Settings

Oslo is the capital city of Norway and had a population of 658,390 as per 1 January 2016 [10]. About one million live in the Oslo metropolitan area. The OAEOC is a primary care emergency outpatient clinic, serving the entire city at all hours. There are about 200,000 consultations per year. The Norwegian emergency health care system is two-tiered with a gate-keeping function. Unless triaged for hospital care by the ambulance service, patients are seen in primary care and cannot present directly to a hospital emergency department (ED). Concerning poisoning with alcohol and/or recreational drugs, the OAEOC functions as a pre-ED for the city hospitals, managing the less complicated cases with limited diagnostic and treatment resources [11]. The OUH is one of four hospitals in Oslo, with both primary and tertiary referral functions. About 4600 patients are seen in the OUH ED per year.

Data from the two centers were analyzed separately. As some patients were transferred from the OAEOC to the OUH, the patient populations have some overlap.

Participants

Using the case definition developed by the European Drug Emergencies Network (Euro-DEN) [12], we registered all patients with acute toxic effects related to recreational drug use presenting at the two centers. Cases involving only alcohol were not included, nor poisonings related to self-harm, suicide attempts, inflicted by others, or accidental exposures. Eligible cases were found by searching the patient registration lists in the electronic patient records.

From this material, we extracted all cases involving at least one central stimulant drug. From the OAEOC total of 3733 cases, 1131 (30.3%) were included. From the OUH total of 457 cases, 211 (46.2%) were included.

Data collection and classification

We collected data from the electronic patient records at both centers and from paper observational charts at the OAEOC, using the Euro-DEN variable set [12]. We registered age, gender, time of presentation, time of discharge, whether the patient was brought by ambulance, disposition from the ED, death in hospital, toxic agents taken, clinical observations at presentation (cardiac arrest, level of consciousness measured by Glasgow Coma Scale (GCS) score, respiratory rate, heart rate, blood pressure, temperature, serum lactate (hospital only)), clinical features of the poisoning episode (hyperthermia (≥39 °C), vomiting, headache, anxiety, hallucinations, agitation, psychosis, seizures, palpitations, chest pain, hypertension (≥180 mmHg), hypotension (≤90 mmHg), arrhythmias), and treatment (intubation, sedation, naloxone, flumazenil, any treatment beyond mere observation).

Classification of toxic agents was based on the clinical diagnosis made by the doctor treating the patient as entered in the patient records, in its turn based on all available information, i.e., information from the patient and companions, and on the clinical picture. The term “unspecified stimulant” was used in the patient records when the patient reported taking a stimulant but did not know which specific drug or when the clinical picture was consistent with a stimulant toxidrome and no information on specific drugs was available.

At the OUH, the clinical diagnosis was supplemented by toxicological analyses in urine samples in 124 (58.8%) cases. For gamma-hydroxybutyrate (GHB), gas chromatography–flame ionization detector (GC-FID) was used until 9 October 2014, when replaced by gas chromatography–mass spectrometry (GC–MS). For other agents, immunological screening was followed by liquid chromatography–tandem mass spectrometry (LC–MS/MS) confirmation.

We co-categorized amphetamine and methamphetamine as amphetamine. Z-drugs were categorized as benzodiazepines.

When converting continuous variables into categorical variables (tachypnoea (respiratory rate≥20/min), bradypnoea (respiratory rate<10/min), tachycardia (heart rate≥100/min), bradycardia (heart rate<50/min), and hyperlactataemia (serum lactate≥3.00 mmol/L)) missing values were treated as the absence of the relevant clinical feature. Commonly used clinical thresholds were used when defining the categorical variables.

For comparisons, we grouped the cases as follows: amphetamine, cocaine, MDMA, other stimulants (all other specified or unspecified stimulants), and multiple stimulants (all cases involving more than one stimulant drug).

Statistical analyses

Analyses were done in IBM SPSS version 26, 27, and 28. We described the data using proportions for categorical variables and medians and interquartile ranges for continuous variables. The chi-square test was used when comparing categorical variables. When chi-square test assumptions were not met as more than 20% of the cells had an expected value of less than 5, we used Fisher’s exact test instead. The Kruskal–Wallis test was used when comparing continuous variables. The level of significance was set at p<0.05.

Comparisons between drugs

Amphetamine was frequently combined with opioids. This explains why naloxone frequently was given in the amphetamine group and likely signifies that an opioid toxidrome often dominated the clinical picture of the combination. Most of these patients were probably injecting drugs, as an estimated 3000–4000 people in Oslo regularly do [13]. In a Norwegian study of people injecting drugs, 60% had injected amphetamine during the last 4 weeks, 43% had injected heroin, and 29% both [14]. In our study, many patients combining amphetamine and opioids had also taken benzodiazepines, signifying dangerous polydrug use in this group of patients [15, 16]. Most drug-induced deaths in Oslo are polydrug poisonings [17], and fatal overdoses from combined stimulant and opioid poisoning are on the rise in the USA [2].

Patients taking cocaine and MDMA were significantly younger than those taking amphetamine, probably reflecting the position of these drugs as party drugs or club drugs [18, 19]. This is consistent with findings across Europe [20]. Along the same lines, ethanol co-ingestion was most frequent among cocaine and MDMA patients, and they more frequently presented on weekends. GHB, often also viewed as a party drug [18, 21] and previously found to have a weekend presentation pattern in Oslo [22], was surprisingly often combined with amphetamine.

Taking multiple stimulants seemed to be a major risk factor for severe toxicity. The largest proportions needing intubation and sedation at the hospital were found in this group. However, the patients in the cocaine group stayed the longest in the hospital. Chest pain, tachycardia, and arrhythmias were mostly seen when cocaine was taken, in line with the established cardiotoxic effects of cocaine [23].

Psychosis is a known risk of amphetamine use [24, 25]. Although most of the patients presenting with psychosis had taken amphetamine, the largest proportion with psychosis and agitation was seen among patients taking unspecified stimulants. The unspecified stimulants may hide undiagnosed stimulant NPS inducing psychosis [26, 27], though it is also possible that these stimulants were registered as unspecified as the patients were too psychotic to give any specific information on the drugs taken.

Nearly all the hospital patients were admitted to intensive care. Hence, triage for hospital treatment seems to have been appropriately targeted by the ambulance service and the outpatient clinic. Concerning the related risk of under-triage, a previous study found that patients with substance use-related poisoning were safely managed in primary care by the procedure in use at the OAEOC [11].

Time trends

Compared to a study from 2012 at the OAEOC [22], there was a 17% increase in the number of amphetamine poisonings per year, while the number of cocaine and MDMA poisonings per year had doubled. A similar trend is seen in the USA, where both fatal and non-fatal overdoses from stimulant poisoning are increasing [2]. The increase in MDMA poisonings matches the increase in MDMA seized by Norwegian police since 2010 [3]. Surprisingly, no corresponding change in the seizure of cocaine has been seen during the last decade [3]. In wastewater analyses, gradually increasing amounts have been found since 2010 for all three drugs: amphetamine, cocaine, and MDMA [4]. Though we found an increasing number of cocaine poisonings, figures are still low compared to other ED settings in Europe [28].

Not applicable.