Post on 14-Apr-2018
7/29/2019 A Critical Review of the Causes of Death Among
1/11
RESEARCH REPORT
2004 Society for the Study of Addiction doi:10.1111/j.1360-0443.2004.00790.x Addiction, 99, 978988
Blackwell Science, LtdOxford, UKADDAddiction0965-2140 2004 Society for the Study of Addiction
99Review ArticleReviewof buprenorphine and methadone deaths
S. Pirnay
et al.
Correspondence to:
Stephane Pirnay
Laboratoire de Toxicologie de la Prfecture
de Police
2 Place Mazas75012 Paris
France
Tel: + 33 1 44 75 47 20
Fax: + 33 1 44 75 47 25
E-mail:
docteur-stephane.pirnay@laposte.net
Submitted 1 October 2003;
initial review completed 10 November 2003;
final version accepted 15 April 2004
RESEARCH REPORT
A critical review of the causes of death among
post-mortem toxicological investigations: analysis
of 34 buprenorphine-associated and 35
methadone-associated deaths
S. Pirnay1,2, S. W. Borron2,4, C. P. Giudicelli3, J. Tourneau1, F. J. Baud2 & I. Ricordel1
Laboratoire de Toxicologie de la Prfecture de Police, Paris, France1, Hpital Fernand Widal, Paris, France2, Service de Mdecine Interne, Hpital Begin, St Mand,
France3 and Department of Emergency Medicine, George Washington University, Washington, DC, USA4
ABSTRACT
Aims To assess the trends in the number, mortality and the nature of forensic
cases involving toxicological detection of buprenorphine or methadone amongtoxicological investigations performed in Paris from June 1997 to June 2002.
Design Retrospective, 5 year study with review of premortem data, autopsy,
police reports, hospital data, and post-mortem toxicological analyses.
Setting and participants 34 forensic cases of buprenorphine and 35 forensic
cases of methadone detection among 1600 toxicological investigations per-
formed at the Laboratory of Toxicology in the Medical Examiners Office in Paris.
Measurements and results Therapeutic, toxic or lethal drug concentrations
were defined based upon the results of blood analyses and the published litera-
ture. Drug concentrations were cross-referenced with other available ante- and
post-mortem data. Subsequently, we classified a clear responsibility, possible
responsibility or not causative role for buprenorphine or methadone in the
death process, or no explanation of death. Buprenorphine and methadone canbe regarded as being directly implicated in, respectively, four of 34 death cases
(12%) and three of 35 death cases (9%), and their participation in the lethal
process is strongly plausible in eight (buprenorphine) and 11 (methadone) addi-
tional deaths.
Conclusions Analysis of causes of death reveals the difficulties in determining
the role of substitution drugs in the death process, as many other factors may be
involved, including circumstances surrounding death, past history, differential
selection of subjects
into either substitution modality and concomitant intake of other drugs (espe-
cially benzodiazepines and neuroleptics). The potential for synergistic or addi-
tive actions by other isolated moleculesparticularly opioids, benzodiazepines,
other psychotropes and alcoholmust be also considered.
KEYWORDS Benzodiazepines, buprenorphine, cause of death, forensic
sciences, gas chromatography, mass spectrometry, methadone, poisoning.
INTRODUCTION
Methadone was developed in Germany during the Sec-
ond World War at a time when the country was unable
to procure opiates for medical use from the Orient.
Methadone is an effective analgesic and suppresses
withdrawal symptoms from other opiates. This latter
property has led to its usage for the treatment of
7/29/2019 A Critical Review of the Causes of Death Among
2/11
Review of buprenorphine and methadone deaths 979
2004 Society for the Study of Addiction Addiction, 99, 978988
heroin-addicted persons in methadone maintenance
programs since the 1960s. In these programs, orally
administered methadone is provided as a substitute for
injected heroin. Methadone does not provide a high as
such, but ideally prevents withdrawal and reduces the
craving for heroin.
Buprenorphine is a partial agonist to opioid receptors
derived from thebaine. At low dosage (0.30.6 mg intra-
venous or intramuscular), buprenorphine is a strong
analgesic drug, 2540 times more powerful than mor-
phine. Buprenorphine is widely prescribed in a dose of
0.2 mg for the treatment of medium to high painful
symptoms and also for premedication in anaesthesiology.
In 1996, high-dosage buprenorphine was made available
on the French market as a substitution treatment for her-
oin addicts. Three dosages exist: 0.4 mg, 2 mg and 8 mg.
In France, 74 300 heroin addicts were treated daily
with buprenorphine in 2001, while only 9600 people
were treated with methadone. A report in 2000 describedthe success of this heroin substitution program, since
fatal heroin overdoses fell from 500 a few years ago to
100 in 1999. Another recent report suggests that high-
dose buprenorphine, like high-dose methadone and
levomethadyl acetate, substantially reduces the use of
illicit opioids [1].
However, deaths have been reported during substi-
tution with buprenorphine in humans [2,3]. Deaths
may result from either misuse or overdose during
buprenorphine substitution treatment [4,5]. More
recently, a retrospective study was published of fatali-
ties linked to buprenorphine in France from mid-1996to March 2000 [6]. These reports show evidence that
buprenorphine is misused and administered by the
intravenous route [5]. The authors highlighted the
association of substitution products (methadone or
buprenorphine) with psychotropic drugs as a major
factor in fatalities among heroin addicts. Moreover,
buprenorphine is widely associated with benzodiaz-
epine use in heroin addicts, and this combination is
considered as a risk factor for lethal outcomes [2,4,7
10]. Fatal intoxications with methadone have also been
reported [1114].
