Severe outcomes following pediatric cannabis intoxication: a prospective cohort study of an international toxicology surveillance registry

Abstract Introduction An increasing number of jurisdictions have legalized recreational cannabis for adult use. The subsequent availability and marketing of recreational cannabis has led to a parallel increase in rates and severity of pediatric cannabis intoxications. We explored predictors of severe outcomes in pediatric patients who presented to the emergency department with cannabis intoxication. Methods In this prospective cohort study, we collected data on all pediatric patients (<18 years) who presented with cannabis intoxication from August 2017 through June 2020 to participating sites in the Toxicology Investigators Consortium. In cases that involved polysubstance exposure, patients were included if cannabis was a significant contributing agent. The primary outcome was a composite severe outcome endpoint, defined as an intensive care unit admission or in-hospital death. Covariates included relevant sociodemographic and exposure characteristics. Results One hundred and thirty-eight pediatric patients (54% males, median age 14.0 years, interquartile range 3.7–16.0) presented to a participating emergency department with cannabis intoxication. Fifty-two patients (38%) were admitted to an intensive care unit, including one patient who died. In the multivariable logistic regression analysis, polysubstance ingestion (adjusted odds ratio = 16.3; 95% confidence interval: 4.6–58.3; P < 0.001)) and cannabis edibles ingestion (adjusted odds ratio = 5.5; 95% confidence interval: 1.9–15.9; P = 0.001) were strong independent predictors of severe outcome. In an age-stratified regression analysis, in children older than >10 years, only polysubstance abuse remained an independent predictor for the severe outcome (adjusted odds ratio 37.1; 95% confidence interval: 6.2–221.2; P < 0.001). As all children 10 years and younger ingested edibles, a dedicated multivariable analysis could not be performed (unadjusted odds ratio 3.3; 95% confidence interval: 1.6–6.7). Conclusions Severe outcomes occurred for different reasons and were largely associated with the patient’s age. Young children, all of whom were exposed to edibles, were at higher risk of severe outcomes. Teenagers with severe outcomes were frequently involved in polysubstance exposure, while psychosocial factors may have played a role.


Introduction
Cannabis is the most used recreational drug worldwide, with over 209 million users [1], and it is a commonly used psychoactive substance in the United States. As of June 2023, legal provisions allowing the recreational use of cannabis have been approved in Canada and Uruguay, as well as in other jurisdictions, including 22 states in the United States and the District of Columbia [2]. Cannabis has become a major industry in states where its recreational use is legal [3]. In Colorado, for example, the first recreational cannabis stores opened in 2014, and since then, cannabis sales have grown sharply to $1.1 billion in 2017 and doubled to $2.2 billion in 2021 [4], while medical cannabis sales have remained stable. Similarly, the legal recreational cannabis retail market in Ontario, Canada, has expanded rapidly since the legalization of cannabis in October 2018. The number of cannabis stores per 100,000 individuals aged 15 and greater years increased more than 100-fold in two years: from 0.13 in April 2019 to 14.0 in July 2021 [5].
Growing evidence suggests that legalization of recreational cannabis has been associated with a variety of public health concerns, such as the increased risk of involvement in motor vehicle collisions [6], increasing emergency department visits for cannabis intoxication and dependence in young adults [7] and higher rates of cannabis use during pregnancy [8]. The increasing availability of recreational cannabis in households following legalization has led to a parallel increase in unintentional pediatric intoxications [9].
While severe cannabis intoxication is infrequent among adults, it is relatively common in children [1,9], and may involve central nervous system (CNS) toxicity, respiratory depression and cardiotoxicity [1], including dysrhythmias and acute coronary syndrome [10]. Canadian studies conducted in the post-legalization era found that the introduction of commercial edible cannabis products was a key driver to increasing rates [11] and severity in intoxicated children [11,12].
With the continued legalization of cannabis across many jurisdictions despite mounting evidence of a subsequent increase in pediatric intoxications, we sought to identify predictors of severe outcomes, defined as intensive care unit admission or in-hospital death, following cannabis intoxication among children.

