söndag 6 april 2014
Fatalities due to use or misuse of pain-killers containing dextropropoxyphene
in collaboration with the Swedish National Board of Forensic Medicine
Foreword by the National Board of Forensic Medicine
Authors’ preface to the English edition
A short summary of the content of this report
The organisation and duties of forensic medicine in Sweden
Results in summary
Dextropropoxyphene – field of application and toxicity
Background and field of application
Pharmacokinetics and pharmacodynamics of DXP
Toxicity of DXP
Fatal DXP poisoning – suicide or accident?
International experiences and measures taken
DXP in the United States
DXP in the United Kingdom
DXP in Denmark
DXP in Norway
DXP in Sweden
An Australian contribution
Dependence on DXP
The scheduling of narcotic drugs
Summary, discussion and conclusion
Analgesic dependence among chronic pain patients
The prevalence of DXP and fatal DXP poisoning among forensic autopsies
DXP and alcohol
The classification of the manner of death in fatal DXP poisoning
Unmixed and combined DXP preparations
Age and kind of preparation in fatal poisoning
DXP in individuals suspected of driving under influence
The analgesic effect of DXP according to meta-studies
Appendix 1 Eight studies on dextropropoxyphene
Appendix 2 References
Foreword by the National Board of Forensic Medicine
Authors’ preface to the English edition
The interest in examining the analgesic preparation dextropropoxyphene (DXP) and its side-effects in respect to dependence and fatal poisonings first emanated from our experiences from teaching in early detection of alcohol and drug problems. On several occasions physicians and other professionals reported that they had seen patients abuse analgesics such as Distalgesic, and that some of them had died suddenly and unexpectedly after having combined alcohol with these pills. These individual referents all considered that DXP was a serious problem that needed attention.
Against this background we started a project at the Department of Forensic Medicine at Uppsala University, under the supervision of Tom Saldeen, MD, PhD, professor and chairman of the department. The project has led to eight scientific papers (1-8), all published in international scientific journals, and one doctoral thesis (9). A second thesis will be presented in March 2001. Unfortunately, results of scientific research are mostly slow in reaching the public. Scientific publications are often read only by a handful of individuals who are interested in that particular subject or journal, and otherwise they are placed on some bookshelf and remain unread.
Early in the project we found a considerably high rate of fatal poisonings due to DXP, both suicidal and accidental. We then became aware of the fact that if human lives were to be saved, it would be necessary to inform the society outside the academic world of the results of the research. We considered it important to inform both the prescribing physicians and the users of DXP about the risks of the most frequently prescribed analgesic preparation in our country. It seemed urgent that we should reach not only abusers but also the ordinary users, since both categories are at risk of becoming intoxicated, all too often with a fatal outcome.
The National Board of Forensic Medicine, which has supported us financially, agree about the importance of disseminating our results and has edited this report, which is intended among others, for general practitioners, journalists, politicians and the general public. The Swedish version of the report has attracted much attention and so far has been printed three times.
Since we have reason to believe that the problem of constantly high rates of DXP fatalities is not limited to Sweden, we hope that the translation of this report into English will constitute a useful contribution to the debate in other countries. Reports have indicated that dextropropoxyphene-related deaths are a persistent problem in the United Kingdom (10), in Australia (11), in New Zealand (12) and in the USA (13). In all countries where DXP is frequently prescribed, a rate of fatal DXP poisoning comparable to that in Sweden can be expected. When related to the population and assuming similar conditions, the number of deaths each year would be about 5,000–7,000 in the United States, about 1,000–1,200 in the United Kingdom, about 300–400 in Australia and about 60-70 in New Zealand.
We wish to express our gratitude to our supervisor, Professor Tom Saldeen. We also thank the National Board of Forensic Medicine for financial support and encouragement in the project and for editing the report of our collected research.
We are also grateful to the Swedish Poison Information Centre, which prior to our studies had recognised the risks of DXP and seriously shared and supported our concern about the constantly high rate of fatal poisonings caused by this preparation. We are indebted to the National Corporation of Swedish Pharmacies, which has supported us with information on the prescribing of DXP preparations during the study years, and the National Board of Health in Denmark, which gave us the opportunity to read all documents regarding the process leading to restrictions of DXP prescription included in their DXP dossier.
Finally, we want to dedicate this book to all the 1,600 persons who were known to be victims of fatal DXP poisoning during the years 1992 - 1999. Their tragic deaths have given us knowledge of the serious risks of our most frequently prescribed analgesic preparation. It is our hope and expectation that this knowledge will contribute to saving lives in the future.
Uppsala, Sweden Picture
Fully certified psychologist, fully certified psychotherapist,
Doctor of Medical Science
Journalist, graduate student in Public Health
Dextropropoxyphene (DXP) is one of the most frequently prescribed analgesic compounds in Sweden today and it is recommended to patients with moderate to serious pain. There are seven preparations containing DXP on the Swedish market; in three of them DXP is the sole analgesic ingredient, while four of them are combinations of analgesics. In Distalgesic, for example, paracetamol is also included. In 1996 the sale of DXP preparations was 14.4 defined daily doses/1,000 inhabitants during a 12-month period (DDD).
A short summary of the content of this report
Chapter 2 gives a general background of the production of DXP in the 1950s and describes how it quickly gained world-wide popularity. Soon after its introduction it was reported, however, that the margin between the therapeutic and toxic doses was small, and in chapter 3 the discussions in the USA, the United Kingdom and the Scandinavian countries following this discovery are presented. We have paid special attention to the process leading to restriction of the prescribing of DXP in Denmark, contrasting this with the lack of a similar debate in Sweden. At the end of this chapter the classification of the manner of death in fatal poisoning is considered. How does the physician establish the intent of the deceased in cases of self-inflicted deaths?
In chapter 4 dependence on therapeutic drugs is discussed and the scheduling of narcotic drugs is described. When DXP was introduced, it was claimed that unlike morphine and some of the more potent analgesics it caused little or no tendency to addiction. However, soon after its introduction case reports showed that just like other opioids, DXP could lead to dependence and abuse.
Our studies are summarised in chapter 5. We first investigated analgesic use disorders among chronic pain patients at a rehabilitation clinic in Sweden. There were indications that some patients did become dependent on analgesics and in 1992 a new admission routine was initiated, whereby all patients were interviewed concerning their drug use. The investigation showed that patients with chronic pain are at considerable risk of developing DXP use disorders, even when on ordinary medication. Since we had been told that some persons had died unexpectedly when consuming DXP, it seemed natural to continue to examine this drug in the context of forensic medicine.
The eight studies summarised in chapter 5 are presented in more detail in Appendix 1 for the benefit of any readers who would like to know more about the methods, the results and the discussions in each study. For further analysis we refer to the original published articles. All references are found in Appendix 2.
Below we will give a brief description of the forensic medicine activity in Sweden and the bases of our autopsy investigations.
The organisation and duties of forensic medicine in Sweden
The primary task of forensic medicine is to investigate deaths that are suspected to be unnatural and persons involved in suspected crimes of violence. The findings and conclusions are presented in legal certificates and form a basis of decisions by the police, prosecutors and courts. The vast and solid experience of forensic medicine has also contributed to the work of prevention of crime, injury and disease conducted by other authorities and organisations, as well as to the efforts of finding the causes of sudden and unexpected deaths.
The organisation of the medico-legal examiner system in Sweden is influenced by the continental system, which is a common model in large parts of Europe. All autopsies in cases of deaths suspected to be unnatural are conducted on behalf of the judicial bodies, mainly the police. Since 1991, all medico-legal work in Sweden is carried out at six departments of forensic medicine that answer to the National Board of Forensic Medicine. The National Board is also responsible for the fields of forensic chemistry, forensic genetics, and investigational forensic psychiatry.
About 90,000 people die in Sweden every year. Of interest to forensic medicine are primarily deaths due to accidents, crimes and suicide, i.e. cases of unnatural death. The frequency of autopsies has gradually decreased, from about 40 per cent in 1980 to 20 per cent in 1995. About 70 per cent of the autopsies are performed within the medical service, while the remaining 30 per cent (between 5,000 and 6,000 per year) are medico-legal autopsies undertaken at the departments of forensic medicine. The medico-legal autopsies include almost all deaths due to homicide, manslaughter, other violence, and suicide. In addition, the departments of forensic medicine deal with more than 90 per cent of the traffic deaths and about half of all other fatal accidents.
A forensic chemistry analysis is required in about 90 per cent of all medico-legal investigations of death. The purpose of these analyses may vary from screening for chemical substances that might explain a case of death, to exclusion of the presence of alcohol, prescription drugs, or other substances. Prescription drugs, primarily sleeping pills, sedatives and analgesics, are commonly found in connection with suicide and deaths from undetermined causes.
The results obtained in the fields of forensic medicine and forensic chemistry provide a useful source of data on unnatural deaths as well as on unexpected natural deaths, and a computer-based system for assignment management (RättsBase for forensic medicine and ToxBase for forensic chemistry) has been developed and put into operation. The two databases contain all assignments since 1992 and constitute valuable resources for research and development work, not only in the forensic sciences but to a great extent also for researchers within other areas of society, such as universities and public authorities and organisations (14).
RättsBase and ToxBase made it possible for us to examine all the medico-legal autopsy cases in Sweden from 1992 to 1996, thus giving us results on a national level. We have since updated the investigation to include all the autopsy cases up to 1999. Beside the results from the published studies we will therefore complete this presentation with the results from the updating.
Results in summary
The prevalence of DXP in the peripheral blood of the deceased increased by 35% from 1992 (6.8%) to 1997 (9.2%). In 1998 and 1999 a tendency to a reduction was noted (8.3% and 8.1% respectively). DXP was the single preparation causing the incomparably largest number of fatal poisonings in Sweden during the study years. DXP caused or contributed to death in a total of 1,592 cases during the period 1992 to 1999, i. e. in about 200 fatal poisoning cases per year (range 158-227).
Middle-aged men, habitual or social drinkers, on medication for pain, are most prone to combine DXP and alcohol and are most liable to become victims of accidental poisoning due to a combination of DXP and alcohol. About 30 men and 10 women died accidentally every year of the study period of this combination.
Unmixed preparations were over-represented among fatal poisonings where DXP caused or contributed to death, compared with preparations containing DXP combined with other analgesics. Among fatal poisoning cases due to DXP, antidepressants or sedatives, DXP predominated among younger people compared with the other drugs.
Accidental DXP fatalities were underestimated. The alternative to a report of suicide on death certificates in these fatalities has become an undetermined manner of death, thus reducing judgements of the deaths as accidents to a minimum. As discussions of suicide rates often include the undetermined cases, there will most certainly be a considerable over-representation of suicide in the case of DXP. And this in turn will increase the risk that knowledge of the high toxicity of DXP will not reach the population consuming this drug.
In order to save a considerable number of human lives each year in our country, we propose strict regulations in the prescribing of DXP, as it seems absolutely necessary to limit the availability of the DXP preparations. The combination of liability to addiction and high toxicity makes DXP especially dangerous, since one of the most important criteria for being dependent is loss of control of the amount taken. As DXP can even be lethal within the therapeutic range when combined with alcohol or sedatives, it may become a death trap not only to those who misuse, but also to ordinary users of this analgesic.
The natural thing would be to remove DXP completely from the market and to replace it with preparations that are less dangerous when it comes to the risk of poisoning as well as the risk of dependence. Today alternative analgesics are available for prescription to the category of patients at present receiving DXP.
field of application and toxicity
Background and field of application
Opium was known to have therapeutic effects on the human body as early as 5,000 years ago among civilisations in the areas of the Euphrates, the Tigris and the Nile. As opium had a depressant effect on what is now known as the central nervous system (CNS), it was used as an analgesic drug. In the beginning of the 19th century chemists discovered that crude opium contained about 20 different alkaloids, which are chemical compounds existing naturally in the vegetable kingdom and which often have strong effects on the human organism. The most important of these alkaloids are morphine and codeine. Morphine is the most powerful analgesic existing today and it also creates a psychological sense of well-being, i. e. euphoria.
