans show little or no addiction potential of dextromethorphan. Jasinski (1977) surveyed the available epidemiological
studies of various morphine substitutes and calculated the availability and relative abuse incidence in relation to morphine. This incidence was zero for dextromethorphan for 1973 to 1975. The safety of antitussive drugs has been monitored by academic, governmental and industrial bodies since 1929 (Eddy 1973; Jasinski 1977; Martin 1977). The international experience was first summarised by the World Health Organization (WHO) in 1970. The WHO, FDA and the American Medical Association consider dextromethorphan to be a non-narcotic, nonsedative drug, presenting no public health problem and not requiring special regulation (Seventeenth Report of the WHO Expert Committee on Drug Dependence 1970). Dextromethorphan has been subject to limited and sporadic abuse, but there does not appear to be any evidence of dependence of the morphine type. McCarthy (1971) studied 20 dextromethorphan abusers and noticed that after a few weeks of abuse the popularity of this drug declined after the abusers started to experience a period of somnambulism, lethargy and ataxia.
4. Recreational Abuse Sporadic abuse of practically all cough suppressants and expectorants has been reported (McCarthy 1971). This may be in the form of a cough
Dextromethorphan Safety
suppressant alone or in combinations containing an antihistamine or pseudoephedrine. The content of alcohol in some older dextromethorphan-containing preparations, as in many other cough mixtures, is as high as 50%. In the 1960s, several cases of misuse of dextromethorphan were reported, often together with psychotomimetic drugs. Neither the number of reports nor the manner of abuse give grounds to suppose that dextromethorphan has properties leading to long-lasting addiction. In contrast, an association of abuse with psychotomimetic effects, especially of short duration, cannot be excluded (Tortella et al. 1989). The earliest reports of dextromethorphan abuse in the US date back to the 1960s. Younes (1969) described a 'kick-inducer' with a dextromethorphan-containing product among teenagers. Intentional abuse of dextromethorphan-containing preparations was studied in Utah between 1985 and 1988 by McElwee and Veltri (1990). In this large population study the authors found increased intake of dextromethorphan-containing cough syrup among teenagers, but the acute health risk appeared to be minimal. In 1990, the Pennsylvania Health Authorities held hearings on dextromethorphan abuse; a further inquiry on this subject was held by the US FDA also in 1990. Data from the National Institute on Drug Abuse consisting of reports from medical examiners and emergency rooms showed that dextromethorphan accounted for only 0.01% of all abuse accounts, ranking fortyfirst among all drugs mentioned. Data from the Drug Abuse Warning Network confirms that dextromethorphan abuse poses no acute public health dangers in the US (McElwee & Veltri 1991a,b). Until 1986 the UK Home Office Drugs Branch had not received any reports of dextromethorphan misuse (Fleming 1986). The UK CSM received no reports of abuse of dextromethorphan between July 1963 and July 1990 (CSM, personal communication). In 1986, Orrell and Campbell (1986) reported the case of a 37-year-old man with a history of abusing a wide range of drugs who had been diagnosed as schizophrenic. Over the previous year, he had abused a cough syrup containing dextro-
195
methorphan, consuming up to 2 bottles (500mg) several times a week. He was admitted after complaining of feeling anxious and aggressive. After I week on antipsychotic medication his mental state improved, but on 2 occasions he became abusive, irritable, and disruptive. A case of pure dextromethorphan hydrobromide powder abuse by a 30year-old man was reported by Fleming (1986). Euphoria and restlessness occurred about 15 min after inhaling the powder and lasted for up to 2h, followed by depression, nausea, tiredness and dizziness. There is evidence of dextromethorphan abuse in Australia before 1965 (McCarthy 1971). During 1968 and early 1969, there was a marked increase in its use among young people in and around Brisbane. The abusers reported increased perceptual awareness, altered time perception and visual hallucinations. The popularity of the drug with younger drug abusers was related mainly to its easy OTC availability. Dextromethorphan was taken mainly in mixture form, the usual dose being I bottle (100mg), often mixed to disguise the taste. Onset of the effects was within Ih, and their duration was 3 to 4h. In the 1980s, teenagers in different parts of Sweden started to abuse dextromethorphan. As a result of these reports, the Swedish National Board of Health and Welfare has enforced compulsory prescriptions by a doctor from April 1986. Dextromethorphan abuse has also been reported from several other countries including Austria, France and New Zealand (Bornstein et al. 1968; McCarthy 1971). In Switzerland, 3 cases of abuse were reported in the 1960s and 3 in the 1980s. In all of these countries, questions regarding the potential for dextromethorphan abuse have been raised by the Health Authorities. According to our data, recreational abuse remains the most important drug safety consideration associated with dextromethorphan.
