Drug Abuse

This report was commissioned by the European Monitoring Centre for Drugs
and Drug Addiction (EMCDDA) as a background paper for the risk assessment
of the compound GHB. The report follows the structure of Annexes A
and B of the risk-assessment guidelines developed by the Scientific
Committee of the EMCDDA.

Pharmacotoxicological evidence

Chemical/pharmaceutical information

Chemical description

Gamma-hydroxybutyric acid (shown below) refers to the protonated form
whereas gamma-hydroxybutyrate refers to the deprotonated form of the carboxylic
acid moiety. The abbreviation GHB refers to both of these chemical
names. Other chemical names include oxybate, 4-hydroxybutanoic acid,
and 4-hydroxybutyric acid. The chemical structure is shown in Figure 2.

The chemical abstracts registration (CAS) number for ‘free’ gamma-hydroxybutyric
acid (GHB) is [591-81-1] and [502-85-2] for GHB sodium salt (NaGHB).

GHB sodium salt is a white solid, soluble in water and methanol.
GHB salts other than sodium can also be formed, for example GHB potassium
salt (KGHB).

PRECURSORS

There are various chemical precursors to GHB. Gamma-butyrolactone (GBL)
and 1,4-butanediol have both been found to undergo in vivo conversion to
GHB in animals and humans. The hydrolysis of GBL to GHB is catalysed in
vivo by a lactonase (Roth et al., 1966). In rat whole blood the half-life conversion
of GBL was only 1 minute, with serum more active than plasma
(Roth et al., 1966). Rat liver was also found to have substantial lactonase
activity, however, human cerebrospinal fluid (CSF) did not. It was found that
muscle tissue can sequester a large part of the initial GBL dose, thereby
delaying conversion to GHB and prolonging the duration of action. GHB formation
from GBL is chemically a reversible reaction and appears to be pH
dependent. Under acidic conditions, GHB can be converted to the lactone,
GBL, a process that has been exploited for gas chromatographic analysis of
the compound (Vree et al., 1976). No endogenous GBL has been detected in
plasma or urine, therefore, it is assumed that this conversion does not occur
in vivo. It has also been reported that 1,4-butanediol is also rapidly
metabolised to GHB in vivo, in a reaction catalysed by the enzyme alcohol
dehydrogenase (ADH) (Maxwell et al., 1971; Roth et al., 1968).

METHODS OF SYNTHESIS

Illicit GHB is reportedly synthesised using various methods. If pharmaceutical-
grade GHB cannot be obtained, users/producers usually exploit the conversion
of GBL to GHB under certain conditions (e.g. alkaline pH > 7).
Notionally this requires the addition of sodium hydroxide (or potassium
hydroxide) with water to GBL. There are various dangers associated with
such a reaction, particularly as the reaction is exothermic and GBL is flammable.
Furthermore, commercially-available domestic or industrial products,
which could be used for synthesis, are not meant for human consumption
and invariably contain other potentially toxic substances, including
heavy metals and other organic solvents such as acetone or toluene. Use of
such products as reagents may result in serious toxic effects if the resultant
impure product is consumed. To aid the producer, ‘GHB kits’ are available
which apparently contain the necessary ‘pure’ ingredients in ‘accurately
weighed’ amounts. Various ‘recipes’ have been presented both on the
Internet and in books (Ward et al., 1998).

IDENTIFICATION

The analytical profile of GHB has been described in numerous papers. In
particular, data pertaining to gas chromatography with mass spectrometry
(GC-MS) and gas chromatography with flame ionisation detection (GC-FID)
are described (Vree et al., 1976; McCusker, et al., 1999; Ferrara et al., 1993;
Couper et al., 2000); analysis usually requires conversion to γ-butyrolactone
(GBL) or chemical derivatisation. A chemical (colour reaction) spot test has
been developed which also requires the conversion of GHB to GBL
(Badcock et al., 1999). GHB is not detectable using traditional field-test kits.

Legitimate uses of GHB

GHB has been used in various pre-clinical and clinical trials since 1960.
GHB was originally evaluated as an anaesthetic particularly in France and
Germany (as Gamma OHTM and SomsanitTM, respectively). It has also been
assessed in the treatment of narcolepsy and associated disorders such as cataplexy,
in addition to its use as an aid to opiate and alcohol withdrawal in
Italy (as AlcoverTM). There are no known reported industrial uses of GHB,
however, GBL and 1,4-butanediol are used as solvents in various industrial
processes.

Pharmaceutical form

GHB is available as either a liquid formulation or as a powder (either loose
or sometimes in a capsule). It has various street names including ‘liquid
ecstasy’, ‘liquid E’, ‘GBH’, ‘easy lay’, ‘scoop’, ‘liquid X’, ‘fantasy’ and ‘cherry
meth’.

Seized GHB material in Europe appears to consist of both powder and liquid
preparations. Seizures of GBL and 1,4-butanediol are predominantly in liquid
form. The following is a list of some common (mostly previously available)
GHB-related products usually sold as ‘nutritional or dietary supplements’

().

‘Blue Nitro’ contains GBL, Vitamin B12 and Potassium
‘RenewTrient’ contains GHB
‘Midnight Blue’ contains GBL
‘SomatoPro’ contains 1,4-butanediol
‘Serenity’ contains 1,4-butanediol
‘Enliven’ contains 1,4-butanediol

GHB and related products are generally perceived to be cheap to purchase
compared to other illicit drugs, in respect of the cost per effective dose.

