A Critical Review of the Causes of Death Among

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:

[email protected]

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


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).


3/11


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


4/11



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.


5/11



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


6/11



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.


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



7/11



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.


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


Table 6 Psychotropic drugs (other than benzodiazepines) associ-

ated with detection of products of substitution.


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


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


8/11



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.


9/11



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


10/11



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.


11/11



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.

A Critical Review of the Causes of Death Among

Documents

Transcript of A Critical Review of the Causes of Death Among