In the year 1874, diacetylmorphine – in daily language heroin – was for the first time produced from morphine. Heroin, which was first recognised as a problem in Sweden as late as in 1975, is a semisynthetic drug, while methadone and DXP are completely synthesised, with effects resembling those of morphine, only weaker (15).
DXP is chemically similar to methadone and pharmacologically to codeine. It was first synthesised in 1953 (16). At that time there was a need for an analgesic that was free from the side-effects of other narcotic analgesics, namely the development of tolerance, which presages a liability to addiction, and respiratory depression, which has serious consequences in general anaesthesia and obstetrics.
In 1958 DXP was presented in JAMA as Dextro Propoxyphene Hydrochloride - under the section of New and Nonofficial drugs (17). It was stated that unlike morphine and some of the more potent analgesics it produced no respiratory depression after therapeutic doses, that its addiction liability was little or none and that its toxicity was low and the margin of safety adequate.
DXP is rapidly absorbed after oral administration and the peak plasma level is usually attained within one hour of a single oral dose. However, since DXP undergoes marked first-pass biotransformation, only a small proportion of the absorbed dose enters the circulation in unchanged form. The remainder of the drug is metabolised to norpropoxyphene. The elimination half-life of DXP is 12 hours, but that of norpropoxyphene is three times longer, 36 hours. With repeated administration of DXP, accumulation leads to plasma concentrations that are at least 500% higher than with single doses, and consequently the elimination half-time may be substantially prolonged. It has been found that alcohol reduces the first-pass biotransformation of DXP, so that the availability of the administered dose may be enhanced by 25-50% (18-21).
DXP preparations immediately became very popular, but a short time after their introduction evidence of high toxicity of DXP began to be reported. The first report on a case of self-poisoning with DXP was published in 1960 (22), and this was soon followed by other reports of acute overdoses of DXP, cases in which the persons had survived (23-25).
In 1964 the first fatality was reported (26), a teenage girl who died after ingestion of an overdose of DXP. The authors concluded that although in normal doses DXP is well tolerated (according to the Department of Forensic Chemistry the blood level of DXP after a therapeutic dose is 0.05-0.75 µg/g), in excessive doses it possesses a malignant potential in terms of central nervous system toxicity. In severe poisoning, rapid loss of consciousness, respiratory depression, hypoxia, and ultimately cardiovascular collapse appear to be the expected course.
In the 1960s it was discovered that the margin between therapeutic and toxic doses was narrow and that the effects of DXP were potentiated after intake of alcohol. When alcohol and other toxic substances were absent, the minimal fatal concentration of DXP in the blood was found to be 2-3 µg/g (3-4 times the therapeutic dose). When DXP was combined with alcohol, death would occur at blood concentrations as low as 1-2 µg/g, in some cases as low as 0.2 µg/g. It was shown that there is a high risk of death if 20 or more Distalgesic tablets, each containing 32.5 mg DXP and 325 mg paracetamol are ingested, and that as few as 6 to 15 tablets can lead to lethal consequences when potentiated by alcohol. Further, it was shown that synergism also exists between DXP and barbiturates (27-31).
Fatal DXP poisoning – suicide or accident?
When a person dies of poisoning it is the task of the forensic medical physician to decide whether the manner of death was suicide or an accident. The manner of death is classified in accordance with the International Classification of Diseases (ICD), which also includes a classification of injuries and causes of death. The revisions of ICD have included some changes in the way in which self-inflicted deaths should be classified. According to the seventh revision, ICD-7 (1958), self-inflicted deaths were to be regarded as suicide if it could not be proved that the death was unintentional. When the eighth revision, ICD-8, was introduced in 1968, an “undetermined” manner of death was added as an alternative to suicide in the classification of self-inflicted deaths. From this it could be understood that self-inflicted death should not be considered as suicide unless the certifier was able to establish that the deceased intended death (32). ICD-9 and ICD-10 follow a similar approach (33).
The rapid death in DXP poisoning, the narrow margin between therapeutic and toxic doses, and the potentiation of DXP when mixed with alcohol may lead to difficulties in assessing the intent of the deceased in cases of self-inflicted fatal poisoning, and in classifying death as accidental, suicidal or undetermined. There is no agreement on valid rates of suicide or accidents associated with this drug. A suicide rate of 50-60% has frequently been reported, while the accident rate has varied between 25% and 49% (34-36).
(Picture: Suicide? Accident?)
There has been considerable scientific documentation of DXP toxicity in some parts of the world since the 1970s, and a survey of the discussion on this subject in the United States, the United Kingdom and Scandinavia, and of the measures taken to prevent DXP intoxication, is presented below.
Early in the 1970s warnings about the risk of DXP toxicity and about the potentiating action of alcohol on the DXP effect were issued in the USA on the bases of clinical reports (37). It was concluded that physicians should be alerted to the potential deleterious effects of indiscriminate use and abuse of DXP, and should warn their patients not to drink alcohol when taking DXP. It was recommended that physicians should observe extreme caution when prescribing DXP to young people.
In extensive surveys it was found that the number of deaths involving DXP was increasing each year, and at a faster rate than the total drug deaths. Deaths attributed to suicide, as well as those judged to be accidental deaths and undetermined deaths had increased (35, 38-40). DXP, according to these studies, did not appear to be a paediatric problem. Nor did the deceased persons belong to the illegal drug abuse population or have any particular propensity for the use of heroin or narcotics. Rather they constituted a particular medical population of those who misuse prescription drugs and alcohol. Almost all of them received a wide range of prescription drugs, particularly tranquillisers, which they often misused by self-medication, multiple drug ingestion, and combining alcohol with their medication.
The reports of fatalities due to DXP led to an information campaign in 1978-80 undertaken by the US Food and Drug Administration (FDA) and the drug’s manufacturer through mailed warnings, face-to-face education of prescribers, press- releases, and labelling changes. The goal was to reduce simultaneous use of DXP and alcohol or other CNS depressants, to reduce the prescription of refills, and to get physicians to stop prescribing DXP for patients at risk of abuse and misuse.
Trends in prescribing and deaths related to DXP overdose before and after this campaign were analysed by Soumerai and collaborators (14). They found that a pre-existing downward trend of about 8% per year continued during the warnings, but halted after them. The no-refill recommendations had no effect on the refill rates. They also noted that the overdose death per DXP prescription had remained about constant since 1979, and considered that stronger, more sustained regulatory or educational measures had to be implemented to achieve any sharper reductions in the misuse of the drug. Further, they suggested that the wide extent of DXP use is mainly due to its overall popularity in relation to competitive products, probably largely on account of effective marketing activities. One example of these activities is described in their report:
As FDA had no ongoing physician education programme of its own, it requested that the manufacturer of DXP should revise the drug labelling and conduct both a mailed and a person-to-person educational campaign. Eli Lilly and Company sent a mailed warning message to 145,000 physicians, but when it came to the face-to-face education an investigation conducted by an FDA official showed that Eli Lilly did not fulfil this commitment. It was only in less than 10% of the "face-to-face educational occasions" that the detailers conveyed suitable information on the new warnings; while over 75% of the detailers left free samples of DXP products on those particular occasions. The FDA official quoted several verbatim messages from Lilly detailers recorded by physicians. “Darvon and Lilly won FDA battle”, “Safe in spite of Nader report”, “OK by the Drug Commission”, “Few if any side effects”.
(Picture: “And here you are, some free samples”)
Today DXP is among the top ten drugs reported by medical examiners to occur in deaths due to drug abuse.
In the United Kingdom it was claimed in the early 1970s that many doctors were unaware of the danger of DXP in overdose and that the problem of this drug in the UK was not fully appreciated. Some clinicians found that deaths occurred very rapidly, suggesting that the narcotic effects of DXP predominate, and that this might explain the sparsity of clinical reports (41).
During the 1980s there was a considerable discussion and research on DXP in the UK (21, 30, 43-50). In 1980 Young and Lawson concluded that no analgesics are devoid of danger when taken in overdose, but stated that in DXP the evidence suggests that its dangers exceed its analgesic properties (30). In 1987 the debate of the 1980s came to a close with a review by Lawson and Northridge of the world-wide situation regarding acute overdose of DXP (21). Among other aspects, they surveyed the effects of different preventive measures adopted in various countries. Their final conclusion was that if effective primary prevention was to be attempted by restricting the prescribing of an analgesic, then alternative agents that were safer in both normal usage and in overdose, and preferably no more expensive, had to be promoted. In the mid-1990s the fact was established that DXP-related deaths were a problem that persisted in the UK (10). As compound analgesic preparations containing DXP were still widely prescribed and deaths due to DXP overdose were persistent, the introduction of stricter prescription control and educational programmes for users was proposed.
As in the UK, reports of DXP poisoning began to appear in Denmark in the early 1970s. A rapid increase in fatal accidental DXP poisonings that was proportional to the increased use of the drug was reported (51). It was pointed out that DXP in the form of a slow-release preparation (Abalgin retard) was especially dangerous, because of repeated dosage by patients attempting to induce the therapeutic effect more quickly.
In 1982 the Danish National Board of Health sent a message to all physicians to the effect that they should not prescribe DXP preparations to known addicts. This led to a small decrease in the prescription rate (from 8.4 DDD in 1982 to 8.0 DDD in 1983) (52). The National Board of Health also started to make enquiries in all cases of fatal DXP poisoning of alcohol and/or drug abusers, and asked the prescribing doctor about the indication and dosing instructions for the drug (National Board of Health, 1998).
In spite of the above-mentioned measures, reports of fatal DXP poisonings continued to be published (53-60) and in the dossier at the National Board of Health (54) DXP was now labelled “the killer drug”, quoting one of the reports (55). In April 1988 the National Board of Health considered total withdrawal of DXP preparations, but as it was concluded from an investigation ordered by the Board that DXP preparations were regarded as a valuable alternative to acetylsalicylic acid and paracetamol by some patients (61), and since there were no satisfactory analgesic alternatives for patients with moderate pain, strict prescription regulations were decided upon instead.
In July 1988 information was sent to all physicians in Denmark to the effect that it was now obligatory that they should submit copies of all prescriptions of DXP preparations for registration. The prescription form for narcotic drugs was to be used, with only one preparation per prescription. The national identity number of both the patient and the prescribing physician should be recorded. Further, it was again emphasised that DXP preparations should not be prescribed to alcoholics or drug addicts, but only to patients without any mental disorders. DXP should not constitute first-hand choice, before comparable analgesics had been tested.
The public attention drawn to the DXP problem by the National Board of Health and finally the prescription regulations led to a decline in the DXP prescription rate and in rate of deaths due to DXP during the late 1980s. The death rate from DXP poisoning decreased by 45% from 1986 to 1992, while the DDD was reduced by 60% from 1985 to 1992 (62-65).
The measures taken in Denmark to reduce prescribing of DXP and the DXP death rate were successful, but it seemed as if the heavier narcotic medical compounds partly compensated for these declines. In Denmark the legal prescribing of strong opioids is higher than in any other country in the world after adjustment for size of population (66).
In Norway a warning concerning the risk of intoxication by DXP was sent to all physicians in 1975 (67). In 1982 the regulations were strengthened, and DXP came to be classified in the same group as morphine. A decline in the death rate from DXP poisoning had already occurred in connection with the discussion preceding the more stringent regulations (68). The DDD has been constant at 1.6 since 1990 (69). The present death rate for DXP poisoning is 4-8 cases per year (70).
No other opioid has compensated for DXP, but generally the use of analgesic drugs is lower in Norway than in many other countries. In 1995 the sale of analgesics in Norway amounted to 64% of that in Sweden and to 60% of that in Denmark. The majority of the prescribed drugs were non-narcotic. The narcotic drugs comprised 5% of the total use of analgesics (69).