While buprenorphine and methadone may be dif-ferentially employed according to the patients unique
clinical condition, it is vital for clinicians to under-
stand the circumstances in which each drug may be
implicated in patient deaths, as this may impact on the
choice of substitution product. Thus, the aim of this
retrospective study was to assess the cause of death in
60 consecutive fatalities in which buprenorphine
(n = 34), methadone (n = 35) or both (n = 9) were
analytically detected from June 1997 to June 2002 at
the Laboratory of Toxicology of the Paris Police
Department.
MATERIALS AND METHODS
Subjects
The population served by the coroners office includes all
of Paris and part of the adjacent suburbs. However, not all
fatal opioid-related deaths are likely to be referred to the
medical examiner. Only a subset of cases is referred to thecoroners office, this decision depending on the officials of
the judiciary. The decision to recommend a post-mortem
toxicological analysis can be made by a police officer, the
examining physician, the prosecutor or the instructing
judge. Obviously, there is a possibility at any stage of the
investigation that the need for toxicological investigation
could be overlooked.
All toxicological analyses of death cases performed at
the Laboratory of Toxicology of the Paris Police Depart-
ment from June 1997 to June 2002 that demonstrated
the presence of buprenorphine or methadone, in whole
blood and/or urine, were reviewed. Each toxicological
case was requested by police services or magistrates,
depending on the chronology of the case, in relation to a
suspicious death, as defined by Article 74 of the French
Penal Procedure Code.
All data reported by the forensic specialist, as well as
police reports, were examined in our retrospective study.
The following parameters were collected: age, sex, ante-
mortem associated pathologies and circumstances of
death.
Toxicological analyses
The research for and quantification of drug substances
were carried out systematically in blood and urine. In the
case of associated substances (other than buprenorphine
or methadone), we also performed qualitative toxicologi-
cal analyses in internal organs, hair and bile. A complete
screening of post-mortem biological samples was
performed in all subjects for carbon monoxide (using
infrared spectrophotometry), for benzodiazepines by
immunoassay (EMIT DAU of Microgenics Laboratories,
Passau, Germany) on a COBAS MIRA analyser (Roche,
Saint-Fons, France), for usual organic solvents, includingethanol (using head space gas chromatography-mass
spectrometry (GC-MS)), and for pharmaceuticals and
drugs of abuse (using liquid chromatography-diode array
detector (LC-DAD) and GC-MS) [15]. The methods
employed were gas chromatography coupled with mass
spectrometry (HP 6890/5973 using Agilent (Massy,
France) technology for simple quadruple mass spectrom-
eter and Varian 2000 (Les Ulis, France) for ion trap), as
well as high-performance liquid chromatography cou-
pled with molecular absorption spectrometry with diode
array detection (Alliance, Milford, MA, USA).
7/29/2019 A Critical Review of the Causes of Death Among
3/11
2004 Society for the Study of Addiction Addiction, 99, 978988
980 S. Pirnay et al.
For quantitation, all analytical conditions were
adapted and deuterated standards were added according
to each identified drug.
Definitions of drug concentration levels
Generally, when interpreting a blood concentration from
a post-mortem case, the toxicologist can find helpful
information in databases presenting therapeutic, toxic
and lethal concentrations. However, it should be noted
that generally the ranges of both therapeutic and toxic
concentrations have been determined empirically. Fur-
thermore, it is noteworthy that the therapeutic, toxic and
even lethal concentrations as defined in the literature do
not take into account the role of tolerance. Therapeutic,
toxic or lethal drug concentrations were defined based
upon blood analyses. We defined high concentration for
ranges, where they exist, as those between therapeutic
and toxic blood concentrations.Unfortunately, the data available for buprenorphine
therapeutic and toxic ranges is limited. We used the pub-
lished therapeutic concentrations of buprenorphine and
methadone provided by Repetto et al. [16] and the toxic
concentrations of buprenorphine and methadone pub-
lished by Kintz et al. [17]. For buprenorphine, therapeutic
blood concentrations were defined as less than 5 ng/mL,
and toxic concentrations were defined as greater than or
equal to 5 ng/mL. The limit of detection of buprenorphine
by GC-MS is 0.5 ng/mL and the limit of quantitation by
GC-MS is 1 ng/mL.
For methadone, therapeutic concentrations weredefined as lower than 500 ng/mL, high concentration
levels were 5001000 ng/mL, toxic concentrations were
defined as 10002000 ng/mL and lethal concentrations
were greater than 2000 ng/mL.
Similar methodology was applied for determining
concentration ranges for all other associated drugs.
Determination of the cause of death
The cause of death was determined by either I.R. or J.T.,
both experts at the Laboratory of Toxicology of the Paris
Police Department, from June 1997 to June 2002 in anon-blinded manner. Thereafter, S.P. reviewed all toxico-
logical analyses, as well as data reported by forensic spe-
cialists, police reports and, when it existed, ante-mortem
clinic background. Thus, using the various sources of
data, we a posteriori defined clear responsibility of
buprenorphine or methadone in the death process when
the case was free of other pathology or violent death, and
when buprenorphine and/or methadone was found at
toxic or lethal levels in blood, with associated drugs at
therapeutic levels or less. A possible responsibility of
buprenorphine or methadone in the death process was
defined when buprenorphine and/or methadone was
found at toxic levels but other associated factors existed,
such as previous illness or associated drugs at toxic or
lethal levels. Buprenorphine or methadone were defined
as not causative in the death process when they were
found at therapeutic levels or lower and associated fac-
tors existed, such as previous illness, evidence of violent
death or the presence of associated drugs at toxic or lethal
levels. No explanation of death was defined when the
case was free of pathology or violent death and when
buprenorphine or methadone, as well as associated
drugs, were found at therapeutic levels or less.