Study design
Employing the Toxicology Investigators Consortium (ToxIC) dedicated pediatric cannabis sub-registry, we conducted a prospective cohort study of all patients <18 years of age, who presented with cannabis intoxication to participating hospitals in the US, Canada and Israel from August 1, 2017, to June 30, 2020. In Canada, the use of medical cannabis was legalized in 2001. In October 2018, Canada legalized cannabis for recreational use in adults. In Israel, discussions regarding the legalization of cannabis for recreational use are ongoing. Inclusion criteria were exposure to cannabis, with signs and/or symptoms associated with cannabis intoxication, confirmed either clinically by the consulting medical toxicology service and/or by laboratory testing, and having a severe enough intoxication that has led to a formal bedside medical toxicology consultation. If polysubstance exposure was documented, such cases were included only if cannabis played the primary or substantial role in the clinical picture, as originally determined by the medical toxicologist who was consulted about the patient in real-time.

Data sources
The pediatric cannabis sub-registry was established in 2017 as a dedicated component of the ToxIC core registry. The ToxIC core registry was established in 2010 by the American College of Medical Toxicology and has been described in detail [13]. The Toxicology Investigators Consortium prospectively compiles de-identified data from 35 participating sites, comprised of >50 medical centers across the US, which represent approximately 60% of all medical toxicology training programs in the country, and also international sites in Canada and Israel. All prospectively identified patients who receive a consultation from a medical toxicologist are entered in real-time into the ToxIC registry. If the case involved pediatric cannabis exposure, additional detailed data are collected and entered into the dedicated pediatric cannabis sub-registry on a Research Electronic Data Capture (REDCap) platform. The latter collects patient-level clinical and demographic data, including circumstances of exposure, substances involved, signs and symptoms on presentation, management, disposition, and outcome. Toxicology Investigators Consortium data are regularly monitored and reviewed by data analysts for completeness with directed questions to the ToxIC site investigators to fix any omissions and inconsistencies.
The Toxicology Investigators Consortium functions under the approval of the Western Institutional Review Board, and all participating sites obtain study approval and a waiver of informed consent from their respective Institutional Review Board. The study adheres to the STROBE guideline for observational studies (eAppendix).

Study patients
The study cohort includes all pediatric patients (<18 years) who presented to the emergency department of a participating ToxIC hospital with cannabis intoxication within the study timeframe. The following demographic and clinical data were extracted from the patients' medical records into pre-defined fields on a secure REDCap data collection interface: age, sex, ethnicity, medical history, psychiatric history (unipolar and bipolar depression, anxiety, post-traumatic syndrome disorder), and social history (previous child protective services involvement, history of substance abuse, prior nonpharmacologic ingestion, high-risk sexual behavior, school failure). Relevant extracted historical data also included a reason for cannabis exposure (intentional or unintentional), administration route, and exposure location. Clinical data identified in the emergency department included vital signs (heart rate, blood pressure, oxygen saturation, temperature, Glasgow Coma Scale [GCS]), toxidrome, signs of respiratory depression or central nervous system (CNS) involvement, and signs of pulmonary involvement (defined as the presence of either acute respiratory distress syndrome, pneumonitis or aspiration). Ancillary data included electrocardiogram, chest X-ray findings, and laboratory results such as blood and urine drug screens. Interventions including cardiopulmonary resuscitation, intubation, bag valve mask ventilation, decontamination, antidote treatment, additional pharmacologic support, emergency department disposition (home, ward, intensive care unit), length of hospital stay, in-hospital death, and child protective services involvement after discharge were also collected.

Outcomes
The primary outcome was a composite "severe outcome" endpoint indicating high illness severity, defined as intensive care unit admission or in-hospital death. While various clinical factors may affect lower acuity disposition decisions, we focused our comparisons on this severe composite outcome, as intensive care unit admission typically represents a need for high-level medical care.