Route of administration and dosage

As GHB is invariably obtained in the form of a powder or liquid formulation,
the primary route of administration is oral. However, it does not preclude the
possibility of the powder being ‘snorted’ or ‘smoked’ or the liquid being
injected — although there are no confirmed reports of these routes of administration.
The powder (usually GHB sodium salt) is invariably mixed with
water prior to consumption. Many of the dangers associated with illicit GHB
use are due to variances in the GHB concentrations of such solutions.
Furthermore, the concentration of ‘pre-prepared’ liquid solutions can also
vary considerably. Many web sites and books which advocate GHB use suggest
that an individual ‘finds the dose they are comfortable with’ and ‘takes
GHB on an empty stomach for a more rapid effect’ (Ward et al., 1998). This
is due to the fact that GHB appears to ‘effect different people in different
ways’ — a euphoric dose for one person could be a sedative dose for another
(Kam et al., 1998). The steep dose-response curve of GHB — where a
small increase in the dose can cause sedation as opposed to just nausea —
could also cause problems in terms of the user selecting the required dosage
or taking subsequent doses in quick succession. However, it is generally suggested
that a 0.5 g dose be taken for relaxation and disinhibition, a 1 g dose
for euphoric effect and a 2–3 g dose for deep sleep (Ward et al., 1998;

; ).

Toxicology and pharmacology in animals and humans

Pharmacodynamics and preclinical safety data

NEUROPHARMACOLOGY

GHB was first synthesised in 1960 by Laborit (1964) in an attempt to study
the effects of butyric acid and GABA (γ-aminobutyric acid), producing a
compound which would interfere with β-oxidation and would cross the
blood-brain barrier. Bessman and Fishbein (1963) later discovered that GHB
is an endogenous compound existing as a proposed metabolite of GABA.
During these studies GHB was isolated in the brain of both rats and humans.
Some researchers postulated that GHB was also a putative neurotransmitter
or neuromodulator (Mandel et al., 1987; Cash, 1994).

There have been many studies detailing the effects of GHB on various neurotransmitter
systems, particularly serotonin (5-HT, 5-hydroxytryptamine),
noradrenaline (NA, norepinephrine), dopamine (DA) and acetylcholine
(ACh). Although these studies have produced variable results, the data suggest
that GHB does have a significant effect on the dopaminergic system.
There may also be an accompanied increase in the release of endogenous
opioids, for example, dynorphin (Hechler et al., 1991).

Giarman and Schmidt (1963) noted that at relatively high doses of GHB,
ACh levels were increased in certain regions of the brain. Early work by
Gessa et al. (1966) studied the effect of GHB on 5-HT, NA and DA in the
brains of rabbits and Long-Evans rats. Rabbits were injected intravenously
(i.v.) and rats were injected intraperitoneally (i.p.) with varying doses of GHB
ranging from 250 mg/kg to 2 000 mg/kg and sacrificed 0–4 hours post dose.
The results of the various experiments indicated that there is a slight increase
in 5-HT and NA levels in the brain; however, they observed a pronounced
increase in brain DA levels (primarily in the caudate nucleus). The maximal
increase in DA concentration occurred 1–2 hours after administration of
2 000 mg/kg of GHB with a slow decline thereafter. Further study of the
effects of GHB on DA involved the administration of L-DOPA and a known
monoamine oxidase inhibitor (MAOI), pargyline. It was found that although
DOPA produced an initial higher increase in rat brain DA, GHB produced a
more sustained increase and co-administration of the two compounds
(DOPA 50 mg/kg i.v. and GHB 2 000 mg/kg i.p.) produced a further increase.
Furthermore it also appeared that DOPA-decarboxylase was not affected by
GHB. Administration of pargyline (80 mg/kg i.p.) to rats produced complete
monoamine oxidase (MAO) inhibition, whereas MAO activity was not inhibited
following a 2 000 mg/kg i.p. GHB dose. It was concluded that GHB
does not appear to be a MAOI.

Other studies concerning GHB and brain DA levels confirmed that DA is
altered in response to GHB (Walters et al., 1973; Bustos et al., 1972; Spano
et al., 1971; Cheramy et al., 1977; Godbout et al., 1995). It appears that
there is an initial inhibition of DA release at the synapse but an increase in
neuronal DA production. After this intracellular increase in DA there is either
a time-dependent (DA increases with time) or dose-dependent non-functional
leak of DA from the neurone (low doses inhibit, high doses stimulate).
Both theories ultimately result in a pronounced increase in brain DA concentration.
However, Feigenbaum and Howard (1996) have reported that
GHB inhibits rather than stimulates DA release and that experiments showing
DA stimulation were performed under anaesthesia or in the presence of
high calcium concentrations; such conditions apparently have been found to
spuriously enhance striatal DA release.

GHB was also found to have an affinity for two receptors in the brain — a
possible GHB-specific receptor and GABAB receptor. GHB appeared to have
no affinity for the GABAA receptor. Evidence for a GHB-specific receptor
came from experiments by Benavides et al. (1982) and Maitre et al. (1990)
involving radiolabelled GHB ((3H)GHB), which bound to the receptor even
in the presence of GABA, and binding inhibition studies using a GHB antagonist
NCS-382, which prevented GHB binding. The highest concentrations
of the GHB binding sites in rat brain were in the olfactory bulbs, hippocampus
and cerebral cortex. Further work using rat brain membranes suggest
that the receptor is linked to the Gi or Go family of proteins (Ratomponirina
et al., 1995). Godbout et al. (1991) reported that there is an increase in spontaneous
firing in prefrontal cortical neurones after administration of low
doses of GHB. As this is inhibited by NCS-382, it suggests that GHB binding
to the GHB-specific receptor mediates this response. DA is known to inhibit
prefrontal nerve cells, suggesting that GHB reduces the DA levels, thus
preventing inhibition of prefrontal cortical neuronal firing. GHB inhibits DA
release by binding to the GHB-specific receptor. However, administration of
high doses of GHB produced inhibition of these neurones. It was postulated
that this was due to an increase in DA levels resulting from GHB-induced
stimulation of a second receptor, GABAB (Nissbrandt et al., 1996; Bowey,
1989; Xie et al., 1992; Williams et al., 1995). GHB has been found to be
only a weak agonist of this receptor, exhibiting a binding affinity of 1 000
times less than GABA and 1 000 times less than binding to the GHB-specific
receptor (Mathivet et al., 1997). Studies using a GABAB antagonist, CGP
35348, indicated that GHB activation of the GABAB receptor produces
hyperpolarisation (Williams et al., 1995). A Na+ dependent GHB transport
has also been discovered which is thought to remove GHB from the synaptic
cleft following neuronal release (Benavides et al., 1982).