Sales figures for high doses of buprenorphine and
methadone in France during the same period
(19962001)
These data were drawn from the annual report of the
Observatoire Franais des Drogues et des Toxicomanies
(OFDT) [18]. We compared the annual number of fatali-
ties associated with buprenorphine and methadone to
the annual sales figures.
RESULTS
From June 1997 to June 2002, at the Laboratory of Tox-
icology of the Paris Police Department, 34 post-mortem
examinations were positive for buprenorphine and 35
others were positive for methadone in biological samples.
In nine of them, buprenorphine and methadone were
detected simultaneously. Over the 5 year period, therewere progressive increases in buprenorphine- and meth-
adone-associated deaths that parallel the increase in sales
figures of both drugs (Fig. 1 a,b).
Buprenorphine was detected in 24 men and seven
women, aged from 20 to 48 years, the median age being
33 years. The gender was unknown in three cases.
Buprenorphine administration by the intravenous route
was known or suspected in seven cases. Methadone was
isolated in 26 men and eight women, aged from 23 to
47 years; the median age was likewise 33 years. The gen-
der was unknown in one case.
Buprenorphine was quantitated in blood in a range ofconcentrations between 1 and 46.1 ng/mL, including 12
cases at toxic concentrations (more than 5 ng/mL) and in
six cases within the therapeutic range (below 5 ng/mL).
It was detected only in urine in 16 cases. Methadone was
quantitated in the blood in a range of concentrations
between 70 and 1960 ng/mL, including five cases within
the toxic range (more than 1000 ng/mL) and 17 cases
within the therapeutic range (below 500 ng/mL). It was
detected only in urine in two cases.
Nine cases are listed in Table 1, where buprenorphine
and methadone were both quantitated in biological
7/29/2019 A Critical Review of the Causes of Death Among
4/11
Review of buprenorphine and methadone deaths 981
2004 Society for the Study of Addiction Addiction, 99, 978988
Figure 1 (a) Number of forensic cases(Paris) related to buprenorphine versus
number of treatments (France), 19962002.
(b) Number of forensic cases (Paris)
related to methadone versus number of
treatments (France), 19962002
0,25
2
3,4
4,5
5,4
77,4
0
1
5
7
6
12
0
4
8
12
Jan. 96 June 96 Jan. 97 June 97 June 98 June 99 June 00 Mar. 01
Numb. of treatments x10000
Numb. of buprenorphine cases
2,5
3,8
5,15,55
6,457
9,5 9,6
0
1
7
9
8
0
4
8
12
Jan. 96 June 96 Jan. 97 June 97 June 98 June 99 June 00 Mar. 01
Numb. of treatments x1000
Numb. of methadone cases
(a)
(b)
Table 1 Nine cases of simultaneous detection of buprenorphine and methadone.
Common cases of
buprenorphine
andmethadone
Buprenorphine
in blood
(ng/mL)
Buprenorphine
in urine
(ng/mL)
Methadone
in blood
(ng/mL)
Methadone
in urine
(ng/mL)
1 < LQ* 12 251 12702 < LQ < LQ 460 < LQ
3 < LQ 11.8 360 14620
4 12.2** 33.8 600 9200
5 < LQ 5.6 351.5 3900
6 < LQ 30.5 804 16.5
7 < LQ 19.7 208 265
8 < LQ 15.1 526 1159
9 < LQ 15 1035** 7000
*LQ (limit of quantification) < 1 ng/mL.
**Toxic concentrations.
7/29/2019 A Critical Review of the Causes of Death Among
5/11
2004 Society for the Study of Addiction Addiction, 99, 978988
982 S. Pirnay et al.
samples. This table shows that there was only one case of
buprenorphine with a toxic concentration and a concom-
itant non-toxic concentration of methadone. Another
case showed a toxic concentration of methadone with a
non-toxic concentration of buprenorphine. The other
seven cases shared non-toxic concentrations of
buprenorphine and methadone.
Except for one case of methadone detection, buprenor-
phine and methadone were uniformly detected in associ-
ation with other drugs. The median number of drugs
associated with buprenorphine was 4.5 (range 110),
while the median number of drugs associated with
methadone detection was 5.0 (range 010). The number
of associated drugs was dispersed randomly through the
cases from 1997 to 2002. There did not seem to be any
modal effect over 5 years (Figs 2 and 3, Table 2).