In Sweden only a few reports have been published on the subject of DXP since the 1970s. In a study in 1971 it was concluded that DXP may be fatal even in moderate doses, that analgesic compounds seem to be associated with an especially high risk, and that there is a risk of dependence with these drugs. As the analgesic effect of DXP is not strikingly high, they suggested that caution should be observed in the prescribing of these preparations, and that the market should be cleansed regarding to the DXP compounds (71).
A literature survey undertaken to assess the risks of abuse and poisoning due to DXP, based on 542 references concerning this drug, was published in 1976 (72). The reviewers found that 112-160 fatal poisoning cases due to DXP occurred every year in Sweden, and drew the same conclusions as in the study mentioned above, namely that caution should be observed when prescribing DXP. In view of the high toxicity, physicians should only prescribe small amounts of the drug, and should not prescribe DXP to patients at risk, for example alcoholics and drug addicts. The physicians should warn the patients of the risk of taking alcohol together with DXP.
The prescription rate temporarily decreased during the last years of the 1970s and nothing of the DXP discussions held in the USA, the UK and the other Scandinavian countries, Denmark and Norway, during the 1980s was reflected in Sweden. There was complete silence until 1993, when an increase in the rate of deaths due to DXP from 88 in 1985 to 119 in 1990 was reported (73). In 1985 the DXP fatalities constituted 64% of all analgesic poisonings, compared with 72% in 1990. DXP was also found to be responsible for the largest number of all fatal poisonings in Sweden during the years 1992–1995 (74).
In 1994 we started research on side-effects of DXP and in 1999 we summarised our research in the Swedish version of this report, supported by the National Board of Forensic Medicine. The scientific articles emanating from our research were all published during the years 1998 and 2000 (1-8). The publication of our results has led to a lively discussion in Sweden, which has received large-scale attention in the media. When this book was first edited the director general of the National Board of Forensic Medicine wrote an official letter to the director general of the Medical Product Agency (MPA) requesting that immediate preventive measures be taken to reduce the serious side-effects of DXP. The problem of DXP fatalities and dependency has also gained political interest. For example a motion was proposed to the Swedish Parliament that attention should be paid to this issue, and the Minister of Health and Social Affairs was asked whether it was to be dealt with or not. The Minister of Health and Social Affairs referred the question to the MPA, who announced in a press notification in August 2000 that the pharmaceutical industry was being requested to supply all DXP packages with extra warning texts in view of the lethal danger in overdosing DXP or in combining it with alcohol. The question of restriction of DXP prescription is at present under consideration by the Federation of Swedish County Councils and the National Board of Health and Welfare.
Prescription data for Australia were compared with information on drugs taken in self-poisoning and suicide, to determine which drugs were over-represented (11). After adjustment for numbers of prescriptions, DXP had the fourth highest odds ratio, after short-acting barbiturates, chloral hydrate and colchicine, and preceding tricyclic antidepressants. It was stated that short-acting barbiturates, chloral hydrate and DXP have little or no advantage over corresponding alternatives, and have excessive toxicity in overdose. It was therefore considered that they should be removed from the market.
Fatalities due to DXP have been reported from several countries for more than three decades. Among these reports there has been over all agreement about the toxicity of DXP. A majority of the studies have established the high risk of DXP poisoning and have led to the proposal that prescription of this drug should be restricted, and sometimes even to the opinion that DXP should be withdrawn from the market. Nevertheless, of the countries mentioned above, it is only the two Scandinavian countries Denmark and Norway that have in fact implemented strict regulations of prescription. In all the other countries the problem of DXP fatalities still persists.
Persons abusing different kinds of prescription drugs may be divided into two main categories: on the one hand the so-called ordinary people, who have narcotic drugs prescribed because of illness and who develop a primary dependence on these drugs, and on the other hand people already abusing other drugs, who want to strengthen the intoxication or use the narcotic drug in the state of withdrawal.
In the first category, of so-called therapeutic abusers, the drug is first taken in normal doses, but later the users develop tolerance and gradually increase their doses to get the same effect as before. Some of these abusers become physiologically dependent and have to continue the medication to avoid withdrawal symptoms, while others may become psychologically dependent, addicted to the euphoric or relaxing effects of the narcotic drugs. These latter persons may develop a pattern of increasing pathological use.
(Picture: I am not dependent on pills! …..but I need more of them!)
According to ICD-10 (33), a person may be diagnosed as dependent on a substance if he or she fulfils at least three of six defined criteria during a period of twelve months. These criteria are summarised in the following:
· a strong craving for or a compulsion to take the drug
· loss of control regarding the amount, the duration or the point in time
· characteristic withdrawal symptoms and/or the substance is taken to relieve or avoid withdrawal symptoms
· marked tolerance
· the drug occupies an increasingly central position in life
· continued use despite experience of physiological as well as psychological consequences of the drug.
Although DXP is an opioid, the possibility of addiction was thought to be low at the time of introduction of the drug, and one early study showed that the potential for addiction was substantially lower than that of codeine (75). Only a few years later case reports of patients showing addiction to DXP began to be published (76-79). First it was suggested that DXP was being used by persons already addicted to some other drug (72), in order to potentiate the effect of the other drug or in situations of withdrawal. Later it was shown that DXP could also be a drug of primary abuse (80) and that dependence most frequently seemed to develop among patients with chronic pain (81).
The narcotic conventions of the United Nations (UN) decide which drugs should be defined as narcotic. These drugs are allowed only for medical and scientific purposes. The reason why a substance is scheduled as narcotic is that it is addictive. Each country that has attached itself to the narcotic conventions of the UN may add other drugs in its own national legislation (82). The actual narcotic conventions of the UN are founded on three international conventions:
1. The general narcotic convention of the UN from 1961. The convention includes substances from the vegetable kingdom that can be abused – such as cannabis, cocaine, opium, morphine, heroin and other opiates. In June 1997 153 UN states had attached themselves to this convention.
2. The psychotropic convention from 1971. This convention includes synthesised drugs such as LSD, amphetamine, barbiturates, benzodiazepines and ecstasy. It is supported by 144 countries.
3. The narcotic convention from 1988. This has been accepted by 138 countries, which means that these countries are obligated to co-operate to prevent the increasing numbers of narcotic crimes.
In the USA the Controlled Substances Act (C.S.A.), Title II of the comprehensive Drug Abuse prevention and Control Act of 1970, is the legal foundation of the government’s fight against abuse of drugs and other substances. This law is a consolidation of numerous laws regulating the manufacture and distribution of narcotics, stimulants, depressants, hallucinogens, anabolic steroids and chemicals used in the illicit production of controlled substances (83).
The C.S.A. places all substances that in some manner are regulated under the existing Federal Law into one of five schedules. This placement is based upon the substance’s medical use, potential for abuse, its safety and its dependence liability. The Act also provides a mechanism for a substance to be controlled, i.e. added to a schedule; de-controlled, i.e. removed from control; or rescheduled, i.e. transferred from one schedule to another. The five schedules are as follows:
Schedule I. The drug has a high potential for abuse, it has no currently accepted therapeutic use in the United States, and there is a lack of accepted safety for use of the drug under medical supervision. Some Schedule I substances are heroin, LSD, and marijuana.
Schedule II. The drug has a high potential for abuse, but it has a currently accepted therapeutic use or currently accepted medical use with severe restrictions. Abuse of the drug may lead to severe psychological or physical dependence. Schedule II substances include morphine, cocaine, methadone.
Schedule III. The drug has a potential for abuse that is lower than that of the drugs in Schedules I and II. It has currently accepted therapeutic use. Abuse of the drug may lead to moderate or low physical dependence or high psychological dependence. Anabolic steroids, codeine and hydrocodone with aspirin or Tylenol, and some barbiturates, are Schedule III substances.
Schedule IV. The drug has a low potential for abuse relative to the drugs in Schedule III. The drug has currently accepted medical use in treatment. Abuse of the drug may lead to limited physical dependence or psychological dependence relative to the drugs in Schedule III. Included in Schedule IV are Darvon, Talwin, Equanil, Valium and Xanax.
Schedule V. The drug has a low potential for abuse relative to the drugs in Schedule IV. The drug has currently accepted medical use in treatment. Abuse of the drug may lead to limited physical dependence or psychological dependence relative to the drugs in Schedule IV. Over-the-counter cough medicines containing codeine are classified in this schedule.
Bulk DXP is in Schedule II, while preparations containing it (e.g. Darvon) are in Schedule IV.
Sweden has joined all three narcotic conventions of the UN, and therefore all the preparations classified as narcotic drugs in these conventions are automatically classified as such in Sweden. The National Corporation of Pharmacies has the responsibility of placing all these substances into one of the five schedules.
Summary, discussion and conclusions
The aim of our DXP project was to test the hypothesis that the side-effects of DXP constitute a serious public health problem in Sweden today. If the hypothesis was verified, we anticipated that the results would provide justification for total withdrawal of DXP from the market in the first place, or in the second place for implementation of strict regulations in the prescribing of this drug.
Analgesic dependence among chronic pain patients (1)
In our first study on DXP, interviews were conducted in 1992 with 265 orthopaedic patients and patients with chronic pain referred to a rehabilitation clinic, using a structured diagnostic instrument (ADDIS/SUDDS) in accordance with DSM-III-R (84), concerning their use of alcohol and drugs. The interview was completed by 243 patients. Twenty-two per cent of these patients met criteria for analgesic use disorders in accordance with DSM-III-R, and 18.5% fulfilled DSM-IV criteria. DXP was the most common analgesic prescribed and was used by 47% of the patients who met criteria for analgesic use disorders. Codeine was used by 18%. Two-thirds of the patients with analgesic use disorders did not have any other substance use disorder, indicating that DXP might be a drug of primary dependence.
As the result confirmed earlier findings (81, 85) that patients with chronic pain using narcotic analgesics seemed to be at considerable risk of developing DXP dependence, it was considered that a classification of preparations containing DXP into schedule IV would be justified.
The combination of high toxicity and a psychoactive effect of DXP makes it especially dangerous to individuals who have developed dependence, since one of the most important criteria of dependence is loss of control over the amount taken (ICD-10). Once a dependent person has started to take a drug with dependence liability – either alcohol or other drugs – he or she has great difficulties in discontinuing it. Many dependent individuals have described how they have tried to control their intake, for example by making a note every time they take a pill. In spite of this they may wake up hours or days later and discover that they have once more lost their control. If DXP is the drug then taken, the risk is considerable that they will never wake up again.
In a second study the prevalence of DXP in the total medico-legal autopsy material in Sweden in 1992 to 1996 was examined and characterised. (In a follow-up we studied the total medico-legal autopsy material for the period 1997-1999). Deceased persons with measurable levels of DXP in the peripheral blood were included. DXP was found in blood samples from 7.5% of the total number of blood samples. The autopsy prevalence of DXP increased by 25% during the study years.
A constantly high rate of fatal poisonings in which DXP caused or contributed to death was found (158-227 cases per year). The mean blood DXP concentration was 1.62 µg/g (the blood level of DXP after a therapeutic dose is 0.05-0.75 µg/g). The cases below 50 years of age had a significantly higher mean concentration than those above 50 years.
Autopsy cases where DXP was not found in the peripheral blood were excluded. DXP was present in hepatic blood or cardiac blood in another 115 cases. Further, we did not investigate the main metabolite norpropoxyphene, and thus cases where DXP could only be traced indirectly by its metabolite were overlooked. This meant that the prevalence might have been higher than was indicated in this study.