5. Overdose Massive overdosage with dextromethorphan may result in respiratory depression, excitation, confusion, toxic psychosis characterised by hyper-
Drug Safety 7 (3) 1992
196
active behaviour, extreme pressure of thought and morbid auditory and visual hallucinations. However, dextromethorphan overdose is not easy to achieve: clinical safety studies with dose escalation to 330 mg administered every 6h have been tolerated with comparatively minor and rapidly reversible neurological symptoms (Steinberg & Bell 1991 ). 5.1 Adults Several cases of overdose have been reported from Sweden. The inappropriate use of dextromethorphan-containing preparations was usually combined with alcohol consumption. The accompanying symptoms, some of them severe, were not life-threatening in any patient of this group. The first 2 cases of fatal overdose with dextromethorphan were reported in 1987 in Sweden (Rammer et al. 1988). The autopsy showed no gross pathology in I of thel'e cases, and signs of pulmonary aspiration of gastric content in the second. Study of tissues and blood showed concentrations of dextromethorphan to be 2- to 3-fold higher than the concentration of its metabolite dextrorphan. The authors commented on the existence of individual variations in the metabolic rate of dextromethorphan, but they did not suggest that any of these cases was a poor metaboliser. On the basis of the blood and tissue dextromethorphan concentrations, this conclusion would appear to be appropriate. McElwee and Veltri (1990) analysed 265 cases of dextromethorphan exposure reported to the Intermountain Regional Poison Control Center (lRPCC) in Utah. In 1989, IRPCC had a catchment population of about 1.7 million people. Most exposures were accidental. Intentional exposure accounted for about 22% of cases. In the cases with known outcome, most resulted in either no effect or minor effect for the patients. None of the patients died or developed life-threatening symptoms or permanent disability from exposure. The authors concluded that there is only minor abuse of dextromethorphan, and the risk to health from those exposures is minimal.
5.2 Children A few serious dextromethorphan overdoses have been reported in children. The clinical effect of such overdoses was predominantly neurological. The cases were characterised by central excitation rather than sedation. All patients treated with naloxone responded dramatically (Boeckx 1987; Henreting et al. 1988; Katona et al. 1986; Shaul et al. 1977). The FDA reviewed all reported overdose cases (Federal Register 1983) and found no fatalities even with doses exceeding 100 times the normal dose. Up to 1983, despite widespread use of preparations containing dextromethorphan, only 33 cases of accidental overdose requiring medical intervention were known. Five of these cases have been the subject of publications. The age of the children concerned was between 10 months and 10 years. The amount taken varied from 15 to 600mg. A fatal outcome following accidental overdose with dextromethorphan alone is very unlikely. Naloxone is available as an effective antidote, and improvement in the child-proof packaging has further increased safety for children.
6. Significance of Animal Neurological Findings In various animal and in vitro models, high doses of dextromethorphan appear to have both neuroprotective and neurotoxic properties. The neuroprotective activity of the drug has been reported amongst others by Choi (1987) who demonstrated that dextromethorphan and dextrorphan can substantially reduce the neurotoxicity of glutamate in vitro. He further suggested that dextromethorphan or related compounds may eventually have clinical therapeutic utility in a variety of neurological disease states. Prince and Feeser (1988) demonstrated a protective action of dextromethorphan against cerebral infarction in a rat model of hypoxia-ischaemia. Engbar and Chase (1988) propose that dextromethorphan might prevent neurotoxicity due to long term exposure to low levels of excitotoxins, as may occur in neurological dis-
197
Dextromethorphan Safety
orders such as Huntington's disease and Alzheimer's disease. Weisand et al. (1990) reported that dextromethorphan injections produced swollen neuronal somata in the retrosplenal cortex of male rats. Evidence suggesting that other glutamate receptor antagonists may have neurotoxic effects in rats was provided by Olney et al. (1989). They demonstrated vacuolar changes in specific populations of brain neurons when the noncompetitive N-methylD-aspartate (NMDA) antagonists MK 801, phencyclidine, tiletamine and ketamine were administered subcutaneously to adult rats in relatively low doses. Anticholinergic compounds were reported to protect neurons from vacuolisation without interfering with neuroprotection from MK 801 (01!:ley et al. 1989). No clinical evidence that the histological findings in animals are relevant to human therapy with dextromethorphan exists at present. The issues of dose, duration, route of administration and metabolic differences between man and rat make the extrapolation of the animal data to humans extremely difficult. The significance of vacuolisation observed in rat brains after subcutaneous injection requires more research.