Among buprenorphine cases, one subject had one
associated drug, four subjects had two associated drugs,
eight subjects had three associated drugs, four subjects
had four associated drugs, five subjects had five
Figure 2 Number of associated drugs in addition to buprenorphine detected in blood of 33 facilities (data are presented chronologically)
5
4
3
6 6
3 3
2
1
3 3
8
6
3
2
6
4
5 5
4
8
9
6
10
3
6 6
2
5 5
3
2
4
7
0
2
4
6
8
10
12
1 4 8 10 1 1 1 2 1 3 1 4 15 1 6 1 7 1 8 1 9 2 0 21 2 2 2 3 2 4 2 5 2 6 2 7 28 2 9 3 0 3 1 3 2 3 3 3 42 3 5 6 7 9
Figure 3 Number of associated drugs in addition to methadone detected in blood of 33 facilities (data are presented chronologically)
5
3
2
4
5
3
6
5
2
5
2
0
4
6
1
8
7
9
5
10
6
3
4
6
3
6
0
2
4
6
8
10
12
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
5 5 5
7 7
6 6
4 4
7/29/2019 A Critical Review of the Causes of Death Among
6/11
Review of buprenorphine and methadone deaths 983
2004 Society for the Study of Addiction Addiction, 99, 978988
associated drugs, seven subjects had six associated drugs,
one subject had seven associated drugs, two subjects had
eight associated drugs, one subject had nine associated
drugs and one subject had 10 associated drugs.
Among methadone cases, one subject had no associ-
ated drugs, one subject had one associated drug, four
subjects had two associated drugs, four subjects had
three associated drugs, five subjects had four associated
drugs, eight subjects had five associated drugs, six sub-jects had six associated drugs, three subjects had seven
associated drugs, one subject had eight associated drugs,
one subject had nine associated drugs and one subject
had 10 associated drugs (Table 3).
For opiates and opioids, among buprenorphine cases,
18 subjects had one associated opioid, three subjects had
two associated opioids and two subjects had three associ-
ated opioids. Among methadone cases, 12 subjects had
one associated opioid, six subjects had two associated opi-
oids and four subjects had three associated opioids.
(Table 4).
Concerning benzodiazepine use among buprenor-
phine cases, 16 subjects had one benzodiazepine, foursubjects had two benzodiazepines, one subject had three
benzodiazepines and one subject had four benzodiaz-
epines. Among methadone cases, 18 subjects had one
associated benzodiazepine, seven subjects had two asso-
ciated benzodiazepines and one subject had three associ-
ated benzodiazepines.
The most frequently associated benzodiazepines were
nordiazepam (buprenorphine: 13; methadone: nine),
bromazepam (buprenorphine: six; methadone: eight)
and diazepam (buprenorphine: four; methadone: 10).
Other associated benzodiazepines (less than three times)
were flunitrazepam, oxazepam and temazepam.Lormetazepam was found in association with buprenor-
phine only once and midazolam was found with metha-
done only once. Concerning flunitrazepam, it is possible
that it was underestimated because of its fragility in
putrid medium (Table 5).
Very few subjects had co-ingested psychotropic drugs
other than benzodiazepines with buprenorphine or
methadone (less than four), except for cyamemazine (a
neuroleptic similar in action to chlorpromazine), which
10 subjects had taken with buprenorphine and eight
with methadone (Table 6).
Table 2 Number of positive drug detections in blood and/or urine in association with buprenorphine or methadone.
Year
Methadone Buprenorphine
Median number ofother associated
drugs (exceptmethadone)
Number ofmethadone
cases during the year
Median number ofother associated
drugs (exceptbuprenorphine)
Number ofbuprenorphine
cases duringthe year
1997 5 1 5 1
1998 4 7 4 51999 5 7 3 7
2000 5 9 4.5 6
2001 6.5 8 5 13
2002 6 3 5.5 2
Table 3 Other substances associated with the finding of buprenor-
phine or methadone in the blood and/or urine.
Buprenorphine Methadone
Opioids/opiates 30 (12*) 36 (13)
Benzodiazepines 31 (2) 35 (4)Cocaine 11 (3) 7
Cannabis 14 18
Phenothiazines and antipsychotics 16 (8) 14 (5)
Antidepressants 8 (1) 11 (2)
Other anxiolytics 9 (2) 9
Ethanol 16 19
Amphetamines 3 0
Paracetamol (acetaminophen) 14 (1) 9
Carbon monoxide 0 1 (1)
Others** 2 6
Total 154 165
*Number of cases where the drug was at toxic or lethal concentration.
**Others include: metoclopramide, hydroxyzine, buclizine, cetirizine,
fluconazole, zidovudine and ephedrine.
Table 4 Other opioids associated with detection products of
substitution.
Buprenorphine Methadone
Heroin 5 (5*) 6 (6)
Morphine 2 (2) 4 (1)Codeine 6 (2) 7 (3)
Dihydrocodeine 0 2
Pethidine (meperidine) 1 (1) 0
Pholcodine 2 3
Propoxyphene 5 (1) 5
Buprenorphine 0 9 (3)
Methadone 9 (1) 0 (0)
Total 30 36
*Number of cases where the drug was at toxic or lethal concentration.
7/29/2019 A Critical Review of the Causes of Death Among
7/11
2004 Society for the Study of Addiction Addiction, 99, 978988
984 S. Pirnay et al.
In half the cases, ethanol was associated (similarly in
both groups) with concentrations exceeding 1 g/L: in five
cases in the buprenorphine group and in eight cases in
the methadone group (Table 7). In one case, the concen-
tration reached 3.54 g/L.
A cross-analysis of toxicological investigations data
permits appreciation of the responsibility of the products
of substitution.
