The use of benzodiazepines has increased sharply since they were first synthesised in the 1950s, and they have gradually replaced barbiturates and other hypnotic and anti-anxiety drugs (1). In 2008, 6 % of the Norwegian population was prescribed at least one benzodiazepine (2). Since the 1990s, z-hypnotics such as zopiclone (Imovane, Zopitin) and zolpidem (Stilnoct), have taken over for insomnia (3).
Benzodiazepines and z-hypnotics are chemically dissimilar, but both act by increasing the effect of the neurotransmitter gamma-aminobutyric acid (GABA) on GABAA receptors in the brain. This reduces neuronal excitability, with a hypnotic, anxiolytic and anticonvulsive effect as a result (4, 5).
Benzodiazepines can also be used as intoxicants, and have adverse long-term effects in the form of impaired cognitive abilities and risk of dependence. Overdosing results in altered level of consciousness, impaired coordination, amnesia and respiratory depression. The risk of an overdose is greatest in combination with other intoxicants with depressant effects. Flumazenil is an antidote, but has a short half-life and may cause seizures (4, 5).
Z-hypnotics cause less intoxication and entail less risk of inducing dependence (5), but the mechanism is otherwise so similar to benzodiazepines that we have chosen to include them in this study. In the following we use benzodiazepines as a collective term.
12 % of benzodiazepine users account for 59 % of prescribed use in Norway (6). A Swiss study found that heavy users often alternate between legal and illegal strategies to acquire benzodiazepines (7). In an anonymous British questionnaire survey (8), sources of benzodiazepines were general practitioners (55 %), friends and family (40 %), online purchasing (27 %), street sales (20 %) and abroad (11 %). Multiple sources were used by 31 %.
We describe acute poisonings treated by the main A&E clinic in Oslo, due to substance abuse of benzodiazepines, with emphasis on the type of benzodiazepine and other intoxicants they were taken in combination with.
Material and method
We included all cases of substance abuse poisoning with benzodiazepines treated by the main A&E clinic in Oslo (the Oslo Accident and Emergency Outpatient Clinic (OAEOC)), and retrospectively reviewed cases registered by the reception nurse in the patient registration lists in the A&E clinic’s electronic record system in the period October 2013–September 2015. Poisonings with suicidal intention were not included.
From patient records we noted age, sex, toxic agent, whether the patient was brought by ambulance, observation period, treatment, disposition and a fixed set of symptoms and signs, in accordance with a data registration template developed by the research network European Drug Emergencies Network (Euro-DEN) (9). Diagnosis of the toxic agent was based on the attending doctor’s assessment as noted in the records, which in turn was based on information from the patient and/or supplied by third parties, and on clinical findings. No toxicological laboratory diagnostic work was performed.
The project was carried out as a quality improvement study and was approved by the Information Security and Privacy Office at Oslo University Hospital.
SPSS version 25 was used to perform analyses. We used the chi-squared test to compare categorical variables.
Results
Benzodiazepines were involved in 1 037 substance abuse poisonings in the course of a two-year period. The median age of patients was 36 years (interquartile range 28–46, range 14–78, and 787 (76 %) were men.
Clonazepam (Rivotril) occurred most frequently, with 575 (55 %) cases, followed by diazepam (Stesolid, Valium, Vival) 158 (15 %) and alprazolam (Xanor) 125 (12 %) (Table 1).
Table 1
Combinations of benzodiazepines (including z-hypnotics) and other intoxicants in acute poisoning cases treated at the main A&E clinic in Oslo from 1.10.2013–30.09.2015. Data are stated as number (%) The total number may be higher than the sums in columns or rows, as several benzodiazepines might be combined with the same intoxicant and several intoxicants could be combined with the same benzodiazepine. GHB = gamma hydroxybutyrate, MDMA = methylenedioxymethamphetamine
|
|
Heroin
|
Methadone
|
Buprenorphine
|
Other opioids
|
Amphetamine
|
Cocaine
|
MDMA
|
GHB
|
Cannabis
|
Ethanol
|
Other / unknown
|
Only benzodiazepines
|
Total
|
|
Clonazepam (Rivotril)
|
318 (66)
|
28 (62)
|
23 (68)
|
21 (29)
|
131 (66)
|
18 (56)
|
10 (56)
|
31 (63)
|
62 (65)
|
159 (50)
|
10 (43)
|
45 (45)
|
575 (55)
|
|
Diazepam (Stesolid, Valium, Vival)
|
48 (10)
|
8 (18)
|
1 (3)
|
10 (14)
|
24 (12)
|
5 (16)
|
2 (11)
|
8 (16)
|
18 (19)
|
82 (26)
|
6 (26)
|
13 (13)
|
158 (15)
|
|
Alprazolam (Xanor)
|
77 (16)
|
4 (9)
|
6 (18)
|
7 (10)
|
27 (14)
|
6 (19)
|
3 (17)
|
5 (10)
|
10 (11)
|
23 (7)
|
1 (4)
|
12 (12)
|
125 (12)
|
|
Oxazepam (Sobril)
|
28 (6)
|
4 (9)
|
3 (9)
|
8 (11)
|
14 (7)
|
2 (6)
|
2 (11)
|
3 (6)
|
8 (8)
|
35 (11)
|
3 (13)
|
14 (14)
|
94 (9)
|
|
Flunitrazepam (Rohypnol)
|
14 (3)
|
1 (2)
|
1 (3)
|
2 (3)
|
7 (4)
|
1 (3)
|
2 (11)
|
4 (8)
|
4 (4)
|
7 (2)
|
-
|
5 (5)
|
31 (3)
|
|
Zopiclone (Imovane, Zopitin)
|
3 (1)
|
-
|
-
|
2 (3)
|
-
|
-
|
-
|
-
|
2 (2)
|
13 (4)
|
-
|
7 (7)
|
25 (2)
|
|
Nitrazepam (Apodorm, Mogadon)
|
7 (1)
|
-
|
1 (3)
|
2 (3)
|
2 (1)
|
1 (3)
|
-
|
2 (4)
|
-
|
5 (2)
|
-
|
5 (5)
|
19 (2)
|
|
Zolpidem (Stilnoct)
|
-
|
-
|
-
|
2 (3)
|
-
|
-
|
-
|
-
|
-
|
4 (1)
|
-
|
5 (5)
|
11 (1)
|
|
Flurazepam
|
-
|
1 (2)
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
1 (<0.5)
|
|
Unspecified
|
62 (13)
|
2 (4)
|
2 (6)
|
26 (36)
|
22 (11)
|
3 (9)
|
1 (6)
|
7 (14)
|
7 (7)
|
35 (11)
|
5 (22)
|
15 (15)
|
142 (14)
|
|
Total
|
484 (100)
|
45 (100)
|
34 (100)
|
72 (100)
|
199 (100)
|
32 (100)
|
18 (100)
|
49 (100)
|
95
(100)
|
321 (100)
|
23 (100)
|
101 (100)
|
1037 (100)
|
Benzodiazepines were combined with other intoxicants in 936 cases (90 %), most frequently heroin 484 (47 %), ethanol 321 (31 %) and amphetamine 199 (19 %).