The high prevalence of DXP in autopsy blood points to a high prescription rate. As mentioned above, among the Nordic countries DXP is most frequently prescribed in Sweden. In 1996 the prescription rate was 14.4 DDD, while the prescription rate of the second most frequently prescribed preparation codeine was 10-11 DDD. During the years 1992-1995, 25 cases of fatal poisoning where codeine contributed to death were reported, and only one case in which codeine was judged to be the cause of death (74).
Simultaneous occurrence of drugs and alcohol is often found among self-inflicted intended poisonings. The aim of our third study was to investigate deaths not classified as suicide, and where both DXP and alcohol were found in the blood. Among the records for the years 1992-1996 we found 425 such cases. An increase in the number was observed, as half of the cases were noted during the last two years of the study period.
In 42% of the cases there was a blood alcohol concentration of less than 0.1%, in one-third this figure was more than 0.2%, and in 11% of the cases it was 0.3%. The majority of these cases were between the ages of 30 and 59 years and 71% were male. Notes on alcoholism were found in 16% of the cases.
In spite of warnings in the public drug compendium and in the corresponding drug compendium for physicians against using DXP while under the influence of alcohol or sedatives, our study showed an increase in the simultaneous occurrence of DXP and alcohol in the blood in autopsy cases during the five years. Our results confirmed those of earlier studies showing that individuals who mixed these drugs did not, as one might perhaps believe, belong to a typical street drug population, or to a group of young people experimenting with alcohol and pills, but were middle-aged men, habitual or social drinkers, on medication for pain. An average of 40 individuals died accidentally every year from the dangerous combination of DXP and alcohol, and it is therefore clear that this risk category needs to be identified by the prescribing physicians.
(Picture: “I suppose it won’t harm if I take another drink”)
The classification of the manner of death in fatal DXP poisoning (4)
In another study we analysed the 956 cases that had been classified as fatal poisoning during the years 1992-1996 and where DXP had caused or contributed to death. Among these cases the manner of death was classified as accidental in 49 (5%) of the cases, suicidal in 542 cases (57%) and undetermined in 365 cases (38%). In comparison with earlier studies (31, 36, 86) the rate of accidental death was very low, and the number of accidents was probably under-reported. The analysis of the death certificates and the police reports showed that cases likely to have been accidents often were classified as undetermined.
In Sweden cases classified as undetermined are frequently included in discussions of suicide rates (87), which in the case of DXP poisoning may lead to an underestimation of the risk of unintentional poisoning. An invalid classification may also lead to unnecessary suffering among close relatives, as a suicide by a family member probably is one of the most traumatic experiences a person can have. In many countries insurance matters are also influenced by the classification of the manner of death. Therefore, in order to classify a death as suicide, unambiguous evidence of intent of the deceased to die should be established, especially in the case of DXP poisoning, in which it is very difficult to determine whether the death was intentional or not.
In a later study (5) we found that the classification of the manner of death was often based on very limited grounds, and it is therefore important to implement new routines to enlarge the base of information.
It is often questioned whether withdrawal of a toxic medical preparation really does lead to a decrease in the number of deaths. Such questioning is often based on the presumption that the majority of the fatal poisonings are suicides, and that a person who really intends to commit suicide will find other means to do so. When it comes to DXP, we do have reasons to believe that the number of cases of accidental poisoning is considerably larger than is officially reported.
Further, it is not the intention in all suicidal attempts that they shall lead to death. Many people, especially youngsters, may act out aggression or disappointment in such attempts. If the pills they grab in their bathrooms for this purpose happen to contain DXP, the impulsive act may lead to a tragic, unwanted death. We suggest that the number of fatal poisoning cases would decrease considerably if DXP were withdrawn from the market.
Unmixed and combined DXP preparations (6)
There has been some discussion as to whether or not the combination of DXP and paracetamol implies an increased risk. According to some investigators the risk of death is doubled by use of Distalgesic, which contains both DXP and paracetamol. First, DXP may cause coma, convulsions and a rapid onset of respiratory depression and apnoea, and secondly, if the patient survives the initial phase of respiratory depression, hepatic failure from the paracetamol may occur a few days later (88).
We decided to examine the relation between the prescription of various DXP preparations and their involvement in fatal poisoning in Sweden during the years 1992-1996, to determine whether any particular kind of preparation was over-represented. There are seven preparations containing DXP on the Swedish market; in three of them DXP is the sole analgesic ingredient, while four of them are combinations of analgesics. The highest ratio of prescription rate to number of deaths, 27, was attributed to unmixed preparations. The ratio for DXP + paracetamol-related deaths was 6.3, and for DXP + phenazone 6.4, while the lowest ratio, 2, was found among the cases of DXP + chlorzoxazone. The unmixed preparations, representing 26% of all DXP prescriptions during the study years, were implicated in 62% of the DXP fatalities, a considerable over-representation.
In spite of the doubled risk associated with Distalgesic, we found that the unmixed preparations were involved in considerably more deaths when the numbers of deaths were correlated with the rates of prescription. In Sweden Distalgesic® has dominated the DXP market ever since it was introduced in 1972 (89), although it is now losing its share of the market, with a reduction from 58% in 1992 to 49% in 1997. On the other hand the unmixed preparations gained in their share of the market during the same years, from 2.5 DDD in 1990 to 4.8 DDD in 1997, an increase of 92%. Thus we may anticipate an increase in the already high DXP fatality rate unless restrictions are placed on the prescribing of this drug.
Next to DXP, antidepressants and sedatives are the most common causes of fatal poisoning in Sweden (56). In order to analyse the characteristics of DXP victims in comparison with those of victims of poisonings due to antidepressants and sedatives, a Swedish autopsy material from the six-year period from 1992 to 1997 was examined.
The comparison showed a significant over-representation of fatal DXP poisonings among younger people. The average ages of the deceased varied significantly (DXP 43 years, antidepressants 51 years and sedatives 59 years). The age distribution among the fatal DXP poisonings differed from that in Swedish sale statistics, which have shown that the prescription rate of all compounds studied increases with increasing age (89).
To explain the over-representation it may be speculated whether younger people are more prone than the elderly to abuse therapeutic drugs by reason of their euphoric effects. In order to get a kick, they might take large doses, and in the case of DXP, with its high toxicity, the overdose might lead to accidental death.
In comparison with other single preparations found in the blood of the deceased, DXP was markedly predominant in causing fatal poisonings. When the number of DXP-related deaths was considered in relation to the prescription rate in the whole of Sweden during the years 1992-1996, the ratio obtained was 66 (956 deaths/14.5 DDD). The ratio for the second most frequent preparation flunitrazepam during the same years was 20. The conclusion drawn was that DXP has a higher toxic potential than other preparations involved in fatal poisonings.
Irrespective of its causes, the considerable over-representation of fatal DXP poisoning should constitute a warning signal, and preventive measures are urgently needed.
DXP in individuals suspected of driving under influence (8)
Along with the increasing polydrug use and abuse, increasing attention has been paid to individuals driving under the influence of drugs, and in 1999 a new Swedish law came into force, implying a strengthened policy regarding driving while influenced by narcotic drugs (90).
To investigate the prevalence of analgesics containing DXP among individuals suspected of driving under the influence of drugs, we analysed all blood samples in which drugs were screened for in cases of suspected driving under influence in Sweden during the years 1992-1997. DXP was found in less than 3%, but when the number of cases was correlated to the DDD during each year of the period, the ratio trebled from 1992 to 1997, a significant increase. The increase in the prevalence of DXP supports the earlier finding of an increase in the prevalence of this drug among autopsy cases during the same period (2). In view of the high toxicity of DXP, especially when combined with alcohol or other drugs, this trend gives reason for concern, since the studied population represents a group of individuals who use large doses of therapeutic or illegal drugs.
The analgesic effect of DXP according to meta-studies
A medical preparation with such serious side-effects as those of DXP can only be justified if its therapeutic value widely exceeds the risks, and if no alternative preparation of comparable therapeutic value is available. The analgesic effects of DXP have been questioned. Comprehensive meta-studies have shown that although it has some analgesic effects, DXP is not superior to other available analgesics recommended for moderate to severe pain.
In 1970 it was concluded from a review of the literature that DXP was not superior to codeine or aspirin in terms of its analgesic effect. It was considered that factors other than an intrinsic therapeutic value were responsible for the commercial success of DXP, for example promotional efforts involving high expenditure for advertising in medical journals, detailing to physicians, and distributing free samples (91).
In a follow-up in 1977 one of the reviewing authors found that it was doubtful that DXP hydrochloride in a dose of 65 mg provided an analgesic effect equal to that of aspirin 650 mg. Further, he did not find any conclusive evidence that combinations of DXP with other analgesics were more effective than DXP or other analgesics alone (92).
Two other reviews, in 1966 and 1984, showed that DXP alone was superior to placebo in doses of 65 mg or more, but that the effects were questionable in doses below 65 mg. The author concluded that DXP was only approximately 1/2-2/3 as potent as codeine. Likewise, DXP in doses of less than 65 mg was certainly no more, and possibly less effective than the generally used doses of aspirin (93, 94).
In 1997 a systematic overview undertaken to assess the analgesic effects of adding DXP to paracetamol (only randomised controlled studies were included) revealed no objective evidence to support prescribing of a combination of paracetamol and DXP in preference to paracetamol alone for moderate pain (95).
When prescribing restrictions were implemented in Denmark in 1988, the arguments put forward against total withdrawal of DXP were that this drug did have a therapeutic value in some patients and that no satisfactory alternatives for patients with moderate pain were available. The meta-studies contradict these arguments, and today, more than ten years later, there are alternative analgesic preparations designed for the same patient category, but which lack the high toxicity of DXP and thereby are considerably less dangerous. Examples are the anti-inflammatory preparations with a prolonged effect, such as ibuprofen, which have been reported to be more effective than DXP (96-98). Another alternative is tramadol, an opioid with a lower addiction liability than DXP and without DXP’s high toxicity (99, 100). Tramadol has only recently been introduced onto the Swedish market, but in Germany, for example, it has been on the market for more than three decades.
The findings in our investigation have shown a clear and distinct pattern. DXP causes the largest number of fatal poisonings in Sweden today, not only in absolute figures, but also when the number is correlated with the prescription rate. Unmixed preparations, i. e. those containing only DXP, are involved in considerably more deaths than the combined preparations, when the number of fatal poisonings is considered in relation to the rate of prescription.
Individuals below the age of 50 years are over-represented among cases of fatal poisoning where DXP has caused or contributed to death, and middle-aged men seem to be most prone to mix alcohol and DXP without any intent to commit suicide. Further, individuals abusing drugs are at considerable risk of dying accidentally from ingestion of DXP, just like younger people experimenting with alcohol and pills or acting out aggression or disappointment.
The number of accidental DXP poisonings are under-reported, and unintended fatal poisonings are often hidden not only among the deaths classified as undetermined, but also among those interpreted as suicides.
Patients on DXP medication for pain run a considerable risk of becoming dependent on their medicine, even within the prescribed doses.
There has been a significant increase in the prevalence of DXP among autopsy cases and among individuals suspected of driving under the influence of drugs or alcohol.
Finally, we agree with the request made by several clinical researchers before us that DXP should be withdrawn from the market today and replaced by analgesics with less serious side-effects. This would save a considerable number of human lives each year.
Eight studies on dextropropoxyphene
In the following we present the papers on which this book is based. The original articles are recommended for those who whish to have more details. Permission to carry out all studies was received from the Ethical Committee of Uppsala University. The persons working in the project have been under obligation to preserve confidentiality. All information about patients has been stored in accordance with current security regulations and consideration has been paid to the privacy of the deceased and of his or her surviving relatives and friends.
I. Analgesic use disorders among orthopedic
and chronic pain patients at a rehabilitation clinic.
Substance Use & Misuse 1998; 33:1375-1385.