Among people having consumed buprenorphine, two
groups were distinguished, according to their blood
concentrations:
For 12 cases, the concentration was toxic. Among
them, seven cases showed other drugs at lethal or toxic
levels, capable in and of themselves of causing death. In
the five other cases, buprenorphine may be considered
as directly responsible for the death; however, four of
the five had alcohol, cannabis and/or from one to eight
other associated substances at non-toxic concentra-
tions, which may have also played a role in the lethalprocess.
Among 22 other cases for which buprenorphine was
positive at non-toxic levels, 18 deaths can be explained
by other associated drugs or violent causes. The
remaining four cases cannot be explained by other
drugs at toxic concentrations. However, it bears men-
tion that there was concomitant absorption of three to
five other drugs in these cases and absorption of etha-
nol among two (0.4 g/L and 1.3 g/L).
In summary, buprenorphine can be regarded as being
implied directly in four of 34 death cases(12%) and its
participation in the lethal process is plausible in eightothers, as is the case for other associated toxicants. In 18
cases, it appears not to have been responsible and in four
cases, no explanation was evident (Table 8).
For five cases, methadone was quantitated at high
toxic concentrations, but was considered responsible for
death in only three cases, since the two other deaths
involved violent circumstances (homicide by firearm,
burns and carbon monoxide intoxication). For 11 other
cases, methadone was associated with another substance
at toxic concentration, or with another plausible cause of
death.
Table 5 Benzodiazepines found in association with the detection of
buprenorphine and methadone.
Buprenorphine Methadone
Diazepam 4 10 (1*)
Bromazepam 6 (2) 8 (1)
Nordiazepam 13 9 (2)
Flunitrazepam 3 3
Oxazepam 2 2
Clobazam 1 0
Temazepam 1 2
Lormetazepam 1 0
Midazolam 0 1
Total 31 35
*Number of cases where the drug was at toxic or lethal concentration.
Table 6 Psychotropic drugs (other than benzodiazepines) associ-
ated with detection of products of substitution.
Buprenorphine Methadone
Antidepressants
Citalopram
1 2
Clomipramine 3 (1*) 1 (1)
Venlafaxine 1 2 (1)
Amisulpride 1 1
Mianserine 0 2
Paroxetine 0 1
Dosulepine 1 0
Maprotiline 0 1
Carbamazepine 1 1
Sedative and anxiolytics
Meprobamate
3 (1) 3
Hydroxyzine 0 1
Phenobarbital 4 (1) 4
Zopiclone 2 2
Phenothiazines and antipsychotics
Cyamemazine
10 (6) 8 (4)
Alimemazine 1 3 (1)
Phenothiazine 1 1
Levopromazine 2 (1) 0
Acepromazine 2 (1) 1
Pipamperone 0 1
*Number of cases where the drug was at toxic or lethal concentration.
Table 7 Blood concentrations of ethanol detected in associated
with products of substitution.
Ethanol
Buprenorphine
(n = 16)
Methadone
(n = 19)
0.060.5 g/L 6 10
0.51.0 g/L 5 1
1.01.5 g/L 2 5
1.52.0 g/L 2 2
2.03.5 g/L 1 1
Table 8 Responsibility of buprenorphine and methadone as a
cause of death.
33 cases of
buprenorphine
33 cases of
methadone
Clear responsibility 4 3
Possible responsibility 8 11
Not causative 22 21
No explanation for death 4 8
7/29/2019 A Critical Review of the Causes of Death Among
8/11
Review of buprenorphine and methadone deaths 985
2004 Society for the Study of Addiction Addiction, 99, 978988
Among 21 other cases for which methadone was pos-
itive at non-toxic levels, 13 deaths can be explained by
other associated drugs or violent causes. For the remain-
ing eight deaths, concentrations of methadone were low
and none of the other identified substances reached a
toxic concentration.
In summary, methadone can be regarded as implied
directly in three of 35 deaths (9%), and its participation in
the lethal process was plausible in 11, as was the case for
other toxicants associated with its use. In 21 cases, it
appears not to have been responsible and in eight cases,
no explanation was evident.
DISCUSSION
Surprisingly, in this retrospective study we found approx-
imately as many deaths with positive detection of meth-
adone as those with positive detection of buprenorphineover the same period of time in spite of a far greater num-
ber of prescriptions for buprenorphine. In the same vein,
a recent retrospective study on severe opioid poisonings
requiring intensive care admission from 1995 to 1999 in
north-east Paris showed that buprenorphine was
detected in 19 cases while methadone was detected in 12
patients [19]. In France, approximately 80 000 heroin
addicts are treated daily by high-dose buprenorphine
while 10 000 heroin addicts are treated by methadone.
According to the sales of methadone and buprenorphine
during this period of time, one might have anticipated a
higher number of deaths associated with buprenorphinethan with methadone if toxicity was similar. Our data do
not allow us to form any firm conclusions regarding the
potential toxicity of each substance, due to the large
number of variables in terms of chronicity of treatment,
associated drug use and other potential sources of bias.
Indeed, different prescribing practices for each drug and
differential patient selection criteria for opioid substitu-
tion may in part explain our findings. While methadone
is delivered in specialized health centres every week
with urine drug testingbuprenorphine may be pre-
scribed in France by any general practitioner for a 28 day
period and delivered by any pharmacy without anyrequirement for drug testing. From the beginning, meth-
adone appears to have had the image in France of a drug
of last resort for the most desperate cases (this is not the
case in the majority of countries, which have employed
only methadone substitution for many years). On the
contrary, buprenorphine is perceived among users and
care givers as a first-line treatment, more simple, more
flexible and thus a priori destined for less serious cases.