NEUROENDOCRINOLOGY

Following an intravenous 2.5 g dose of GHB in six male human volunteers,
a significant increase in both plasma prolactin and growth hormone (GH)
was observed at 30, 45, 60 and 90 minutes post dose (Takahara et al., 1977).
Five of the six patients fell asleep. These effects were not observed in the
saline controlled group. As DA is known to inhibit prolactin production, the
results suggested there was a GHB-induced reduction in DA. However, as
growth hormone secretion is known to be increased by dopaminergic stim-
ulants, it was concluded that the growth hormone increase in this case was
not due to GHB-inhibition of DA release. Other work had indicated that 5-
HT and a precursor (5-hydroxytryptophan) stimulated prolactin and growth
hormone secretion in rats and humans (Kato et al., 1974; Smythe et al.,
1975). It was therefore speculated that GHB may induce prolactin and
growth hormone release by modifying the release of 5-HT from the nerve terminals.
Further postulation suggested that GHB acts directly on neurons in
the hypothalamus and stimulates or blocks the release of GH-releasing or
GH-release inhibiting and prolactin-release inhibiting hormones. The slowwave
and REM (rapid eye movement) sleep apparently induced by GHB (see
Effects on brain function) is also thought to be the periods of sleep where GH
production is at its greatest (Chin et al., 1992).

CARDIOVASCULAR AND RESPIRATORY EFFECTS AND THERMOREGULATORY RESPONSES

Laborit (1964) observed a constant but short drop in blood pressure in rabbits
after administration of GHB, but in dogs there was either no effect or a
slight progressive increase in blood pressure (even under controlled ventilation
conditions). In all animals, a constant bradycardia was observed. GHB
also appeared to elevate the sensitivity threshold of the pressure receptors in
the rabbit and dog, without having any obvious action on the chemoreceptors.
Laborit and Leterrier (1964) also observed a strong hepatic and renal
vasodilating action, particularly during haemorrhagic shock in animals, indicating
that GHB has ‘antishock activity’. In humans, after a 2–4 g injection
of GHB there appeared to be no effect on blood pressure, unless during
surgery when, in the absence of adequate neuroplegic premedication, a progressive
hypertensive episode occasionally occurred. In addition, there were
no unfavourable effects observed in 50 human atherosclerotic patients under
GHB anaesthesia. However, a frequent decrease in the amplitude of the Twave
was noted, but this appeared to be due to the hypokalaemia (reduction
in serum potassium levels) associated with GHB (Laborit, 1964). This was
reversed by the administration of potassium. A study in Poland of 100
patients also suggested that administration of GHB resulted in a constant
drop in blood cholesterol levels (Laborit, 1964).

Laborit also observed in both animals and humans that GHB-induced sleep
is not accompanied by a decrease in oxygen consumption. At low hypnotic
doses of GHB, a decrease in ventilatory rate was reported with an increase
in amplitude. At high (sleep-inducing) doses of GHB, a Cheyne-Stokes
rhythm appeared (including periods of apnoea, often observed in coma
patients); however, the respiratory centre remained sensitive to an increase
in carbon dioxide (pCO2). Both Laborit and Gessa (1966) reported a slight
drop in body temperature of animals given GHB. Gessa noted that this
appeared particularly pronounced in rats receiving 2 g/kg GHB kept at 18 oC
compared to those kept at 37 oC (room temperature).

EFFECTS ON BRAIN FUNCTION

Many researchers have recorded the effects of GHB on brain function in
animals and humans using an electroencephalogram (EEG) (Laborit, 1964;
Winters et al., 1967; Marcus et al., 1967; Scotti et al., 1978; Mamelak et al.,
1977; Metcalf et al., 1966; Entholzner et al., 1995). The results have been
contradictory to some extent, with GHB producing various EEG patterns in
various animal and human models. Some animal studies report apparent
epileptiform (epileptic/seizure-like) EEG changes which have not been
observed in human volunteer studies following GHB administration.
Random clonic movements of the face and extremities have been reported
to be associated with GHB-induced anaesthesia without epileptiform EEG
changes. In fact, Jouany et al. observed that GHB apparently controlled
chemical-induced seizures (using ammonium chloride, strychnine, cardiazol
and isoniazide) to some extent (Laborit, 1964).

Based on behavioural and electroencephalographic criteria, GHB-induced
sleep has been described as being indistinguishable from natural sleep, that
is unlike coma. The natural stages of sleep 1–2–3–4–REM (rapid eye movement)
all occur in their normal sequence (Mamelak et al.,1977). GHB has
been noted to lengthen stages 3–4 (delta/slow-wave sleep) followed by REM
sleep. The effect of GHB-enhanced sleep appears to wear off after 3–4 hours
at ‘normal’ doses, with no apparent side effects.

TOXICOLOGY

At present there are no animal or human data concerning reproductive toxicity,
neurotoxicity or the mutagenicity and carcinogenic potential of GHB.

Toxicity in animals
Laborit (1964) found sleep could be induced in the rat with 0.5 g/kg GHB
(i.p.) and in rabbits and dogs using 1 g/kg (i.v.). In rats, the LD50 (i.p.) was
1.7 g/kg and the LD100 was 2 g/kg. The cause of death was reported to be respiratory
depression; however, using artificial respiration rabbits tolerated
doses up to 7 g/kg. With respect to weight, bone marrow, liver and kidneys,
there were no significant differences observed between controls and rats
receiving 0.17 g/kg GHB daily for 70 days.

During the course of the various experiments involving the administration of
GHB to animals at numerous doses, the following observations have been
made regarding the toxicity of GHB in animals. The toxicity of GHB appears
to be dose-dependent and can induce various degrees of sleep, bradycardia,
a decrease in body temperature and possible seizures/spasms. Death has
been reported to be due to respiratory depression in rats.