The main purpose of study I was to document the prevalence of past and present analgesic use disorders, including those with drugs containing DXP or codeine, among orthopaedic patients and patients with chronic pain treated at a Swedish rehabilitation clinic in 1992. The patients were referred by the Swedish National Health Insurance or by private doctors for further evaluation. Myalgia, myositis and unspecified muscular pain, disc degeneration and changes in the cervical spine were the most common diagnoses on referral.
To assess the occurrence of substance use disorders among these patients, they were all interviewed concerning their consumption of alcohol and drugs. The interview was introduced as part of the admission procedure and designed to elicit information permitting a diagnosis of substance use disorders in accordance with the DSM-III-R criteria (85). The DSM-III-R diagnosis was classified into dependence and abuse.
To evaluate the attitudes of the staff members towards the new routine, a structured interview with these members was conducted. The attitudes of the patients were evaluated by means of a questionnaire sent to them after discharge.
The interview was completed by 243 patients. Of these, 54 patients (22%) were found to show dependence or abuse of analgesics according to DSM-III-R. DXP was the preparation most often prescribed and was used by 47% of the patients fulfilling the criteria for dependence or abuse. The difference between the group with no analgesic use disorder and the group meeting the DSM-III-R criteria for such a disorder was significant (p=0.003) regarding DXP (Table 1).
Table 1. The prevalence of use of different analgesic preparations by the total group, by patients with no analgesic use disorders and by patients with analgesic use disorders.
Total No analgesic Analgesic
group disorder use disorder
(n=243) (n=189) (n=54)
DXP 32% 26% 47%*
Codeine 12% 9% 18%
Paracetamol 30% 30% 30%
ASA 5% 5% 5%
ASA=acetylsalicylic acid *p=0.0003
Patients who were dependent on analgesics and/or sedatives were helped to reduce their use of drugs during their stay at the hospital. They were offered alternative pain relievers, physical training, heat baths and medical information. In the cases where the gradual reduction was not completed, this was continued by the patient’s own general practitioner. Three patients who had developed dependence on alcohol were referred directly to the treatment centre.
Our results confirm the suggestion of earlier studies (81, 85) that patients with chronic pain are at risk of developing dependence on DXP when using it for pain relief. Two-thirds of the patients with analgesic use disorders did not have any other substance use disorder, indicating that the risk of developing an analgesic dependence syndrome among chronic pain patients is not linked to other kinds of drug problems. This finding also supports earlier results (80). Thus periodic assessment of patients using narcotic analgesic drugs is important, even when they follow the prescribed doses. The evaluation of the attitudes of the staff members and patients towards the new routine showed that both staff and patients were overwhelmingly positive to the interview routine, and the use of a structured interview for documenting drug use among patients with chronic pain can therefore be recommended.
Forensic Medicine research
II. The prevalence of dextropropoxyphene in autopsy blood samples .
Forensic Science International 1998; 96:135-142
The aim of this study was to determine the current prevalence of DXP in the total forensic material in Sweden during the five-year period from 1992 to 1996, in relation to age and blood concentration of DXP.
The analysis was made on 23,691 blood samples sent by the six forensic institutes to the national laboratory in 1992-1996. The blood samples were analysed for DXP. All cases with a measurable amount of DXP in peripheral blood were included in the study. Simultaneous findings of paracetamol and alcohol were considered in the analyses. Death certificates were analysed in all cases concerning the cause and manner of death.
DXP was found in blood samples from 1,782 (7.5%) of the 23,691 blood samples. The DXP prevalence increased significantly by 25% from 1992 to 1996. The mean blood DXP concentration was 1.62 µg/g (the DXP level in the blood corresponding to therapeutic doses ranges from 0.05-0.75 µg/g). Of the 1,782 cases 60% were men and 40% were women. The mean age was 54 years (range 15 to 96). A significantly higher DXP level (2.36 µg/g) was found in the group aged < 50 years compared with the cases > 50 years of age (1.04 µg/g).
As seen in Table 2, cases with a blood DXP level of > 0.75 µg/g were distributed fairly equally between the years 1992, 1993, 1994 and 1995 (n = 139, 162, 154 and 163 respectively), but there was an increase to 217 cases in 1996.
Table 2. The number of blood samples, the prevalence of DXP in peripheral blood and
the number of cases with a blood DXP concentration < and > 0.75 µg/g in each year between 1992 and 1996.
Blood DXP DXP < 0.75 > 0.75
samples in blood prevalence µg/g µg/g
n n %
1992* 4,511 308 6.8 169 139
1993 4,581 330 7.2 168 162
1994 4,714 359 7.6 205 154
1995 4,953 364 7.3 201 163
1996 4,932 421 8.5 204 217
total 23,691 1,782 7.5 947 835
Paracetamol was found in 53% of the 1,782 cases (mean 75 µg/g; therapeutic level 2.5-25 µg/g). The cases < 50 years of age had a significantly higher level of paracetamol than those aged > 50 years. A significantly higher level of paracetamol was also found in the group with a blood DXP level > 0.75 µg/g compared with the group with blood DXP < 0.75 µg/g.
Alcohol was found in the blood in 43% of the 1,782 cases (mean blood alcohol concentration 0.14%). A significantly higher percentage of the cases < 50 years of age than of those aged > 50 years had alcohol in the blood, and this was also true in the group with blood DXP > 0.75 µg/g compared with < 0.75 µg/g.
Fifty-four per cent (956) of the 1,782 cases died of fatal poisoning, and in 707 (74%) of these the blood DXP level was > 0.75 µg/g.
Discussion and conclusion
The prevalence of DXP in autopsy cases in our study increased by 25% from 1992 to 1996 and the study confirmed the constant rate of 112-160 cases per year of fatal poisoning with DXP in Sweden since the early 1970s (72). The mean death rate among the fatal poisonings with a blood DXP concentration > 0.75 µg/g in this study was 141 cases per year. The follow up during the years 1997-1999 showed that the prevalence increased significantly by 35% from 1992 to 1997 (9.2%) and that the number of fatal DXP poisonings remained high during the years 1997, 1998 and 1999 (224, 195 and 217 respectively).
A striking finding was the significantly higher levels of both DXP and paracetamol in cases < 50 years of age. This together with the significantly higher prevalence of alcohol in the blood in cases with DXP may confirm some alarming reports about young people using analgesics to potentiate the effects of alcohol (101). The high availability of DXP preparations makes it a drug of great interest to adolescents (102). The constantly high rate of fatal DXP poisonings cannot be expected to decrease unless preventive measures are taken. Other Scandinavian countries, such as Norway and Denmark, have managed to decrease the rate of fatal DXP poisonings through government regulations for prescription. As the defined daily dose of DXP preparations in Sweden is six times as high as in Denmark and nine times as high as in Norway (Table 3), the introduction of similar regulations in Sweden should be considered.
Table 3. Sales of dextropropoxyphene in different Scandinavian countries. DDD = defined daily dose/1000 inhabitants during a 12 - month period.
1990 1994 1996 mean
DDD DDD DDD DDD
Sweden 13.9 14.7 14.4 14.5
Norway 1.6 1.6 1.6 1.6
Denmark* NA 2.7 2.0 2.4
*) use in hospital not included. NA = not available
III. Middle-aged men – a risk category regarding fatal poisoning due to dextropropoxyphene and alcohol in combination.
Preventive Medicine 2000; 31:103-6
The aim of study III was to determine the characteristics of non-suicidal deceased persons with both alcohol and DXP in the blood, among the total medico-legal autopsy material in Sweden during the years 1992 to 1996. The requirements for inclusion in the study were that DXP was present in peripheral blood, that a blood alcohol concentration > 0.01% was found, and that death was not classified as suicide.
As mentioned above (study II), in the study above DXP was found in the blood in 1,782 cases, and in 766 of these the death was classified as suicide. These 766 cases were therefore excluded from study III. Alcohol was noted in 425 of the remaining cases. Almost half of the 425 cases (47%) with both alcohol and DXP in the peripheral blood were found in the two last years (1995-1996). The majority of the cases were 30–59 years old and 71% were male (Fig. 1). The mean blood alcohol concentration was 0.14%. Notes on alcoholism were found in 16% of the cases and drug addiction in 8%.
Figure 1. The age distribution of males and females with both DXP and alcohol in the blood.
n = 425
The cause of death was classified as fatal poisoning, with DXP causing or contributing to death, in 220 (52%) of the deceased. The manner of death among these 220 cases was accidental in 26 cases (12%) and undetermined in 194 (88%). Among the 205 (48%) cases not classified as fatal poisoning the causes of death were distributed as follows: 83 (19.5%) natural, 25 (6%) chronic alcoholism, and 97 (22.5%) other injuries.
Discussion and conclusion
As many as 42% of the analysed cases had a blood alcohol concentration of < 0.1% (29% < 0.05%), and of these 12.5% were classified as alcoholics and 10% as drug addicts. Thus the majority of these 42% were “normal” drinkers, without a developed alcohol tolerance. We hypothesise that such persons mix DXP and alcohol when they are in pain, a possibility which might be supported by the 83 cases classified as natural deaths, or when they are in situations where it is common to use alcohol.
The fact that the majority (71%) of the fatal poisoning cases were between 30 and 59 years old indicates that death from the effects of a combination of DXP and alcohol does not typically occur among young people, but rather among middle-aged men (71% were male).
The alcohol consumption in Sweden is among the lowest in the European Union (EU). The general drinking pattern, however, which is characterised by occasional heavy consumption (compared with the pattern in many other countries where frequent but low consumption is more common), constitutes a high risk when it comes to fatal poisoning caused by combination of alcohol and other drugs.
In the public drug information compendium (FASS), the presentation of DXP preparations includes a warning against using the drug while under the influence of alcohol or sedatives. In the corresponding drug compendium for physicians, the information on all DXP preparations includes a warning against prescribing the drug to alcoholics or drug addicts. The increasing proportion of cases with both alcohol and DXP in the blood, during the years of our study shows that warnings to patients and recommendations to physicians are not sufficient to lead to any decrease in death rates due to simultaneous poisoning with alcohol and DXP.
In the absence of strict regulations by the authorities concerning prescription of DXP, it should be the responsibility of the physicians before prescribing this drug to ask the patient whether he or she uses alcohol, to describe the life-threatening risks of simultaneous intake of alcohol and to actually ask the patient if he or she is willing to abstain from alcohol during the treatment with DXP. If the patient seems to be unsure or negative about being abstinent, the physician should consider a less hazardous analgesic.
IV. The manner of death among fatalities
where dextropropoxyphene caused or contributed to death
Forensic Science International 1998; 96:181-187
The aim of study IV was to investigate the manner of death among fatalities where DXP caused or contributed to death (= DXP fatality) in Sweden during the five-year period from 1992 to 1996. As mentioned above (study II), the certifying physicians classified 956 cases as DXP fatalities. The autopsies were performed by a total of 40 physicians.
Figure 2 shows the distribution by age of the 956 autopsy cases classified as DXP fatalities. Most cases, 229 (24%), were found in the age-group 40-49 years.
Figure 2. Histogram showing the number of fatalities due to DXP in the different age-groups. n = 956
The mean age of the 956 fatal DXP poisoning cases was 47 years (range 15-96) and 56% were men and 44% women. The mean DXP concentration in the blood in the DXP fatalities was 2.60 µg/g.
Among the 956 DXP fatalities, the manner of death was recorded as accidental in 49 cases (5%), suicidal in 542 (57%) and undetermined in 365 (38%). Table 4 presents the mean frequency and range of the three manners of death in the different forensic medicine districts during the study years (Fig. 3). The rate of accidental death differed significantly between district C and districts E and F. District B had a significantly higher rate of suicide than all the other districts apart from D, while district E had a significantly lower rate than all the others.
Table 4. The mean frequencies and ranges of the different manners of death among fatalities due to DXP, in the six forensic medicine districts A-F, during the study years 1992 – 1996 (%).