Candidate predictors of severe outcome
Based on a literature review [1,14,15] and the clinical experience of the investigators, we investigated potential a prioridefined risk factors for severe outcome following cannabis intoxication. The following potential demographic and clinical risk factors were examined: age, due to physiological differences suggesting that younger children may be at increased risk, exposure intent (intentional versus unintentional), as volition may affect outcomes, edibles ingestion because edibles have been linked with high tetrahydrocannabinol concentrations [1] co-exposure to other substances and medications (i.e., stimulants, sedative-hypnotics, anticholinergics/antihistamines, opioids, alcohol) due to potential synergistic toxicity [14,15], baseline chronic disease, which may represent an additional risk, and psychosocial history that was previously associated with increased severity [16]. We defined a positive psychosocial history as any documented mental health condition (e.g., major depression), previous drug intoxication, known polysubstance abuse, prior self-harm attempt, school failure, homelessness, reported high-risk sexual behavior, or child protective services involvement prior to the index hospital presentation.

Statistical analysis
Baseline characteristics were compared between patients who experienced severe outcomes versus those who did not. Chisquare or Fisher's exact test were used for the comparison of categorical variables, and a two-tailed t-test was employed for continuous variables. Categorical variables were described by percent proportions, and continuous variables were described as a mean, a standard deviation (SD), or median and interquartile range (IQR), as appropriate. For cells with counts less than five (e.g., death), data was suppressed to protect patient privacy.
We performed the analysis in two stages: first, a univariate analysis was performed to compare patients with severe outcomes versus those without as the binary dependent variable and the plausible candidate predictors of severe outcomes as independent variables. A two-tailed P < 0.05 value was considered statistically significant. Then, in the second stage, all parameters with a significance level of P < 0.2 in the univariate analysis and those considered clinically plausible, as explained above, were entered into the multivariable logistic regression analysis model to assess which unique risk factors independently predicted severe outcomes. All analyses were performed using SPSS Statistics, version 28 (SPSS Inc, Chicago, Illinois).

Exposure characteristics and outcomes
The demographics, intoxication, and clinical details of patients with vs. without severe outcome are summarized in Table 1. Overall, 60 (43%) patients were intoxicated by cannabis at home. Seventy-six (55%) children were intoxicated by cannabis intentionally, and 62 (45%) unintentionally. Self-harm intent was documented in 8 (6%) patients. One hundred and thirty (94%) children were exposed to recreational cannabis, while eight (6%) have been exposed to medical cannabis prescribed to a family member. Seventy-two (52%) patients were exposed to cannabis via ingestion, all of whom were exposed to cannabis edibles. Of them, 16 to baked goods, 23 to candies, 14 to other (tetrahydrocannabinol wax, liquid, and edible bar), and 19 to unknown cannabis edibles. Fifty-seven (40%) patients were exposed to cannabis via inhalation, 3 of them through vaping. Nine patients (7%) had an unknown mode of exposure.
Sixty patients (43%) reported prior cannabis use, 33 (24%) of them reported regular use (daily, weekly, or monthly). Laboratory blood/urine confirmatory testing was done in 111 (80%) patients, and in the remainder, exposure was clinically confirmed by a medical toxicologist at the bedside based on the obtained history, physical evidence if brought in, and physical examination. By the a priori study design, cannabis played a central role in all intoxications, including polysubstance exposures. Twenty-six (19%) children were exposed to at least one more substance (polysubstance exposure), mostly to sympathomimetics (n ¼ 9, 7%) and opioids (n ¼ 9, 7%). Overall, 75% of the patients had documented CNS involvement, and 10% had pulmonary manifestations. One hundred twenty-three (89%) pediatric patients were admitted to the hospital, with a median length of stay of 2 days (IQR 1-4 d). Twenty-three (16%) children were intubated and mechanically ventilated.

A comparison between children and adolescents
Children 10 years or younger had significantly higher rates of unintentional intoxication (P < 0.01), home location during exposure (P < 0.01), ingestion as the route of exposure (P < 0.001), CNS involvement (P < 0.01), hospitalization (P ¼ 0.01), and child protective services involvement (P < 0.01), than older children. Older children (aged >10 years) had significantly higher rates of polysubstance exposure. The common co-ingested substances were amphetamines, opioids, alcohol, sedative-hypnotics, and antidepressants. Children older than 10 years also reported more background psycho-social conditions compared to their younger counterparts ( Table 1).