Toxicity in humans
Early reviews concerning the use of GHB, particularly in anaesthesia, suggested
that GHB was non-toxic (Laborit, 1964). Short amnesia and hypotonia
have been associated with an oral dose of 10 mg/kg GHB (Chine et al.,
1992). REM sleep can be induced in humans using an oral dose of between
20-30 mg/kg GHB (Mamelak et al., 1986; Yamada et al., 1967). 50–70 mg/kg
GHB given intravenously produces hypnosis but has little analgesic effect
(Appleton et al., 1968). This dose may also cause hypotonia, bradycardia,
nausea, vomiting, random clonic movements of the face and extremities and
Cheyne-Stokes respiration (Laborit, 1964; Chin et al., 1992). Following a
typical 65 mg/kg intravenous dose of GHB, sleepiness can occur within
5 minutes, followed by a comatose state lasting for 1–2 hours or more, after
which there is a sudden awakening (Vickers, 1968). High oral doses of GHB
(greater than 60 mg/kg) can also result in coma, usually lasting up to 4 hours
(Mamelak, 1989). Table 1 shows a summary of resultant concentrations following
various GHB doses.

In 1964, Helrich et al. reported that blood GHB concentrations exceeding
260 mg/l were associated with deep sleep, 156–260 mg/l associated with
moderate sleep, 52–156 mg/l associated with light sleep and levels less than
52 mg/l were associated with wakefulness.

Interactions with other drugs or medicines
There have been various published reports of GHB intoxication, however,
the frequent presence of other drugs may have complicated the clinical presentation.
Typical presentation appears to be various degrees of consciousness,
euphoria (‘high’), aggressive behaviour, ataxia, amnesia, somnolence,
bradycardia, confusion, hallucinations, respiratory depression and apnoea,
vomiting and random clonic movements (sometimes reported as being
seizures) (Kam et al., 1998; Chine et al., 1998; Li et al., 1998a; Li et al.,
1998b). Presenting patients have been reported to have initial GCS scores
(Glasgow coma score) of between 3 (severe decrease in consciousness) and
15 (wakeful) (Chin et al., 1998; Williams et al., 1998). There have been no
detailed studies concerning the interaction of GHB with other drugs or medicines.
However, it is believed the adverse effects of GHB intoxication are
exacerbated by the presence of other sedatives or depressants such as opiates
(e.g. heroin or morphine), benzodiazepines, barbiturates or alcohol (e.g.
ethanol) and possibly other psychoactive compounds (e.g. amphetamine).
Depending on the nature of the interaction, resultant effects may also
depend on the order in which the drugs are administered — for example,
there may be potential problems if amphetamine is ingested after GHB due
to the resultant release of neuronal dopamine.

Various possible reversal/antagonizing agents have been tested against the
clinical effects of GHB toxicity. Commonly used coma reversal agents such
as naloxone (opiate/opioid antagonist) and flumazenil (GABA, benzodiazepine
antagonist) had no effect (Mamelak et al., 1986; Yamada et al.,
1967; Vickers, 1968). In addition, various anticonvulsant and other agents
have been tested using animal models (e.g. ethosuximide, sodium valproate,
clonazepam, diazepam, L-dopa, phenobarbitone); however, although there
were some EEG changes, the results appeared to be species specific
(Mamelak et al., 1986). An investigation by Henderson et al. (1976) showed
that intravenous physostigmine was effective in reversing the anaesthetic
action of GHB in 25 patients. These results were confirmed by Schöntrube
et al. in 1993. Due to the rapid gastro-intestinal absorption of GHB, gastric
lavage and administration of activated charcoal are of limited use. Treatment
of GHB intoxication is therefore largely supportive and intubation with
mechanical ventilation is sometimes used (particularly to protect the airway
if the patient is vomiting) (Appleton and Burn, 1968). However, in the majority
of cases the patient awakes spontaneously within approximately 7 hours
(presumed to be due to the short elimination half-life of GHB).

Cases of GHB intoxication in humans

Non-fatal cases

As GHB is not usually detected during routine toxicological analysis
(Badcock and Zotti, 1999; Williams, 1998; Elliott, 2000) the evidence for
GHB or related product ingestion (e.g. GBL or 1,4-butanediol) is usually
based on anecdotal or circumstantial evidence.

There have been many reported cases of intoxication linked to GHB, however,
there also appear to be many more unconfirmed/anecdotal reports

(; ).

There have been other
reports of toxicity resulting from ingestion of GBL or 1,4-butanediol; the
patients presented with identical symptoms to cases involving GHB ingestion
(CDC, 1999; Dyer et al., 1997; Rambourg-Schepens et al., 1997). This
is consistent with the reported in vivo conversion of these compounds to
GHB (Poldrugo and Snead, 1984; Lettieri and Fund, 1978).

The majority of reported cases have occurred in the United States (Couper
and Logan, 2000; Chin et al., 1992; Chin et al., 1998; Li et al., 1998; Stokes
and Woekener, 1998; CDC, 1999; CDC, 1997; FDA, 1990; FDA, 1997;
CDC, 1990; CDC, 1997; Dyer et al. 1991; Steele and Watson, 1995; Dyer,
1991; Viera and Yates, 1999; Eckstein et al., 1999) and Europe

(;

Williams et al., 1998; EMCDDA; Elliott, 2000; Kouagie et
al., 1997; Vandevenne et al., 2000; Hovda et al., 1998; Personne and
Landgren, 2000; Knudsen, 2000; Hunderup and Jorgensen, 1999) although
abuse of GHB has also been reported in Australia (Australian Drug
Foundation). Based on documented cases and reports to Reitox national
focal points, it can be estimated that there have been at least 200 presumed
GHB overdose cases in Europe (EMCDDA); global estimates range from hundreds
to thousands of cases (CDC, 1997a; FDA, 1990; FDA, 1997; CDC,
1990; CDC, 1997b). In Sweden and the United Kingdom alone, there have
been at least 100 apparent GHB-related hospital admissions since 1996.
Eight cases have been reported in the Netherlands, 12 cases in Denmark,
two cases in Belgium, two cases in Finland, three cases in Norway and one
case in Spain (EMCDDA).