The differences in prescribing rules and the disparity with
regard to access to the two products clearly have a strong
influence on their reputations. Another reason for
prescribing one drug versus the other may have to do
with the perception that methadone, due to its pure m
agonist properties, has greater anxiolytic effects than
buprenorphine. This would encourage use of the former
in subjects presenting with psychopathology complicated
by a greater degree of anxiety. These elements seem to be
corroborated by a study that attempted to identify predic-
tive factors for response to treatment by high-dose
buprenorphine and that showed that social insertion and
previous attempts at abstinence were favourable factors
while, inversely, psychiatric disturbances that had not
been treated previously were unfavourable factors [20].
However, authors are not unanimous in this regard. Cer-
tain consider that buprenorphine is a first-line substitu-
tion treatment, reserving methadone as a utilizable
treatment in selected cases [21]. A similar point of view
has been expressed by a Swiss group, who concluded that
it is not appropriate to treat all patients with a pure ago-
nist like methadone and that initial induction of treat-ment with buprenorphine is often preferable with the
possibility to resort to methadone treatment [22]. It
might thus be argued that the indications for buprenor-
phine and methadone are sufficiently different that com-
parison of mortality in their use serves little purpose. On
the contrary, we find the comparison to be critical. In
locations where both substitution products are available,
this data may be used to determine more appropriate pre-
scribing conditions for each drug. Conversely, in locales
where only one of the drugs is available, this comparison
should encourage a review of the existing formulary.
Tolerance to opioids and/or benzodiazepines would beexpected to play a significant role in both their individual
and combined toxicities. Tolerance has been reported
with methadone as well as buprenorphine. In vivo behav-
ioural studies showed that higher efficacy m agonists
appear more resistant to tolerance than do lower efficacy
agonists. Treatment with escalating doses of morphine or
buprenorphine produced greater tolerance to the lower
efficacy buprenorphine than to the higher efficacy ago-
nist, morphine [23]. Furthermore, repeated morphine
treatment results in less cross-tolerance to higher efficacy
agonist methadone than to morphine [24]. Unfortu-
nately, in our forensic study, because the ante-mortemhistories of drug use of the subjects were often unobtain-
able, we cannot comment on the role that tolerance, or
lack thereof, may have played in these deaths.
On the contrary, comparison of the incidents caused
by the abuse of each tends to suggest that the risk
of deaths would be higher if methadone alone were
available [25].
Tracqui described 20 deaths attributed to intoxica-
tion by buprenorphine, despite blood concentrations
close to the therapeutic range [3]. However, in five cases
out of 20, there were additional associated opioids.
7/29/2019 A Critical Review of the Causes of Death Among
9/11
2004 Society for the Study of Addiction Addiction, 99, 978988
986 S. Pirnay et al.
Only the study of Kintz [6] collected forensic cases,
and is thus directly comparable with ours. In this series of
117 deaths, only one case involved buprenorphine taken
alone, quantitated at 0.8 ng/mL, and death was listed as
being caused by aspiration pneumonia. Furthermore, the
author did not consider the lethal concentrations of var-
ious substances associated with buprenorphine.
Kintz recognized, like others, the determining role of
benzodiazepines in the death process when they were
employed with buprenorphine [19] or with methadone
[14]. It is known both experimentally [26] and clinically
[27] in anesthesiology that the respiratory depression
induced by the opioids can be worsened by the addition of
benzodiazepines.
Our discussion is based on analytical findings per-
formed on very selective and sensitive analytical equip-
ment. However, we did not quantitate norbuprenorphine,
the metabolite of buprenorphine, which was demon-
strated by Ohtani and colleagues as a strong toxicantplaying a predominant role in the respiratory depression
associated with buprenorphine [28]. For methadone, our
dosages did not distinguish D and L methadone since we
did not use a chiral column on our chromatographs.
With regard to our blood and urine post-mortem sam-
ples, we noted that buprenorphine was detected only in
the urine in 16 cases, and methadone was detected only
in the urine in two cases. This could artificially increase
the total number of buprenorphine overdoses. Thus, it
would be logical to remove those buprenorphine and
methadone cases detected in urine only. Even though the
principal buprenorphine metabolite is active, and meth-adone metabolites are not active, if we remove those
buprenorphine and methadone urine-only positive cases
then the global number of deaths associated with meth-
adone rises.
Our classification of clear responsibility, possible
responsibility or not causative roles of buprenorphine
and methadone in the death process, or no explanation
of death, refers to the past history of the subjects, which
is often not well known. Moreover, those data are only
provided by officials in a retrospective context.
In order to appreciate the action of the various sub-
stances analysed, it is appropriate to consider the diffi-culty inherent in interpreting the concentrations found
in the corpse using toxic concentrations measured in the
living. Specific references for these situations are gener-
ally unavailable. For the living subject, the analyst uses
serum or plasma for which effective therapeutic, toxic
and lethal concentrations have been established. For the
corpse, the use of lysed whole blood may explain the ori-
gin of dispersions from published data. One may add to
these difficulties the variability of biological environ-
ments after death [22,29,30]: development of microbial
flora able to produce or to consume alcohol; quick
degradation of flunitrazepam (less than 24 hours) [22];
re-circulation of blood within the vessels under the action
of gases of putrefaction [29,30]; exchanges between
intra- and extracellular biological compartments; exit of
drugs from tissues; contamination of blood from the
digestive tract; and external contamination. In summary,
it is almost always difficult to determine the precise role of
a drug in the death process.