7. Interaction with MAOIs An interaction between dextromethorphan and MAOIs is well documented in the literature (Harrison et al. 1989; Rivers & Homer 1970; Shamsie & Barriga 1971). The interaction manifests itselfas a serotonergic syndrome: high blood pressure, hyperpyrexia, arrhythmia (Browne & Linter 1987) or myoclonus. The interaction may be fatal (Shamsie & Barriga 1971). The pharmacological mechanism for the interaction of dextromethorphan with MAOIs may be that of dextromethorphan blocking the neuronal reuptake of serotonin (Sinclair 1973). Lethal doses of dextromethorphan in combination with an MAOI in rabbits were very high (Sinclair 1973). Many reported cases of dextromethorphan interaction with MAOIs involve OTe preparations which also contain other potentially hazardous
components. The spontaneous report data base has obtained 1 fatality report of a case where a patient took overdoses of both dextromethorphan and an MAO!. The clinical relevance of the interaction between dextromethorphan and MAOIs remains controversial. Administration of relatively high levels of the MAOI toloxatone failed to significantly alter the metabolic ratio of dextromethorphan in humans (Perault et al 1991). Similarly, the use of dextromethorphan and MAOIs in clinical practice has been reported without incident (Bazire 1987). Nevertheless, in consideration of the possible risk, dextromethorphan-containing formulations should not be used under ordinary circumstances in conjunction with MAOI therapy.
8. Risk-Benefit Considerations Unproductive, persistent cough is not uncommon and often distressing, but the symptoms are rarely serious enough to justify the use of other than very safe drugs. In comparative clinical trials dextromethorphan has been shown to be superior or equal to other antitussives (Matthys et al. 1983). All the efficacious antitussives cause adverse effects. For comparison, other commonly used antitussives have their own safety considerations. Opiates in general have more central depression and addiction potential than dextromethorphan. Antihistamines and anticholinergic agents can, in larger doses, inhibit the secretion of the bronchial glands and also will cause central depression. Ammonium chloride can occasionally cause clinically significant acidosis. Iodide is not safe in certain patients and can cause allergic reactions. Sympathomimetics are associated with a number of side effects. Lastly, benzocaine can cause allergic reactions or sensitise a patient.
9. Conclusions Dextromethorphan has proven itself in long term clinical use to be a highly effective nonopioid antitussive drug. From an analysis of adverse drug events drawn from published case records and
198
spontaneously reported events, dextromethorphan has a wide margin of safety. Dextromethorphanrelated adverse drug reactions are infrequent and usually not severe. Overdose is rare both in adults and children. However, serious concerns surround the probable interaction between MAO! drugs and dextromethorphan. In addition to adverse drug reactions, the safety profile of dextromethorphan is affected by outbreaks of abuse: this constituted the most significant hazard identified by analysis of spontaneous reporting. Pharmacokinetic polymorphism does not appear to present a clinically significant safety risk if dextromethorphan is used for short term treatment. Given the very large volume of dextromethorphan use and the comparatively small number of serious ADEs reviewed here, the risk is low. In conclusion, the safety profile of dextromethorphan is extremely favourable.
References Bazire s. MAOls and narcotic analgesics. British Journal of Psychiatry 151 : 701-701, 1987 Bickerman HA. Antitussive drugs. In Modell W (Ed.) Drugs of Choice 1984-1985, pp. 483-499, CV Mosby Company, St Louis, 1984 Boeckx RL. False positive EMIT® D.A.V. phencyclidine assay as a result of an overdose of dextromethorphan. Clinical Chemistry 33: 974-975, 1987 Bornstein S, Czermak M, Postel J. A propos a case of voluntary drug poisoning with dextromethorphan hydrobromide. Annals of Medical Psychology I: 447-451,1968 Brosen K, Gram LF. Clinical significance of the sparteine/debrisoquine oxidation polymorphism. European Journal of Clinical Pharmacology 36: 537-547, 1989 Browne B, Linter S. Monoamine oxidase inhibitors and narcotic analgesics. British Journal of Psychiatry 151 : 210-212, 1987 Chen ZR, Somogyi A, Bochner F. Dextromethorphan: pharmacogenetics, and a pilot study to determine its disposition and antitussive effect in poor and extensive metabolisers. European Journal of Clinical Pharmacology 183: 1573-1574, 1990a Chen ZR, Somogyi A, Bochner F. Simultaneous determination of dextromethorphan and three metabolites in plasma and urine using high-performance liquid chromatography with application to their disposition in man. Therapeutic Drug Monitoring 12: 97- 104, 1990b Choi DW. Dextrorphan and dextromethorphan attenuate glutamate neurotoxicity. Brain Research 403: 333-336, 1987 Eddy NB. The National Research Council involvement in the opiate problem: 1928-1971. National Academy of Sciences, Washington, D.C 1973 Eichelbaum M. Genetic polymorphism of sparteine-