Williams et al. (1998) reported six cases of probable GHB intoxication
occurring between 1995 and 1996 in London, United Kingdom. These cases
are summarised in Table 2. The clinical observations in these cases confirm
those of other cases where it appears that patients present in various states
ranging from initial confusion, dizziness or euphoria, leading to collapse,
vomiting and loss of consciousness/coma (Chin et al., 1992; Chin et al.,
1998; Li et al., 1998; Li et al. 1991; Dyer et al., 1991; Steele and Watson,
1995; Dyer, 1991; Viera and Yates, 1999; Eckstein et al., 1999).
Administration of naloxone and flumazenil did not appear to have a significant
effect and in the majority of cases activated charcoal was administered
and the patient was intubated. All patients eventually recovered and were
either discharged or self-discharged. The reported ‘dose’ of GHB varied,
however the true amount/concentration of GHB ingested was unknown, as
the exact composition of the GHB product was not usually
ascertained/analysed. Furthermore, in these particular cases it was not
known/confirmed if other drugs were ingested which may have exacerbated
the effects; however, the co-ingestion of alcohol (ethanol) was frequently
mentioned.

In some cases, however, extensive drug screening has been performed and
the presence of GHB has been confirmed and the concentration measured/
estimated in biological fluid (Couper and Logan, 2000; Elliott, 2000;
Louagie et al., 1997; Vandevenne et al., 2000, le Gatt et al., 1999; Baselt,
2000; Dyer et al., 1994). A selection of these cases is presented in Table 3.

Fatal cases

Approximately 65 deaths in the United States have been linked to GHB since
1990 (FDA, 1997a). In Europe, approximately 11 deaths in which GHB has
been implicated have been reported since 1995. United Kingdom (four deaths
— September 1995, March 1996, November 1997 and January 1999),
Sweden (four deaths — February 1996, March 1997, 1998–2000), Finland
(two deaths — 1998 and 1999) and Denmark (one death — January 2000)

(; ; EMCDDA; Elliott, 2000).

However, recently there have been a further two unconfirmed cases in the
United Kingdom and one case in Sweden (EMCDDA; Mixmag, 2000). Table 4
shows reported cases involving GHB or GBL ingestion. Due to in vivo conversion
of GBL to GHB, only GHB is usually detected in biological fluids
analysed in such cases. The majority of cases have involved the ‘recreational’
abuse of GHB for its apparent euphoric or ‘high’ effects, primarily by young people.

There are certain factors that should be noted in GHB cases:

 The presence of other drugs (particularly alcohol and opiates/opioids e.g.
heroin, codeine, methadone and morphine).

 Some researchers describe the presence of GHB in post mortem blood
specimens, in cases where there has been no evidence of GHB.

 The GHB concentration found is sometimes low.

The mode of abuse of GHB frequently involves the use of other drugs such
as alcohol or MDMA, therefore, deaths involving solely GHB appear to be
rare. The presence of alcohol and other depressant drugs is widely believed
to exacerbate the toxic effects of GHB ingestion. Therefore, the presence of
such drugs in deaths involving GHB should be taken into consideration
when assessing fatalities attributed to GHB intoxication. Ferrara et al. (1995)
reported a death involving GHB and heroin (diacetylmorphine). A high concentration
of morphine was detected in the blood (770 mg/l). In most of the
other reported GHB deaths, ethanol has also been involved at significant
concentrations (EMCDDA; Hale; Davis, 1999). In the United Kingdom case
in 1995 involving both GHB and ethanol, the mechanism of death was stated
to be respiratory depression (Hale).

Recently, several researchers have reported that GHB was present in significant
concentrations in post mortem blood, even in cases where the decedents
had died in circumstances apparently unrelated to GHB (Fieler et al.,
1998; Anderson and Kuhwahara, 1997; Stephens et al., 1999). In 1998,
Fieler, Coleman and Baselt detected GHB in 15 out of the 20 post mortem
blood specimens analysed. The apparent concentrations ranged from
3.2–168 mg/l (average = 25 mg/l) using GC-MS analysis. Subsequent
reanalysis using GC-FID confirmed these findings. No GHB was detected in
the blood or urine of living patients, in addition, no GHB was detected in
eight post mortem urine specimens analysed. They suggested that GHB is a
product of post mortem decomposition. Further work by Stephens, Coleman
and Baselt was published in 1999 indicating that certain storage conditions
could elevate the concentration of GHB in post mortem blood samples;
namely if the sample was stored in a non-fluoridated container above 4°C.
Again, they found concentrations within the range (9–433 mg/l) in post
mortem blood (average = 57 mg/l) and only detected GHB in 3 out of 17
post mortem urine specimens. If confirmed by further studies this phenomenon
has profound implications for the interpretation of post mortem GHB
concentrations.

In the majority of GHB-related deaths the concentration in post mortem
blood has been found to be ‘high’, however in several cases the concentration
was found to be relatively low, e.g. less than 50 mg/l. Such concentrations
are within the range of GHB concentrations apparently produced post
mortem, as stated above. Furthermore, in living persons, similar concentrations
have been detected in unconscious patients who awake a few hours
later with no obvious side effects. Due to the rapid absorption and metabolism
of GHB, however, it is difficult to predict how much of the original dose
such post mortem concentrations represent.

In conclusion, more research and thorough analysis of GHB in fatalities and
poisonings is still required before the true involvement of GHB can be established
and accurate mortality and morbidity figures produced.

PHARMACOKINETICS

In 1969, Roth and Giarman demonstrated that [3H]GABA is converted to
[3H]GHB via succinic semialdehyde (intermediate compound) in brain tissue.
This was later confirmed by Anderson et al. (1977). The conversion is
catalysed by the enzymes; GABA aminotransferase and succinic semialdehyde
reductase (Figure 3).

Succinic semialdehyde reductase has been found to be different between
species; in human and pig brain the enzyme is dimeric (MR between 82 000
and 110 000 Da), whereas it exists as a monomeric protein in rat and bovine
brain tissue. The enzyme has also been isolated in the mitochondria and as
the substrate for succinic semialdehyde is synthesised in mitochondria, it has
been postulated that the mitochondrion is the site of GHB synthesis, with
subsequent transport to the cytosol. As previously mentioned, GHB can also
be produced after administration of γ-butyrolactone (GBL) or 1,4-butanediol.