Predictors of severe outcome
Fifty-two (38%) patients had severe outcome (intensive care unit admission including one in-hospital death), and 86 (62%) were either discharged home from the emergency department (n ¼ 15) or admitted to the ward (n ¼ 71). Their characteristics are presented in Table 2.

Severity analyses stratified by age group
To test the robustness of our findings, and based on the exposure differences between younger and older pediatric <0.01 a Acute respiratory distress syndrome, pneumonitis or aspiration. b Psychosocial risk factors: psychiatric history, unipolar and bipolar depression, anxiety, post-traumatic syndrome disorder, and social history; previous child protective services involvement, history of substance abuse, prior non-pharmacologic ingestion, high-risk sexual behavior, school failure. Some cases are represented in more than one category.
patients who present to the emergency department with cannabis intoxication depicted in Table 1, we replicated the univariable analyses, stratified by age groups ( 10 years versus >10 years).

Severity analysis of patients aged £ 10 years
All 60 patients aged 10 years were exposed to cannabis by ingestion. Of these, 32 (53%) had a severe outcome. Those with severe outcome had lower GCS at emergency department presentation and were more likely to undergo intubation compared to their peers without severe outcome (Table 4a).

Severity analysis of patients aged >10 years
Among 78 patients older than 10 years, 20 had severe outcome. Of these, 16 were intubated: 11 involved polysubstance intoxication and five were exposed to cannabis onlyfour were exposed by inhalation (three developed acute respiratory distress syndrome and one had seizures). The <0.01 a Acute respiratory distress syndrome, pneumonitis or aspiration. b Psychosocial risk factors: psychiatric history, unipolar and bipolar depression, anxiety, post-traumatic syndrome disorder, and social history; previous child protective services involvement, history of substance abuse, prior non-pharmacologic ingestion, high-risk sexual behavior, school failure. Some cases are represented in more than one category. .24 a Psychosocial risk factors: psychiatric history, unipolar and bipolar depression, anxiety, post-traumatic syndrome disorder, and social history; previous child protective services involvement, history of substance abuse, prior non-pharmacologic ingestion, high-risk sexual behavior, school failure. Some cases are represented in more than one category. Table 4a. Demographic, exposure, clinical and management characteristics of pediatric patients 10 years following cannabis intoxication with versus without severe outcome.
Non-severe clinical outcome (n ¼ 28) (%) Severe clinical outcome (n ¼ 32) (%) P value Age, years (mean ± SD) 3.8 ± 2. 0.59 a Acute respiratory distress syndrome, pneumonitis or aspiration. b Psychosocial risk factors: psychiatric history, unipolar and bipolar depression, anxiety, post-traumatic syndrome disorder, and social history; previous child protective services involvement, history of substance abuse, prior non-pharmacologic ingestion, high-risk sexual behavior, school failure. Some cases are represented in more than one category.
other intubated teenager with exclusively cannabis intoxication was exposed to edibles. Teenagers with severe outcome more likely involved in polysubstance ingestion, present to emergency departments with lower GCS scores, have CNS involvement, be intubated and received intravenous fluids compared to those without severe outcome (Table 4b).
A regression analysis was performed for patients older than 10 years, including age, sex, edibles ingestion, polysubstance intoxication, and psychosocial risk factors. Polysubstance abuse remained a significant independent predictor for severe outcome (adjusted OR 37.1; 95% CI 6.2-221.2; P < 0.001; Table  5). With respect to patients up to 10 years, as all of them ingested edibles, a similar dedicated regression analysis could not be performed.