In our studied series, buprenorphine and methadone
were present at more or less elevated concentrations,
sometimes clearly toxic, when compared with analytical
data usually observed in the living subject. However, the
imputability of the death to these two molecules can be
established only rarely (less than 12% of the cases for
buprenorphine and less than 9% for methadone) and
must be really excluded in 64% of the cases of buprenor-
phine and 60% of the cases of methadone. For the
remaining 24% of buprenorphine-associated and 31% of
methadone-associated cases, the participation of thesedrugs in the process that caused death is plausible at best.
In four of the 34 deaths involving buprenorphine, no
explanation was evident. We note that in three cases
there were co-ingestions of buprenorphine and metha-
done and two to four other drugs [temazepam, cannabis,
propranolol, cyamemazine, paracetamol, alcohol (0.3 g/
L) and cocaine] at therapeutic blood concentrations. The
fourth case had cocaine at therapeutic blood concentra-
tions and alcohol at 1 g/L.
In eight of the 35 cases involving methadone, no
cause of death was evident. Three of these cases involving
both buprenorphine and methadone have just beendescribed. Four other cases implicated methadone along
with two to six associated drugs at therapeutic blood con-
centrations (bromazepam, codeine, morphine, cannabis,
paracetamol, cocaine, diazepam, dihydrocodeine, cetiriz-
ine). The final case among them associated methadone
and bromazepam at therapeutic concentrations, along
with ethanol (0.27 g/L).
CONCLUSIONS
This paper reports a review of forensic cases involving 34cases of buprenorphine detection and 35 cases of metha-
done detection among 1600 toxicological investigations
performed at the Laboratory of Toxicology of the Paris
Police Department within the period of June 1997 to June
2002.
Buprenorphine was detected in post-mortem whole
blood by GC-MS in a range from 0.5 ng/mL to 46.1 ng/
mL, while methadone was measured at 20 ng/mL to
1960 ng/mL.
The study of the role of each drug, based on total blood
concentrations of these and associated substances and
7/29/2019 A Critical Review of the Causes of Death Among
10/11
Review of buprenorphine and methadone deaths 987
2004 Society for the Study of Addiction Addiction, 99, 978988
the reported death circumstances, leads us to assess the
clear responsibility of buprenorphine only in four cases
(12%), and of methadone only in three cases (9%). The
potential for synergistic or additive actions by other iso-
lated molecules, particularly opioids, benzodiazepines,
other psychotropic drugs and alcohol, must also be con-
sidered. Finally, even though both buprenorphine and
methadone may be used as heroin substitution products,
their precise roles may differ according to clinical circum-
stances. This should be taken into account in both the
choice of prescription of these drugs and in any compar-
ison of associated mortality.
Acknowledgements
This study was supported by a grant from MILDT
(Mission Interministerielle de Lutte contre les Drogues et
Toxicomanies) and from Schering-Plough Company,
France.
Conflict of interest statement
This work was supported by Schering-Plough, the distrib-
utor of buprenorphine in France.
References
1. Johnson, R. E., Chutuape, M. A., Strain, E. C., Walsh, S. L.,
Stitzer, M. L. & Bigelow, G. E. (2000) A comparison of
levomethadyl acetate, buprenorphine and methadone foropioid dependence. New England Journal of Medicine, 343,
12901297.
2. Reynaud, M., Petit, G., Potard, D. & Courty, P. (1998) Six
deaths linked to concomitant use of buprenorphine and
benzodiazepines. Addiction, 93, 13851392.
3. Tracqui, A., Tournoud, C., Flesch, F., Kopferschmitt, J.,
Kintz, P., Deveaux, M., Ghysel, M. H., Marquet, P., Pepin, G.,
Petit, G., Jaeger, A. & Ludes, B. (1998) Intoxications aigus
par traitement substitutif base de buprnorphine haut
dosage. 29 observations cliniques20 cas mortels. [Acute
poisoning during substitution therapy based on high-dos-
age buprenorphine. 29 clinical cases20 fatal cases].
Presse Medicale, 27, 557561.
4. Tracqui, A., Kintz, P. & Ludes, B. (1998) Buprenorphine-related deaths among drug addicts in France: a report on 20
fatalities.Journal of Analytical Toxicology, 22, 430434.
5. Robinson, G. M., Dukes, P. D., Robinson, B. J., Cooke, R. R. &
Mahoney, G. N. (1993) The misuse of buprenorphine and a
buprenorphine-naloxone combination in Wellington, New
Zealand. Drug and Alcohol Dependence, 33, 8186.
6. Kintz, P. (2001) Deaths involving buprenorphine: a com-
pendium of French cases. Forensic Science International, 121,
6569.
7. Gaulier, J. M., Marquet, P., Lacassie, E., Dupuy, J. L. &
Lachatre, G. (2000) Fatal intoxication following self-admin-
istration of a massive dose of buprenorphine. Journal of
Forensic Sciences, 45, 226228.
8. Drummer, O. H., Syrjanen, M. L. & Cordner, S. M. (1993)
Deaths involving the benzodiazepine flunitrazepam. Ameri-
can Journal of Forensic Medicine and Pathology, 14, 238243.