GHB is purported to be metabolised via succinic acid and the citric acid
cycle (TCA cycle/Krebs cycle), ultimately producing carbon dioxide and
water. GHB conversion to succinic semialdehyde can be catalysed by
cytosolic GHB-dehydrogenase (accounts for majority of GHB metabolism in
the young animal foetus) or mitochondrial GHB-ketoacidtranshydrogenase
(responsible for majority of GHB metabolism in adult animals) (Kaufman et
al., 1979; Nelson and Kaufman, 1994). Although GHB has the potential to
produce GABA, this was not observed after injecting mice with radiolabelled
GHB (De Feudis and Collier, 1976). Laborit (1964) also postulated that GHB
‘orientated’ glucose-6-phosphate (G6P) into the pentose phosphate pathway
(produces ribose for nucleic acid synthesis and NADPH).

In humans, GHB is rapidly absorbed, with peak plasma concentrations (Cmax)
occurring within 20–60 minutes post oral dose (tmax = 20–60 min.). This is
consistent with the onset of effects occurring approximately 15 minutes after
an oral dose and can last for up to seven hours, depending on the dose
(Galloway et al., 1997). Effects of intravenous dosage have been reported to
occur within minutes, post administration (Takahara et al., 1977). Following
a 12.5 mg/kg dose, the half-life was 20 minutes (Vickers, 1969). Only 2–5 %
is eliminated as unchanged drug in urine (Laborit, 1964; Hoes et al., 1980).

Clinical experience

Clinical safety data

Preclinical studies

 Laborit (1964) observed that in women in labour, GHB had a ‘spectacular
action on the dilation of the cervix’, an effect which was apparently independent
of the anti-anxiety and reduced consciousness obtained.
Furthermore, in 1962, Barrier reported that GHB was beneficial in obstetric
surgery due to the absence of respiratory depression in the infant and its
anti-shock property against possible cardiac anoxia (Laborit, 1964).

 Several researchers have observed an anti-anxiety effect of GHB. This was
reported in a preliminary study by Danon-Boileau et al. in 1962, involving
schizophrenic patients. 500 mg of GHB four times a day produced a temporary
‘disinhibiting effect’ and relaxed the patients (Laborit, 1964).
However, a large proportion of reports regarding GHB’s anti-anxiety
effects appear to remain anecdotal.

 In 1972, Laborit remarked on GHB’s ‘aphrodisiac’ actions in humans.
There have been many anecdotal reports which suggest that GHB has four
sexual enhancing effects; disinhibition (e.g. relaxation), heightened sense
of touch, enhancement of male erectile capacity and increased intensity of
orgasm.

 The clinical evidence pertaining to GHB’s possible antidepressant effects
are largely anecdotal. However, Laborit (1964) suggested that the increase
of acetylcholine and dopamine levels in the brain and the apparent
increase in cerebral protein synthesis, serotonin turnover and aspartic acid
levels by GHB, may correct metabolic disturbances secondary to depressive
states.

Clinical studies

GHB AS AN ANAESTHETIC AGENT

In the 1960s, early work involving GHB assessed its potential as an anaesthetic
agent (Laborit, 1964; Appleton and Burn, 1968; Vickers, 1969).
Anaesthetic doses within the range 60–70 mg/kg were given intravenously to
a patient. GHB has been reported to be involved in over 6 000 cases in general
anaesthesia, and Laborit noted various advantages compared to other
general anaesthetics, including: non-hypotensive bradycardia, muscle relaxant
properties, absence of respiratory depression while the response of the
respiratory centre to CO2 is maintained, anti-shock activity, allows easy
induction and maintenance of hypothermia, no venous irritation and apparent
low toxicity. However, various disadvantages have also been noted
including: lowers serum potassium levels, duration of action is too unpredictable,
only produces complete general anaesthesia in children and poor
pain control. The autonomic nervous system remains active — therefore, as
for other anaesthetics, administration of other agents are required such as
opioid analgesics. Mainly because of the unpredictable duration of action,
GHB was nearly displaced as an anaesthetic agent. However, owing to the
rapid metabolism of GHB and the reliable induction of sedation and anaesthesia
without depressing either respiratory or cardiocirculatory parameters
or liver and kidney function, GHB is being re-evaluated as an agent in emergency
and critical care medicine, mainly in long-term sedation of patients
(Diedrich et al., 1996; Kleinschmidt et al., 1995; Pichlmeier and Schneck,
1991; Pospiech and Schmidt, 1993).

USE OF GHB IN THE TREATMENT OF NARCOLEPSY AND ASSOCIATED CATAPLEXY

Various researchers have studied the use of GHB as a potential treatment for
narcolepsy (Mamelak et al., 1986; Broughton and Mamelak, 1979; Scarf et
al., 1985; Scrima et al., 1990; Delay et al., 1993) due to its sleep-inducing
properties. It was thought that in narcoleptic patients GHB would act to ‘normalise’
sleep patterns and reduce the problems associated with the disorder
such as cataplexy (sudden loss of muscle tone), sleep paralysis, daytimedrowsiness
and hypnagogic events (hallucinations that occur at the onset of
sleep). Mamelak obtained clinical data on 48 narcoleptic patients who had
been treated with GHB for up to 9 years. As GHB-induced sleep wears off
after about 3–4 hours post dose, patients took 2.25–3.0 g of GHB two or
three times a night (i.e. upon waking) (Mamelak et al., 1986). Within the first
few weeks of treatment, many of the patients reportedly felt more alert during
the day and there was a reduction in hallucinations, cataplexy and sleep
paralysis (although this did intensify on the first or second night). Symptoms
appeared to intensify during periods of stress, however, few adverse effects
were observed. A degree of weight loss was also reported in some obese
patients. Daytime-drowsiness continued to occur in many of the patients and
some were prescribed stimulants in the morning such as ‘Dexedrine’ (damphetamine)
as part of their treatment regimen, in order to achieve the
optimal levels of sleep at night and wakefulness during the day. Other studies
noted the occurrence of intermittent episodes of sleepwalking in some GHBtreated
patients and if sleep is resisted the patient may become confused and
emotionally labile (Mamelak et al., 1986; Scarf et al., 1985).