District Accident Range Suicide Range Undetermined Range
% % % % % %
A 6 0 – 10 54 37 -75 40 17 – 53
B 5 0 - 11 73 59 - 88 22 8 – 30
C 9 2 - 17 60 55 - 67 31 16 - 48
D 6 0 - 11 62 48 - 79 32 12 – 52
E 1 0 - 3 33 12 - 50 66 47 - 88
F 1 0 - 3 55 41 - 63 44 37 - 56
Mean 5 3 - 9 57 53 - 60 38 34 - 44
Among the 25 physicians with ten or more autopsies, the accident classification rate varied from 0% to 17%, the suicide classification rate from 25 to 83% and the rate of undetermined manner of death from 8% to 71%.
Discussion and conclusion
In our study the percentage number of deaths classified as accidental was very low, a result quite different from that of earlier studies (31, 85), and we suggested that accidents were probably under-reported. We discovered a number of deaths among those classified as undetermined that might have been classified as accidental if the physicians had had access to more information.
Figure 3. The six forensic medicine districts
Our study supports some alarming reports that have shown that some young people use analgesics to potentiate the effects of alcohol (101, 102). In our study as many as 64% of the fatal poisoning cases of ages 15-29 years had alcohol in the blood. In 5% of the cases the manner of death was classified as accidental, in 59% as suicidal and in 36% as undetermined. We suggest that rate of accidental death among these young people was probably underestimated and that many of these young men and women died without any intention to do so.
In Sweden the alternative to a report of suicide on death certificates in fatal poisoning cases has become an undetermined manner of death, thus reducing judgements of deaths as being accidental to a minimum. Further, under-reporting of accidents will increase the risk that knowledge of the high toxicity will not reach the population consuming this drug, nor will it reach the physicians, the producers or the responsible authorities.
Although the rates of the different manners of death were fairly stable during the study years in all Sweden, our investigation showed that the classification criteria seemed to vary between the Swedish forensic medicine districts and between the individual physicians.
Since valid death statistics concerning the manner of death in DXP fatalities are needed to provide a basis for preventive measures, special attention should be paid to the classification process, in order to increase the uniformity of the assessments among the different physicians, and also to avoid under-reporting of accidental cases.
V. Suicides may be over-reported and accidents under-reported
among fatalities due to dextropropoxyphene.
Journal of Forensic Sciences 1999; 44:334-338
In order to analyse the basis of the process leading to classification of the manner of death in DXP fatalities, a set of operational criteria (103) was applied retrospectively to an autopsy material at one of the six departments of forensic medicine in Sweden (Uppsala) examined during the six-year period from 1992 to 1997.
These criteria were divided into explicit and implicit expressions of the deceased person’s intent to die. Unambiguous verbal and written statements made by the deceased showing his or her intent to die were regarded as explicit criteria, while statements from relatives, acquaintances and others formed the basis of the implicit criteria. Implicit evidence of the deceased person’s intention to die was condensed into 11 criteria, for example expressions of hopelessness, a previous history of suicide attempts or suicide threats, a history of stressful events or a significant loss, and a history of serious depression or a mental disorder.
During the six-year period 1992-1997, a total of 4,306 autopsies were performed at the forensic medicine department. Among these deaths, 113 (2.6%) were classified as fatal DXP poisoning. Suicide was recorded in 84 (74%) of the 113 DXP fatalities and an undetermined manner of death in 24 (21%). In three of the cases the death was judged to be accidental. Overdose of DXP was reported to be a contributory cause of death in two cases, where chronic alcoholism was judged to be the primary cause.
Explicit expressions of the deceased’s intent to die were documented in 29 (26%) of the 109 analysed cases (In four cases no analysis could be performed). In 46 cases only implicit and no explicit criteria were found. The total number of implicit criteria in individual cases never exceeded three and in 34 cases no criteria of any type were documented.
The mean blood DXP concentration in the 109 cases where toxicological analyses were available was 3.33 µg/g. The mean blood DXP concentration was significantly higher (95% confidence interval) among the suicide cases (3.74 µg/g) than among those classified as undetermined (1.90 µg/g).
Discussion and conclusion
The classification of the manner of death in DXP fatalities was often based on very limited grounds. Information from relatives, friends and others concerning the deceased person was rarely available. The shortage of information probably led to deficiencies in the death statistics concerning DXP fatalities.
Considerable under-reporting of accidents and probable over-reporting of suicides were found. In the 34 cases where no criteria of any type were documented, 15 cases were classified as suicide and 14 as undetermined, showing that suicide was chosen as often as undetermined as the manner of death even when no suicidal criteria supported the classification.
A high blood DXP concentration per se seemed to be a criterion of suicidal intent. When it comes to addicts it is difficult to interpret drug concentrations, since development of tolerance and loss of control of the amount taken are important criteria of substance dependence (ICD-10). These dependence syndrome behaviours might lead to accidental fatalities, especially when DXP is the consumed drug.
A case where a DXP fatality in a drug addict was classified as suicide is presented to illustrate the difficulty in interpreting the blood DXP concentration.
A man aged 39 died of poisoning. A blood alcohol concentration of 0.05% and a blood DXP concentration of 4.5 µg/g were found. According to hospital records he was a known alcoholic and abuser of therapeutic drugs, and treatment with the anti-alcoholic agent disulfiram had been withdrawn within the last two weeks. The man died in the apartment of his wife’s grandmother. He had gone there to use her telephone. He made some telephone calls, and suddenly the grandmother heard a sound, and found him lying on the floor, without any signs of life. On arrival of the emergency team he was dead.
The death was classified as suicide. This could instead be an example of an accidental death caused by a combination of DXP and alcohol. Since he was an abuser of therapeutic drugs, his tolerance of DXP was probably increased, but when the disulfiram was withdrawn and he started to drink alcohol again, the combination became lethal.
In order to guarantee valid death statistics, the amount of information constituting the basis of the certifying process has to be enlarged. This would require implementation of new routines, including interviews of relatives, acquaintances and significant others, for example psychiatrists, to obtain the information required to assess the deceased’s intention to die. This is especially important when DXP is the consumed drug.
A set of operational criteria would facilitate the difficult certifying process, and the authors recommend use of the set of explicit and implicit criteria applied in this study.
VI. Correlation between prescription of various Dextropropoxyphene
preparations and their involvement in fatal poisonings.
Forensic Science International 1999; 103:125-132.
The aim of this study was to examine the relation between the prescription of various DXP preparations and their involvement in fatal poisoning in Sweden during the years 1992-1996, to determine whether any particular kind of preparation is over-represented.
DXP is the sole analgesic ingredient in Dexofen (Astra) and Doloxene (Eli Lilly), each containing 50/100 mg DXP, and in Dolotard (Nycomed), containing 150 mg DXP. As Dolotard represented only 4% of the sales of unmixed preparations during the study years, this category consisted mainly of Dexofen and Doloxene.
The four combined preparations are Doleron (Astra), containing 100 mg DXP, 350 mg acetylsalicylic acid (ASA) and 150 mg phenazone; Paraflex Comp (Astra), containing 45 mg DXP, 500 mg ASA and 125 mg chlorzoxazone; and Distalgesic (Eli Lilly) and Dexodon (Tika), containing 32.5 mg DXP and 325 mg paracetamol. Prescriptions of Dexodon (not introduced onto the Swedish market until 1995) constituted less than 4% of all prescriptions of DXP + paracetamol combinations in 1996.
Specifications of the therapeutic substances primarily responsible for the poisoning are not always stated on the death certificate, which is the reason why all fatal poisoning cases where DXP was present in peripheral blood were included in the study. Since alcohol and other drugs can potentiate the effects of DXP so that it becomes fatal even in doses within the therapeutic limits, cases with DXP levels below the upper therapeutic limit (0.75 µg/g) were also included.
The assessment of which DXP preparation the deceased had consumed was based on toxicological analyses. When no other analgesic compound was found besides DXP in the peripheral blood, we concluded that the consumed drug was DXP alone.
It was found to be a complicated question to decide whether paracetamol in the blood indicated intake of a compound consisting of DXP + paracetamol, or whether it meant that the deceased had consumed an unmixed DXP preparation together with some other preparation containing paracetamol. We therefore calculated the ratios between paracetamol and DXP in the blood in 30 cases where, according to the death certificate, a compound containing both DXP and paracetamol was the consumed drug. The mean ratio of paracetamol/DXP in these samples was 80 (range 35-183).
A compound consisting of DXP + paracetamol was believed to have been used when the paracetamol/DXP ratio in the peripheral blood was in the range 35-185. With ratios < 20 the consumed drug was considered to have been an unmixed preparation. Cases in the range 20-35 and > 185 were not categorised, since they could have belonged to either of the two categories.
The study comprised 834 cases. As shown in Table 5, the highest mean ratio of DXP-related deaths/DDD, 27, was found among the unmixed preparations. The death ratio attributed to DXP + paracetamol was 6.3 and that to DXP + phenazone, 6.4, while the lowest ratio, 2, was related to DXP + chlorzoxazone. The death ratio of each of the preparations seemed to be fairly constant during the study years (Table 5).
Table 5. Distribution of fatal poisonings ascribed to DXP alone, DXP + paracetamol, DXP + phenazone and DXP + chlorzoxazone during the years 1992-96, the prescription rate in defined daily doses (DDD), the number of fatal poisoning per DDD (n/DDD), and the mean ratios of each preparation during the 5-year period
Year Total DXP alone DXP + paracetamol DXP + phenazone DXP + chlorzoxazone
n DDD n/DDD n DDD n/DDD n DDD n/DDD n DDD n/DDD
1992 132 80 3.2 25 38 8.8 4.3 10 1.4 7.1 4 1.7 2.3
1993 177 92 3.3 27.8 72 8.5 8.5 10 1.4 7.1 3 1.5 2.3
1994 155 91 3.8 24 56 8.4 6.6 4 1.0 4.0 4 1.4 2.9
1995 170 118 4.2 28 44 8.5 5 6 0.9 6.7 2 1.3 1.5
1996 200 136 4.6 29.5 57 8.0 7 5 0.7 7.1 2 1.3 1.5
Mean 834 517 3.8 27 267 8.4 6.3 35 1.1 6.4 15 1.4 2.0
Discussion and conclusions
The unmixed DXP preparations were considerably over-represented among the fatal poisonings when related to the prescription rate. If all the excluded cases with ratios between 20 and 35 and >185 had been included in the DXP + paracetamol category, the death ratio would have been 8.7, which would still be one-third of the ratio of unmixed preparations. Any other distribution of the 98 cases would lead to an increase in the death ratio attributable to preparations containing DXP alone. Thus the main finding in our study would not be invalidated by the inclusion of these cases.
Unmixed preparations, with their higher content of DXP, may be more attractive to many consumers because of their narcotic (euphoric) effects rather than by reason of any analgesic superiority. Another possibility is that unmixed preparations may erroneously have been considered safer than combinations of DXP and paracetamol, as reports of poisonings have been most frequently concerned compounds containing both drugs, probably because of the predominance of these mixed compounds on the market. In Sweden Distalgesic has dominated the DXP market ever since it was introduced in 1972, although it is now losing ground. On the other hand the market share of the unmixed preparations is increasing, and has doubled from 1990 to 1997, a fact which could probably lead to an even higher rate of fatal DXP poisoning.
VII. Among fatal poisonings dextropropoxyphene predominates in younger people, antidepressants in the middle-aged and sedatives in the elderly.
Journal of Forensic Science 2000; 45:7-10.
Next to DXP, antidepressants and sedatives are the most common causes of fatal poisoning in Sweden (73). In order to analyse the characteristics of DXP victims in comparison with those of victims of poisonings due to antidepressants and sedatives, a Swedish autopsy material from the six-year period 1992 to 1997 was examined. Cases where DXP, antidepressants or sedatives contributed to death in combination with each other or with other drugs were excluded.