9. Hammersley, R., Cassidy, M. T. & Oliver, J. (1995) Drugs
associated with drug-related deaths in Edinburgh and Glas-
gow, November 1990 to October 1992. Addiction, 90, 959
965.
10. Reynaud, M., Tracqui, A., Petit, G., Potard, D. & Courty, P.
(1998) Six deaths linked to misuse of buprenorphine-
benzodiazepine combinations. American Journal of Psychia-
try, 155, 448449.
11. Milroy, C. M. & Forrest, A. R. (2000) Methadone deaths: a
toxicological analysis.Journal of Clinical Pathology, 53, 277
281.
12. Vormfelde, S. V. & Poser, W. (2001) Death attributed to
methadone. Pharmacopsychiatry, 34, 217222.
13. Caplehorn, J. R. & Drummer, O. H. (2002) Fatal methadone
toxicity. signs and circumstances, and the role of benzodiaz-
epines. Australian and New Zealand Journal of Public Health,
26, 358362.
14. Ernst, E., Bartu, A., Popescu, A., Ileutt, K. F., Hansson, R. &
Plumley, N. (2002) Methadone-related deaths in Western
Australia 199399. Australian and New Zealand Journal ofPublic Health, 26, 364370.
15. Pirnay, S., Ricordel, I., Libong, D. & Bouchonnet, S. (2002)
Sensitive method for the detection of 22 benzodiazepines by
gas chromatography-ion trap tandem mass spectrometry.
Journal of Chromatography A, 954, 235245.
16. Repetto, M. R. & Repetto, M. (1997) Habitual, toxic and
lethal concentrations of 103 drugs of abuse in humans.
Journal of Toxicology-Clinical Toxicology, 35, 19.
17. Kintz, P. (1998) Toxicologie et Pharmacologie Mdico-Lgales
[Forensic Toxicology and Pharmacology]. Collection Option
Bio. Paris: Elsevier.
18. Costes, J. M. & Bello, P. Y. (2003) Substitution aux Opiacs en
France, Synthse des Informations Disponibles de 1996 2001
[Substitution to Opiates in France, Summary of Available Databetween 1996 and 2001]. Paris: Observatoire Francais des
Drogues et des Toxicomanies.
19. Gueye, P. N., Megarbane, B., Borron, S. W., Adnet, F.,
Galliot-Guilley, M., Ricordel, I., Tourneau, J., Goldgran-
Toledano, D. & Baud, F. J. (2002) Trends in opiate and opi-
oid poisonings in addicts in north-east Paris and suburbs,
199599. Addiction, 97, 12951304.
20. Gasquet, I., Lancon, C. & Parquet, P. (1999) Predictive fac-
tors for patient maintenance on buprenorphine high dosage
treatment: a naturalistic study in primary care. Encephale,
25, 645651.
21. Reisinger, M. (1994) Comparative value of buprenor-
phine and methadone. Annales de Medecine Interne, 145,
2325.22. Uehlinger, C., Deglon, J., Livoti, S., Petitjean, S., Waldvogel,
D. & Ladewig, D. (1998) Comparison of buprenorphine
and methadone in the treatment of opioid dependence.
Swiss multicentre study. European Addiction Research, 4,
1318.
23. Walker, E. A. & Young, A. M. (2001) Differential tolerance
to antinociceptive effects of mu opioids during repeated
treatment with etonitazene, morphine or buprenorphine in
rats. Psychopharmacology, 154, 131142.
24. Craft, R. M. & Dykstra, L. A. (1990) Differential cross-
tolerance to opioids in squirrel monkeys responding under a
shock titration schedule.Journal of Pharmacology and Exper-
imental Therapeutics, 252, 945952.
7/29/2019 A Critical Review of the Causes of Death Among
11/11
2004 Society for the Study of Addiction Addiction, 99, 978988
988 S. Pirnay et al.
25. Auriacombe, M., Franques, P. & Tignol, J. (2001) Deaths
attributable to methadone vs buprenorphine in France.
Journal of the American Medical Association , 285, 45.
26. Borron, S. W., Monier, C., Risede, P. & Baud, F. J. (2002)
Flunitrazepam variably alters morphine, buprenorphine
and methadone lethality in the rat. Human and Experimental
Toxicology, 21, 599605.
27. Henry, J. (1999) Fatal opioid toxicitya clinical perspec-
tive. Addiction, 94, 974975.
28. Ohtani, M., Kotaki, H., Sawada, Y. & Iga, T. (1995)
Comparative analysis of buprenorphine- and
norbuprenorphine-induced analgesic effects based on
pharmacokinetic-pharmacodynamic modeling. Journal of
Pharmacology and Experimental Therapeutics, 272, 505510.
29. De Letter, E. A., Clauwaert, K. M., Belpaire, F. M.,
Lambert, W. E., Van Bocxlaer, J. F. & Piette, M. H.
(2002) Post-mortem redistribution of 3,4-
methylenedioxymethamphetamine (MDMA, ecstasy) in
the rabbit. Part I: experimental approach after in vivo intra-
venous infusion. International Journal of Legal Medicine, 116,
216224.
30. De Letter, E. A., Belpaire, F. M., Clauwaert, K. M.,
Lambert, W. E., Van Bocxlaer, J. F. & Piette, M. H.
(2002) Post-mortem redistribution of 3,4-
methylenedioxymethamphetamine (MDMA, ecstasy) in
the rabbit. Part II: post-mortem infusion in trachea or
stomach. International Journal of Legal Medicine, 116,
225232.