USE OF GHB IN ALCOHOL AND OPIATE WITHDRAWAL

The use of GHB in alcohol withdrawal has been investigated by various
researchers. In 1989, Fadda et al. treated alcohol-addicted rats with either
GHB (at various doses), ethanol or a placebo, 8 hours after the last dose of
alcohol. The degree of withdrawal tremor was observed. It was found that
GHB appeared to reduce the tremor over a 2-hour period. Gallimberti et al.
(1989) assessed the effectiveness of a solution of GHB (AlcoverTM) in 23 alcoholic
humans. One group received 50 mg/kg of GHB and the other group
received a similar tasting placebo. A withdrawal symptom score was
obtained at baseline and 1–7 hours after treatment. The score was based on
the occurrence of tremors, sweating, nausea, depression, anxiety and restlessness.
It was found that GHB-treated patients had a significantly consistent
reduced withdrawal score, post treatment, compared to the placebo
control group. These results appeared to support those observed in rats. In
an additional study involving 82 patients, Gallimberti et al. (1992) showed
that GHB was effective in reducing alcohol consumption and craving for
alcohol. However, no long-term outcome was evaluated.

Further evidence for the effectiveness of GHB in the treatment of alcohol
withdrawal syndrome has involved randomised, controlled clinical studies
to compare GHB with well established benzodiazepines. In one single blind
study involving 60 alcoholic patients, an oral dose of 50 mg/kg GHB was
compared with 0.5–0.75 mg/kg diazepam for 10 days (Addolorato, 1999).
No significant difference in the overall efficacy was observed. A second
study involving 43 alcoholic patients, compared an intravenous GHB dose
(50 mg/kg then continued by 10–20 mg/kg/h per infusion) against intravenous
flunitrazepam (0.2–2 mg bolus continued by 0.015–0.08 mg/kg/h)
for up to 30 days in intensive care patients (Lenzenhuber et al., 1999). If
necessary, clonidine and haloperidol were administered to treat autonomic

signs of withdrawal and hallucinations, respectively. Again, no difference in
overall efficacy was observed between the two compounds. However, GHBtreated
patients required significantly higher haloperidol (possibly due to
hallucinogenic properties of GHB) and lower doses of clonidine.

In 1993, Gallimberti et al. treated 27 heroin and methadone-dependent
patients with 25 mg/kg of GHB. All patients were in withdrawal and an additional
14 patients were used as placebo controls. A similar withdrawal score
to that used for the alcohol study was obtained up to 3 hours after treatment
and finally assessed on the eighth day of treatment, before and after administration
of naloxone. The results showed that GHB was effective in reducing
all the signs of opiate withdrawal symptoms, except for diarrhoea and
insomnia, over the 8 days. Although three methadone-dependent subjects
and two heroin-dependent subjects reported transient dizziness or vertigo on
the second/third day, no other side effects were attributable to the administration
of GHB. One limitation of the use of GHB for withdrawal was noted
to be its short duration of action, as frequent doses of GHB would be
required.

Non-clinical use of GHB (including subjective effects in humans)
It appears that GHB or related products (e.g. GBL and 1,4-butanediol) are
used by various groups of people. The use and abuse of GHB has increased
since 1990 and has been accompanied by an increased presence of GHBrelated web sites on the Internet.

 Bodybuilders exploit the possible growth hormone promoting properties of
GHB in an attempt to increase muscle mass. GHB is therefore illicitly
sold/distributed in gymnasiums or advertised on the Internet or related
web sites. Some people therefore erroneously refer to GHB as an anabolic
steroid, which is not the case, as its chemical structure does not resemble
a steroid.

 Other people sometimes use GHB as an apparent appetite suppressant or
weight loss product, although there is very little definite scientific data to
support these claims.

 Due to GHB’s sleep-inducing effects, various people suffering sleep disorders
such as insomnia or narcolepsy use GHB products in an attempt to
normalise their sleep patterns.

 Some groups have actively promoted (again usually via the Internet) the
potential anti-ageing effects of GHB due to claimed indirect anti-oxidant
properties of the compound by stimulating the glial cell pentose phosphate
pathway producing NADPH for the reduction of oxidised glutathione

(South, ).

 GHB is also used as a sexual adjunct to enhance libido and sexual function,
by both heterosexuals and homosexuals. Therefore, various GHB or
related preparations are also sold in ‘sex shops’.

 The apparent primary mode of abuse worldwide has been the use of GHB
for its subjective hypnotic, euphoric and hallucinogenic properties.
Although some users reportedly use GHB ‘to relax’ and may use it as an
alternative to alcohol, many users attempt to attain a desired ‘high’, similar
to that sought from ‘ecstasy’ (e.g. MDMA). Hence, liquid GHB is sometimes
referred to as ‘liquid ecstasy’, ‘liquid X’ or ‘liquid E’, although the
mode of action and chemical structure of MDMA and GHB are considerably
different. GHB has been found to be associated with social gatherings
such as parties, nightclubs, music events (e.g. ‘raves’ or festivals), drinking
establishments, etc. In such situations there is the danger of concomitant
ingestion of other drugs or alcohol, which will potentiate the effects of
GHB. The majority of reported hospital admissions and deaths have been
related to such instances of abuse.

 Recently, there has been the suggestion that GHB has allegedly been used
for illicit sexual activity or ‘date rape’, due to the potential incapacitating
and sleep-inducing effects of GHB (and GBL or 1,4-butanediol) (Smith,
1999; Sohley and Salamone, 1999). As GHB is colourless and easily dissolves/
mixes in aqueous solutions (e.g. water and other liquids), it can be
surreptitiously introduced into beverages. The required dosage to cause
such effects, however, may require the introduction of possibly large
noticeable quantities of GHB powder or liquid depending on the formulation
and purity of the GHB used. Furthermore, if GHB sodium salt or solution
is used, a slight salty taste may be noticeable, particularly if introduced
into a previously tasteless liquid such as water (Ward et al., 1998).
Despite this, the use of GHB in such illicit activity is a contentious area of
GHB abuse, as unfortunately it is usually difficult to prove, given the rapidity
of GHB metabolism and elimination.