Patients were brought to the A&E clinic by ambulance in 569 cases (55 %). On admission 47 patients (5 %) were comatose (Glasgow Coma Scale (GCS) score ≤ 8), 619 (60 %) somnolent (GCS score 9–14) and 111 (11 %) had a respiratory rate < 12/min. During the course, 150 (14 %) patients were agitated, 42 (4 %) experienced anxiety and 41 (4 %) were hypotensive. Naloxone (antidote against opioids) was administered in 187 cases (18 %), sedation in 13 (1 %) and flumazenil in 2 (0.2 %). The median observation period was 4 h 25 min (interquartile range 2 h 30 min – 6 h 11 min, range 3 min – 36 h 53 min). In 168 cases (16 %) the patient was hospitalised in a somatic hospital, and in 20 (2 %) in a psychiatric unit. No patients died at the A&E clinic.
A respiratory rate of < 12/min was more common when benzodiazepines were combined with opioids, 101/601 (17 %) cases vs 10/436 (2 %) (p < 0.001), and naloxone was administered more frequently for this combination: 171/601 (28 %) cases vs 16/436 (4 %) (p < 0.001).
Discussion
In substance abuse poisonings, benzodiazepines were very often taken in combination with other intoxicants – probably to enhance the effect, counteract undesirable side effects or relieve abstinence symptoms. This is in line with other European studies based on a clinical assessment of toxic agents (10, 11). Even more cases prove to involve benzodiazepines, when clinical assessments are supported by laboratory analyses (12).
Clonazepam was the dominant toxic agent. This stands in contrast to the Norwegian prescription pattern. Z-hypnotics accounted in 2014–2015 for two thirds of prescriptions, and 50 % more of both diazepam and oxazepam was prescribed than clonazepam, and almost the same amount of nitrazepam as clonazepam (3). However, police seizures clearly reflect the distribution of the most frequently occurring benzodiazepines in substance abuse poisoning. Clonazepam accounted for 52 % of the number of benzodiazepine seizures and 47 % of the quantity seized in 2014, followed by diazepam (20 % and 33 %) and alprazolam (13 % and 18 %), while the quantity of z-hypnotics seized was minimal (13). Z-hypnotics seldom occur in substance abuse poisoning, probably because the intoxicant effect is relatively limited (5). Although Lyphout et al. found that national prescription patterns correlated with the distribution of benzodiazepines in substance abuse poisonings (11), the intoxication potential and availability on the illegal market appear to be of greater importance than prescription by doctors.
Altered level of consciousness was the most prominent clinical finding, as expected (4, 5). Patients with assumed substance abuse intoxication are observed systematically at the OAEOC according to a locally developed procedure (14). Antidote is administered only in cases of respiratory depression. Most patients with respiratory depression had taken opioids in addition. Naloxone is a safer antidote than flumazenil and is therefore administered first. The fact that flumazenil was only administered in two cases shows that naloxone was virtually always sufficient and that respiratory depression is rare in cases involving benzodiazepines alone. If the respiratory rate does not improve, or if the patient has not woken within four hours and can give an account of themselves, the patient is sent on to hospital. Although the OAEOC has good observation possibilities and a short distance to hospital, we believe that our results and observations might be applicable to other A&E clinics with observation possibilities, and to hospital emergency departments.
Strengths and weaknesses
Toxicological screening is not carried out at the OAEOC. Diagnoses of toxic agents were accordingly based mainly on self-reporting, which may be deficient or incorrect owing to patients’ altered level of consciousness, tablets of unknown content from street and online sales, unwillingness to cooperate and/or fear of legal reprisals. There may also be variation in doctors’ record-keeping and in the inclusion and registration practice of data collectors.
Although the majority of substance abuse poisonings in Oslo are treated at the A&E clinic, the sickest patients are taken directly to hospital by ambulance (10). The picture of benzodiazepine poisonings in the capital is therefore not complete. Nor does the study cover benzodiazepine poisonings with suicidal intent, a far larger share of which are sent on from the A&E Clinic to hospital (10). Our material is five years old, and the prevalence of the different intoxicants may have changed since then.
Conclusion
Benzodiazepines are most often taken in combination with other intoxicants in substance abuse poisonings, most frequently heroin, ethanol and/or amphetamine. Clonazepam was the most prevalent benzodiazepine.