During 1992-1997, a total of 4,306 autopsies were performed at the forensic medicine department under study. In 202 cases DXP, antidepressants or sedatives caused death. Among these, DXP caused death in 78 cases (39%), antidepressants in 49 cases (24%) and sedatives in 75 cases (37%). The figure for DXP may be compared with the 30 cases (15%) caused by the second most common compound found, flunitrazepam.
The victims of poisoning by DXP, antidepressants or sedatives shared a similar history of alcohol/drug abuse (20%), depression (31%) and somatic illness (25%). They were mostly living alone at the time of death (> 60%), the majority died at home (81%), and suicide was the most frequent manner of death (73%).
The mean age of the cases of DXP fatalities (43 years, confidence interval [CI] 39-47) was significantly lower than that of the cases of fatal poisoning due to antidepressants (51 years, CI 48-54) and sedatives (59 years, CI 55-63), and the mean age of the sedative poisoning cases was significantly higher than that of the other two drug categories.
The age differences between the three types of drug poisoning are illustrated in Figure 4. There was a predominance of DXP in the three younger age groups (75% of all cases), a fairly equal distribution of the three drugs in the age group 40-49 years and a predominance of antidepressants and sedatives in the age group 50 to 59 years. In the older age groups, above 59 years, the majority of the poisonings were due to sedatives (77% of all cases).
Figure 4. The distribution of fatal poisoning by age group and type of drug. n = 202
Discussion and conclusions
Age seemed to be an important characteristic when it came to the choice of drug. The mean age of the DXP cases was significantly lower, and that of the sedative cases significantly higher than the mean ages of the other two drug groups. Swedish sales statistics have shown that the prescription rate of all the compounds studied increases with increasing age (89). Table 6 shows the distribution of the prescription rates of DXP, antidepressants and benzodiazepines by age in 1996 in the studied area.
Table 6. The prescription rates of DXP, antidepressants and benzodiazepines, in DDD, in 1996 in the studied area.
Age-group, yrs DXP Antidepressants Benzodiazepines
20 - 29 2.0 12.7 4.1
40 - 49 9.2 44.0 30.
70 - 79 41.7 58.0 97.0
The predominance of sedatives as cause of death among the elderly in cases of fatal poisoning might be explained by a very high prescription rate of such drugs at older ages, but the prescription rate of DXP could not explain the predominance of DXP fatalities among younger people. It might rather have been expected that the antidepressants, which were mostly prescribed among younger people, would have been predominant in the younger group.
An analysis of the occurrence of mental disorders in relation to age revealed no differences that could explain the differences in drug choice. Somatic diseases were found in significantly more cases among the elderly, giving no explanation for why DXP was used more by younger persons.
One reason for the over-representation of younger people among the DXP victims could be that they are more prone than the elderly to abuse therapeutic drugs by reason of their euphoric effects. In order to get the kick, they might take large doses, and in the case of DXP, with its high toxicity, the overdose might lead to accidental death. Sedatives are not as toxic as DXP, and victims of accidental overdose of sedatives are thus more often rescued. Since antidepressants lack direct euphoric effects, they are not abused to the same extent as DXP and sedatives, which may explain the predominance of DXP in fatal poisonings among younger people in spite of the higher prescription rate of antidepressants.
VIII. The prevalence of analgesics containing dextropropoxyphene or codeine in individuals suspected of driving under the influence of drugs
Forensic Science International 2000; 112:163-169
The aim of study VIII was to investigate the prevalence of analgesics containing DXP or codeine in the blood among individuals suspected of driving under the influence of drugs during the years 1992-1997.
During the years 1992-1997, drugs were screened for in 4,896 cases. DXP was found in 130 (2.7%) and codeine in 388 (7.9%) of these blood samples In 32 cases both drugs were found.
Table 7 gives the number of cases involving each preparation in relation to the DDD during each year of the period 1992-1997. The mean ratios for DXP and codeine for this six-year period were 1.6 and 5.9 respectively. The DXP ratio trebled from 0.99 in 1992 to 2.89 in 1997, while the codeine ratio showed a slight decrease, by 9% (Table 7).
Table 7. Distribution of cases with DXP and codeine during the years 1992-1997. The prescription rates of the respective drugs, presented as defined daily dose per 1000 inhabitants and year (DDD), and the number of cases per DDD (N/DDD), are given. The mean ratios for DXP and codeine during the 6-year period are also given.
DXP DDD/ N/DDD Codeine DDD/ N/DDD
N 1000 inh. (DXP) N 1000 inh. (codeine)
1992 16 16.3 0.99 61 10.1 6.04
1993 15 12.8 1.17 52 10.4 6.04
1994 18 13.9 1.30 85 11.3 7.52
1995 19 13.8 1.37 66 11.6 5.69
1996 27 13.7 1.97 64 11.3 5.67
1997 35 12.1 2.89 60 10.9 5.50
mean 1.62 5.90
Other drugs were found to the same extent among the DXP and codeine cases and benzodiazepines were present in as many as 346 (71%) of the 486 cases. Amphetamine and/or cannabis was found in 184 cases (38%).
Discussion and conclusions
The prevalence of DXP (< 5%) and of codeine (< 9%) among individuals suspected of driving under the influence of drugs was rather low during the study years, and it is concluded that analgesics containing DXP or codeine are not drugs of primary interest in this specific population.
Although the mean ratio of codeine cases to DDD was almost four times that of DXP during this period, the DXP/DDD ratio trebled, while the codeine ratio seemed fairly constant, with a slight decrease (9%) between 1992 and 1997. The significant increase in the prevalence of DXP supports the earlier finding of an increase in the prevalence of this drug among autopsy cases during the same period (2). In view of the high toxicity of DXP, especially when combined with alcohol or other drugs, this trend gives reason for much concern, since the studied population represents a group of individuals who use large doses of therapeutic or illegal drugs.
1. Jonasson U, Jonasson B, Wickström L, Andersson E, Saldeen T. Analgesic use disorders among orthopedic and chronic pain patients at a rehabilitation clinic. Substance Use & Misuse 1998; 33:1375-1385.
2. Jonasson U, Jonasson B, Holmgren P, Saldeen T. The prevalence of dextropropoxyphene in autopsy blood samples. Forensic Sci Int 1998; 96:135-142
3. Jonasson U, Jonasson B, Saldeen T. Middle-aged men - a risk category regarding fatal poisoning due to dextropropoxyphene and alcohol in combination. Prev Med 2000; 31:103-6.
4. Jonasson B, Jonasson U, Saldeen T. The manner of death among fatalities where dextropropoxyphene caused or contributed to death Forensic Sci Int 1998; 96:181-187
5. Jonasson B, Jonasson U, Saldeen, T. Suicides may be over-reported and accidents under-reported among fatalities due to dextropropoxyphene. J Forensic Sci 1999; 44:334-338
6. Jonasson U, Jonasson B, Saldeen T. Correlation between prescription of various dextropropoxyphene preparations and their involvement in fatal poisonings. Forensic Sci Int 1999; 103:125-132.
7. Jonasson B, Jonasson U, Saldeen T. Among fatal poisonings dextropropoxyphene predominates in younger people, antidepressants in the middle aged and sedatives in the elderly. J Forensic Sci 2000; 45:7-10.
8. Jonasson U, Jonasson B, Saldeen T. The prevalence of analgesics containing dextropropoxyphene or codeine in individuals suspected of driving under the influence of drugs. Forensic Sci Int 2000; 112:163-169.
9. Jonasson B. Studies on dextropropoxyphene – with special reference to dependence among chronic pain patients, classification of the manner of death in fatal poisoning and characteristics of the fatal poisoning victims. Acta Universitatis Upsaliensis, Uppsala, 2000.
10. Obafunwa JO, Busuttil A, Al-Oqleh AMH. Dextropropoxyphene-related deaths – a problem that persists? Int J Leg Med 1994; 106:315-318.
11. Buckley NA, Whyte IM, Dawson AH, McManus PR, Ferguson NW. Correlations between prescriptions and drugs taken in selfpoisoning - Implications for prescribers and drug regulation. Med J Australia 1995; 162: 194-19.
12. Dore GM. The dangers of dextropropoxyphene. Australian and New Zealand Journal of Psychiatry 1996; 30:864-866.
13. Soumerai SB, Avorn J, Gortmaker S, Hawley S. Effect of government and commercial warnings on reducing prescription misuse: The case of propoxyphene. Am J of Public Health 1987; 77:1518-1523
14. The Swedish National Board of Forensic Medicine. The development of Swedish forensic medicine – the history, the present situation and the future. (Swe.). Summary in English. RMV-report 1997:1, Stockholm, 1997.
15. Jonasson U. Drugs that can create dependency. (Swe.). The Education Company of Brevskolan, Stockholm, 1995.
16. Pohland A., Sullivan HR. Analgesics: Esters of 4-dialhylamino-1.2-diphenyl- 2butanols. J Am Chem Assoc 1953; 75:4458-4461.
17. Kautz HD. New and nonofficial drugs. JAMA 1958; 166:1481-1483.
18. Gibson TP, Giacomini KM, Briggs WA, Whitman W, Levy G. Propoxyphene and norpropoxyphene plasma concentrations in the anephric patient. Clin Pharm Ther 1980;27:665-670.
19. Inturrisi CE, Colburn WA, Vereby K, Dayton HE, Woody GE, O´Brien CP. Propoxyphene and norpropoxyphene kinetics after single and repeated doses of propoxyphene. Clin Pharm Ther 1982;31:157-167.
20. Young RJ. Dextropropoxyphene overdosage: pharmacological considerations and clinical management. Drugs 1983; 26:70-79.
21. Lawson AAH, Northridge DB. Dextropropoxyphene overdose, epidemiology, clinical presentation and management. Med Toxicol 1987;2:430-444.
22. Cann HM, Verhulst HL. Convulsions as a manifestation of acute dextropropoxyphene intoxiction. Am J Dis Child 1960; 99:380-382.
23. Hyatt HW. Near fatal poisoning due to accidental ingestion of overdose of dextro-propoxyphene hydrochloride by two-years-old child. New Eng J Med 1962;267:710.
24. Querish EH. Propoxyphene hydrochloride poisoning. JAMA 1964; 188:170-171.
25. Cawood R, Thirkettle JL. Poisoning by propoxyphene hydrochloride (Doloxene). Br Med J 1966; 1324.
26. McCarthy GWH, Keenan MRL. Propoxyphene hydrochloride poisoning. J Am Med Ass. (JAMA) 1964; 187:460-461.
27. Lund A. Nielsen G.D. Dextropropoxyphene concentrations in blood in cases of fatal poisoning. Zeitschrift für Rechtsmedizin 1972; 71:148-152.
28. Christensen H. Dextropropoxyphene and norpropoxyphene in blood, muscle, liver and urine in fatal poisoning. Acta pharmacol et toxicol 1977; 40:298-309.
29. Obafunwa JO, Busuttil A, Al-Oqleh AMH. Dextropropoxyphene-related deaths -
a problem that persists? Int J Leg Med 1994; 106:315-318.
30. Young RJ, Lawson, AAH. Distalgesic poisoning - cause for concern. British Medical Journal 1980; 1:1045-1047.
31. Whittington, R.M. and Barclay, A.D. The epidemiology of dextropropoxyphene (Distalgesic) overdose fatalities in Birmingham and the West Midlands. Journal of Clinical and Hospital Pharmacy 1981; 6:251-257.
32. O´Carroll PW. A consideration of the validity and reliability of suicide mortality data. Suicide and Life-Threatening Behavior 1989;19:1-15.
33. World Health Organisation, International Statistical Classification of Diseases and Related Health Problems, 10th Rev., ICD-10. World Health Organisation, Geneva, WHO, 1992.
34. Sundkvist L, Petrovics J. Fatal poisoning with dextropropoxyphene- containing analgesics - suicide or not? Acta Med Scand 1974;196:467-472.