Dependence potential in humans

Dependence potential

There have been few studies regarding the dependence/abuse potential of
GHB. However, during the numerous studies involving administration of
GHB to patients at varying concentrations, no dependence has been
observed at low doses of GHB. At prolonged high doses, however, physical
dependence as evidenced by a withdrawal syndrome has been noted in
some cases and includes symptoms of insomnia, muscular cramping, tremor
and anxiety (Galloway et al., 1997).

Further studies indicated that GHB maintained self-administration marginally
above saline and water (in monkeys and rats, respectively) (Colombo et al.,
1995). Studies of the reinforcing and discriminative stimulus effects of GHB
in monkeys and rats have indicated that GHB was partly substituted for by
morphine, LSD, chlordiazepoxide and GABA-mimetics such as muscimol,
GBL, baclofen and 3-aminopropane sulfonic acid in rats (Winter, 1981).
GHB did not appear to substitute for d-amphetamine, pentobarbital,
diazepam and triazolam in rhesus monkeys (Woolverton et al., 1999).
Woolverton et al. (1999) concluded that GHB has, at most, a low abuse
potential.

Psychological risk assessment (cognition, mood and mental
functioning)

Acute and chronic effects

In general, there are few published data concerning specific psychological
effects of GHB either acutely or chronically. However, recently, Ferrara et al.
(1999) studied the effects of single dose GHB on psychomotor performance
and monitored the patients’ subjective feelings. GHB at doses of 12.5 mg per kg
and 25 mg per kg, including placebos were given to six male and six female
volunteers. The subjects’ psychomotor performance was evaluated at baseline
and at 15, 60, 120 and 180 minutes post dose using critical flicker
fusion, response competition test, critical tracking task, choice reaction time
and visual vigilance task. The subjects’ mood was assessed before and
120 minutes after treatment using 16 visual analogue scales. The results indicated
that GHB at either dose had no effect on vigilance, attention, alertness,
short-term memory or psychomotor coordination. Calmness increased at the
lower dose and the subjects apparently felt more contented at both dose regimens.
Observed adverse effects consisted of subjective dizziness and dullness
but these feelings disappeared 30–60 minutes after administration.

Various observations made in studies involving the administration of GHB to
patients include relaxation, nausea, agitation and sedation/coma (London
Toxicology Group; Takahara et al., 1977; Chin et al., 1992; Mamelak et al.,
1986; Vickers, 1969). No long-term effects have been reported. After awakening
patients are usually alert and feel ‘refreshed’, however, there have
been some reported instances of transient mental disturbance (Steel, 1968).
Galloway et al. (1997) reported that one patient using GHB (dose unknown)
described effects similar to those experienced after alcohol ingestion and
also that it impaired his ability to drive. Two cases have been reported where
GHB has been detected in the blood and urine of drivers. In one case the
driver was found asleep at the wheel after stopping in a traffic lane (Stephens
and Baselt, 1994) and a second driver involved in a road traffic accident was
reported to be confused, combative and delirious (Baselt, 2000).

Other researchers have evaluated the effect of GHB in the treatment of schizophrenia
as it was thought that GHB-induced inhibition of dopamine neuronal
firing would be beneficial to afflicted individuals. Levy et al. (1983)
and Schulz et al. (1981) in two separate studies generally did not observe
any significant antipsychotic efficacy of GHB.

Conclusions

 GHB is not a new synthetic drug. It was synthesised in 1960 by Laborit but
was later found to be a naturally occurring compound in mammalian brain
and other tissues.

 Evidence relating to the activity of GHB on neurotransmitter systems is
largely contradictory, however, it appears that GHB is a hyperpolarising
compound that blocks dopamine release at the synapse and produces an
increase in intracellular (neuronal) dopamine.

 GHB has been reported to produce enhanced slow-wave/delta sleep without
a decrease in oxygen consumption while the respiratory centre
remains sensitive to carbon dioxide. It also induces anaesthesia but does
not provide pain relief. An increase in growth hormone and prolactin
release has been reported in one study of six human subjects.

 GHB can cross the blood-brain barrier and can be produced in vivo as a
product of GABA metabolism and after administration of GBL (γ-butyrolactone)
or 1,4-butanediol. GHB is rapidly absorbed and metabolised,
possessing a plasma half-life of approximately 20 minutes and has a steep
dose-response curve. Following an oral dose, effects usually occur after
15 minutes and can last up to seven hours, depending on the dose.

 GHB has been evaluated for various potential therapeutic uses including
obstetrics, anaesthesia, alcohol/opiate withdrawal and treatment of narcolepsy
and cataplexy.

 Reports indicate that GHB is used for various reasons and by various sections
of society. These include: its sexual enhancing effects, growth hormone
promoting effects (e.g. apparently increasing muscle bulk), relaxation
and antidepressant effects, postulated anti-ageing properties and
more recently its apparent euphoric (‘high’) effects. There have also been
reports of GHB allegedly being used in cases of so-called ‘date rape’.

 There is limited published data concerning specific psychological effects
of GHB either acutely or chronically, therefore the exact effect of GHB on
cognition, mood and psychomotor ability is unclear.

 Animal and human studies indicate that GHB toxicity is dose-dependent
and can result in coma, random clonic movements, decrease in body temperature,
hypotonia, nausea, vomiting, bradycardia, respiratory depression
and apnoea. A possible physical dependence has been observed at prolonged
high dosage.

 Other depressant or sedative drugs (e.g. opiates, benzodiazepines, alcohol
and barbiturates) and possibly other psychoactive compounds (e.g.
amphetamine) may exacerbate the effects of GHB.

 In humans, non-fatal instances of intoxication and also deaths implicating
GHB have been reported.

(7) This report was written by S. P. Elliott of the Regional Laboratory for Toxicology (City
Hospital NHS Trust, Birmingham, United Kingdom) for the risk-assessment meeting on GHB;
Lisbon 25 and 26 September 2000, EMCDDA.

< Prev		Next >