Discussion
In this prospective multicenter cohort study of children who presented to the emergency department with cannabis intoxication we found that cannabis exposure in young children (all involved edibles) and polysubstance exposure in teenagers were powerful predictors of intensive care unit admission or in-hospital mortality. A notable strength of this study is prospective data collection by medical toxicologists caring at the bedside for all included patients [13].
Cannabis edibles are often highly concentrated and can lead to severe intoxication in children, manifested by delayed and prolonged effects, compared to cannabis inhalation [18]. When ingested, tetrahydrocannabinol, a potent psychoactive cannabinoid, is absorbed over 1-3 h, compared to almost instantaneous absorption by inhalation. This time lag may allow continued consumption of edibles by children before CNS and other symptoms appear. The predominance of CNS depression following cannabis intoxication in young children may also represent a greater weight-based CNS dose [19]. Several jurisdictions have implemented measures to mitigate pediatric exposures to cannabis edibles, such as restrictions of the amount of tetrahydrocannabinol in a package and limiting the attractiveness of product packaging [20]. However, such measures may not be sufficient to curb rises in pediatric intoxications, particularly from illegal products that do not comply with regulations.
All children aged 10 years and younger ingested cannabis edibles, and more than half of them were admitted to an intensive care unit. While previous research also reported higher rates of CNS depression and ventilatory support in young children compared to adolescents [1,20,21], our figure is significantly higher than the 10%-18% rates of intensive care unit admissions reported previously for this age group [1,9,20]. The Toxicology Investigators Consortium may capture more severe patients for whom frontline physicians requested medical toxicology consultation at the bedside. In addition, previous single-center studies may have underestimated the true burden associated with pediatric cannabis intoxications. Notably, there may be a general lower threshold of intensive care unit admission in young children. In Tables 4a and 4b, among the 32 patients 10 years who 2) 0.29 a Psychosocial risk factors: psychiatric history, unipolar and bipolar depression, anxiety, post-traumatic syndrome disorder, and social history; previous child protective services involvement, history of substance abuse, prior non-pharmacologic ingestion, high-risk sexual behavior, school failure. Some cases are represented in more than one category. <0.001 a Acute respiratory distress syndrome, pneumonitis or aspiration. b Psychosocial risk factors: psychiatric history, unipolar and bipolar depression, anxiety, post-traumatic syndrome disorder, and social history; previous child protective services involvement, history of substance abuse, prior non-pharmacologic ingestion, high-risk sexual behavior, school failure. Some cases are represented in more than one category.
were admitted to the intensive care unit, the mean GCS was 9.8, and 7 (22%) were intubated. In the adolescent group, 80% of the intensive care unit patients underwent intubation, and they had a lower mean GCS (7.9). Polysubstance exposure including cannabis was common in adolescents and was associated with markedly increased likelihood of severe outcome. A multi-center study found that most cannabis users used cannabis concomitantly with other drugs or alcohol [17]. Polysubstance use is often associated with greater consciousness impairment, need for higher level care (e.g., intensive care unit), and longer hospital stay than a single substance exposure. Further, a Canadian study of 22,484 patients with versus without polysubstance-related disorders (cannabis with another drug and/or alcohol-related disorder) reported a greater frequency of emergency department visits and hospitalizations in those with polysubstance use disorders compared to those with isolated cannabis use disorders [15]. Polysubstance use may also result in a mixed clinical presentation, which may present more complex diagnostic and management dilemmas than a single drug exposure. The common co-ingested substances in our cohort, such as alcohol, amfetamines, sedative-hypnotics, have likely contributed to CNS manifestations.
Several limitations merit mention. First, most hospitals affiliated with the ToxIC network are large referral centers. These characteristics may result, on average, in a population of patients presenting with more severe toxicity compared to patients presenting to community hospitals, in which medical toxicology bedside consultation service may not be available. In addition, some patients with cannabis exposure presenting to participating hospitals may not have been consulted by medical toxicology service and thus were not included in our data. The relatively small number of deaths compared to intensive care unit admissions did not permit a stratified sub-analysis. Intensive care unit admission decisions may be somewhat influenced by local practices. Finally, as all patients younger than 10 years were exposed to cannabis via edibles, age was dropped from the primary (unstratified) multivariate model in favor of keeping edibles. Hence, the adjusted odds ratio for edibles is very likely influenced by age in a way we could not measure.

Conclusion
Severe outcomes in pediatric patients who presented to emergency departments with cannabis intoxication occurred for different reasons and were largely associated with the patient's age. Young children (all of whom were exposed to edibles) were at higher risk of severe outcomes. Teenagers with severe outcomes were frequently involved in polysubstance exposures, while psychosocial factors may have played a role. Prevention efforts should target these risk factors to mitigate poor outcomes in pediatric patients with cannabis intoxication.