35. Finkle BS. Self-poisoning with dextropropoxyphene and dextropropoxyphene compounds: the USA Experience. Human Toxicol 1984; 3:115-134.
36. Kaa E, Dalgaard JB. Fatal dextropropoxyphene poisoning in Jutland (Denmark).
Z Rechtsmed 1989; 102:107-115.
37. Sturner WQ, Garriott JC. Deaths involving propoxyphene, a study of 41 cases over a two-year period. JAMA 1973; 223:1125-1130.
38. Finkle BS, McCloskey KL, Kiplinger GF, Bennett IF. A national assessment of propoxyphene in postmortem medicolegal investigation, 1972-1975. J Forensic Sci 1976; 21:706-741.
39. BS, Caplan YH, Garriott JC, Monforte JR, Shaw RF, Sonsalla PK. Propo-xyphene in postmortem toxicology 1976-1978. J Forensic Sci 1981; 26:739-757.
40. Hudson P, Barringer M, McBay AJ. Fatal poisoning with propoxyphene: Report from 100 consecutive cases. South Med J 1977; 709:938-942.
41. Carson DJL, Carson ED. Fatal dextropropoxyphene poisoning in Northern Ireland. Review of 30 cases. The Lancet 1977; 894-897.
42. Oliver JS, Smith H, Yehia B. Poisoning by dextropropoxyphene and paracetamol.
J Forensic Sci Soc 1980; 21:207-210.
43. Young RJ. Dextropropoxyphene overdosage: pharmacological considerations and clinical management. Drugs 1983; 26:70-79.
44. Osselton MD, Blackmore RC, King LA, Moffat AC. Poisoning-associated deaths for England and Wales between 1973 and 1980. Human Toxicol 1984; 3:201-221.
45. Meredith TJ, Vale JA. Epidemiology of analgesic overdose in England and Wales. Human Toxicol 1984; 3:61-74.
46. Buckley BM, Vale JA. Dextropropoxyphene poisoning: problems with interpretation of analytical data. Human Toxicol 1984; 3:95-101.
47. Vale JA, Buckley BM, Meredith TJ. Deaths from paracetamol and dextropro-poxyphene (Distalgesic) poisoning in England and Wales in 1979. Human Toxicol 1984; 3:135-143.
48. Whittington RM. Dextropropoxyphene deaths: Coroner´s report. Human Toxicol 1984; 3:175-185.
49. Roberts I, Flanagan RJ, Strang J, Prescott LF. Toxicological studies in a Distalgesic addict. Human Toxicol 1986; 5:207-210.
50. Ruane BJ, Glover G, Varma MPS. Survival after an overdose of Distalgesic (dextropropoxyphene and paracetamol). The Ulster Med J 1989; 58:187-189.
51. Simonsen J, Pedersen P. Accidental deaths from poisoning with dextropropoxyphene. Ugeskr Laeg 1975; 137:2605-2609.
52. Bjaeldager PAL, Breum L, Munck LK, Nordestgård AG, Hunding A. Cases of oral propoxyphene poisoning in the Centre for the treatment of poisoning in 1980. Ugeskr Laeger 1985; 147:2968-2971.
53. The Danish National Board of Health, J nr 2699-11-1983, Copenhagen,
54. Gertinger P, Theilade P. Mode of death in poisoning with dextropropoxyphene. Ugeskr Laeg 1983; 145:1294-1297.
55. Schou JS. Dextropropoxyphene - "the killer drug" (Dan.) I Praktisk laegemiddelinformation Kampmann JP (ed), Köpenhamn, Danmark, 1986.
56. Melsen NC, Schurizek BA. Poisoning with dextropropoxyphene. (Dan.) Ugeskr Laeg 1987; 149:1634-1637.
57. Schou JS. Dextropropoxyphene - a medical drug of danger. (Dan.) Ugeskr Laeg 1987; 149:1624.
58. Theilade P. Death due to Dextropropoxyphene: Copenhagen experiences. Forensic Sci Int 1989; 40:143-151.
59. Kaa E, Dalgaard JB. Deaths from dextropropyxyphene in Jylland in 1985. (Dan.) Ugeskr Laeger 1987; 149: 1629-1634.
60. Ott P, Hansen PB, Dalhoff KP, Loft SH, Poulsen HE. Analgetics: Consumption of, overdose and fatal poisoning with analgesics in Denmark 1979-1986. Ugeskr Laeger 1990; 152:250-252.
61. Christensen CB, Helleberg L. The analgesic effect of dextropropoxyphene, The Department of Pharmacology, the University of Copenhagen, and the gout hospital in Skaelskoer, 1987.
62. Kaa E, Gregersen M. Fatal poisoning in Jutland (Denmark) during 1980s. Int J Leg Med 1992; 105:133-138.
63. Kaa E. Drug abuse in Western Denmark during the eighties. II. Fatal poisonings among drug abusers. Forensic Sci Int 1992; 55:75-82.
64. Segest E, Harris CN, Bay H. Dextropropoxyphene deaths in Denmark from the health authority point of view. Med Law 1993; 12:141-151.
65. Worm K, Steentoft A. Deaths from poisoning in East Denmark 1984-1991. (Dan.) Ugeskr Laeger 1994; 156:3039-3043.
66. Workshop in Fredriksborg, Denmark, 1998.
67. Jöldal B, Halvorsen I. Sales Statistics in the control of drug abuse in Norway.
Bulletin on Narcotics 1982; 34:57-68.
68. Teige B. Sudden deaths from drugs and alcohol outside hospital (Nor.) Reykjavik: Nord Retsmed Foren, 1985.
69. The Norwegian Board of Health (Norska Medicinaldepot AS) Oslo. Norway, 1997.
70. Teige B. Personal communication, 2000.
71. Grönvall G, Malmlund H-O, Matell G. Acute poisoning with drugs containing dextropropoxyphene. Opusc Med Bd 1971; 16:338-347.
72. Gunne LM, Petrovics J, Änggård E. Dextropropoxyphene - assessment of risks of abuse and intoxication. Läkartidningen 1976; 73:1731-1735.
73. Persson H, Irestedt B. Deaths caused by dextropropoxyphene poisoning. Swedish Poison Information Centre 1993.
74. Druid H, Holmgren P. A compilation of fatal and control concentrations of drugs in postmortem femoral blood. J Forensic Sci 1997; 42:79-87.
75. Fraser HF, Isbell H. Pharmacology and addiction liability of dl- and
d-propoxyphene. Bulletin on Narcotics 1960; 9-14.
76. Elson A, Domino EF. Dextro propoxyphene addiction. Observation of a case. JAMA 1963; 183: 186-189.
77. Claghorn JL, Schoolar JC. Propoxyphene hydrochroride, a drug of abuse. J of Am Med Ass (JAMA). 1966; 96:1089-1091.
78. Collins MS, Mathis JL. Addiction to intravenous Darvon, case report. Oklahoma State Med. Assoc. J 1967; 60:609.
79. Wolfe RC, Reidenberg M, Vispo RH. Propoxyphene (Darvon®) addiction and withdrawal syndrome. Ann Intern Med 1969; 70:773-776.
80. Ng B, Alvear M. Dextropropoxyphene addiction - a drug of primary abuse. Am J Drug Alcohol Abuse 1993; 19:153-158.
81. Strode SW. Propoxyphene dependence and withdrawal. American Family Physician 1985; 32:105-108.
82. Olsson P. Den lagliga narkotikan - Läkemedel på villovägar. Rapport från Riksförbundet Narkotikafritt Samhälle, Stockholm, 1997.
83. The schedule of controlled substances in the 21 CFR (Code of federal regulations) 1308.12.
84. American Psychiatric Association (1987). Diagnostic and Statistical Manual of Mental Disorders, Third Edition - Revised. DSM-III-R. Washington D.C.
85. Hoffmann NG, Olofsson O, Salen B, Wickström L. Prevalence of abuse and dependency in chronic pain patients. The Int J of Addiction 1995; 30:919-27.
86. Segest E. Poisoning with dextropropoxyphene in Denmark. Human Toxicol 1987; 6:203-207.
87. Allebeck P, Beskow J. Självmord 1970-88. I Självmord i Sverige - en epidemiologisk översikt, redaktion Beskow J, Allebeck P, Wasserman D, Åsberg M, Stockholm, 1993.
88. Prescott LF. Poisoning with salicylates, paracetamol and other analgesics. Regional Poisoning Treatment Centre, The Royal Infirmary, Edinburgh, 1979; 169-175.
89. The National Corporation of Swedish Pharmacies, Pharmaceutical Section, Stockholm, Sweden, 1997.
90. The Swedish Code of Statutes, The Ministry of Justice, 1999, p. 649.
91. Miller RR, Feingold A, Paxinos J. Propoxyphene hydrochloride - a critical review. JAMA 1970; 213:996-1006.
92. Miller RR. Propoxyphene: a review. Am J Hosp Pharm 1977; 34:413-423.
93. Beaver WT. Mild analgesics: a review of their clinical pharmacology (Part II). Am J Med Sci 1966; 251:576-599.
94. Beaver WT. Analgesic Efficacy of dextropropoxyphene and dextropropoxyphene-containing combinations: a review. Human Toxicol 1984; 3:191-220.
95. Li Wan Po A, Zhang WY. Systematic overview of co-proxamol to assess analgesic effects of addition of dextropropoxyphene to paracetamol. Brit Med J 1997; 315:1565-1571.
96. Shira RB, Analgesic activity of ibuprofen (Motrin) in postoperative oral surgical pain. Oral surg 1978; 45:159-166.
97. Larkin RM, van Orden D, Poulson AM, Scott JR. Dysmenorrhea: treatment with an antiprostaglandin. Obstretics & Gynecology 1979; 54:456-460.
98. Morrison JC et al. Analgesic efficacy of ibuprofen for treatment och primary
dysmenorrhea. South med J 1980; 73:999-1002.
99. Spiller HA et al. Prospective multicenter evaluation of tramadol exposure. Clin Toxicol 1997; 35:361-364.
100. Collins SL, Edwards JE, Moore RA, McQuay HJ. Single dose dextropropoxyphene,
alone and with paracetamol (acetaminophen), for postoperative pain. Cochrane Database Syst Rev 2000; 2:CD00144.
101. Lader M. Abuse of weak opioid analgesics. Human Toxicol 1984; 3:229-236.
102. Elston Hooper H, Santo Y. Use of propoxyphene (Darvon) by adolescents admitted to drug abuse programs. Contemporary Drug Problems, 1980; fall:357-368.
103. Rosenberg ML, Davidson LE, Smith JC, Berman AL, Buzbee H, Gantner G, Gay GA, Moore-Lewis B, O´Carroll PW, Jobes D. Operational criteria for the determination of suicide. J Forensic Sci 1988; 33:1445-1456.
due to use or misuse
This report is based on the research concerning the analgesic preparation dextropropoxyphene undertaken by Birgitta Jonasson, Dr Med. Sci., and Ulf Jonasson, graduate student in Public Health in Sweden. Their research, based on forensic medicine data, has led to eight scientific papers, all published in international scientific journals, and one doctoral thesis. A second thesis will be presented in March 2001.
Early in the project they found a considerably high rate of fatal poisonings due to DXP, both suicidal and accidental. They then became aware of the fact that if human lives were to be saved, it would be necessary to inform the society outside the academic world of the results of the research.
The National Board of Forensic Medicine agree about the importance of disseminating their results and has edited this report, which is intended among others, for general practitioners, journalists, politicians and the general public. The Swedish version of the report has attracted much attention and so far has been printed three times.
Since there is reason to believe that the problem of constantly high rates of DXP fatalities is not limited to Sweden, the authors and the National Board of Forensic Medicine share the hope that the translation of this report into English will constitute a useful contribution to the debate in other countries