Cannabis use disorder linked to ~60% higher risk of adverse cardiovascular events, Canadian study finds

By Tarun Sai LomteReviewed by Danielle Ellis, B.Sc.Oct 2 2023

A recent study published in Addiction evaluated the associations of cannabis use disorder (CUD) with cardiovascular disease (CVD) outcomes.

Background

Cannabis is used by over 200 million individuals worldwide; the adverse effects associated with cannabis have profound implications. CUD affects 27%-34% of cannabis users and has become a significant public health priority, given the absence of treatments and limited access to behavioral interventions.

Although research suggests adverse health outcomes due to cannabis usage, the relationship between cannabis and CVD is less explored. Current evidence suggests an increased rate of cardiovascular events among young cannabis users. Cannabis is also linked to severe events like stroke, myocardial infarction, arrhythmias, atherosclerosis, and cardiomyopathies.

About the study

In the present study, researchers assessed the associations between CUD and CVD outcomes using health administrative databases from Alberta, Canada. They used population-level data from the Discharge Abstract Database (DAD), National Ambulatory Care Reporting System (NACRS), Alberta Practitioner Claims Database, Pharmaceutical Information Network, and Alberta Provincial Population Registry from 2012 to 2019.

The NACRS, Practitioner Claims Database, and DAD were used to identify individuals with CUD. Each subject with CUD was age- and gender-matched to an unexposed control individual with no documented CUD diagnosis code. The index date of cases when they were first diagnosed with CUD was also assigned to the matched control.

The study’s primary outcome was incident CVD events (myocardial infarction, ischemic heart disease, unstable angina, peripheral vascular disease, cardiac dysrhythmias, ischemic stroke, or heart failure). Individuals with a prior history of CVD events were excluded. Each individual was followed up until an incident CVD event, death, or December 31, 2019.

The study covariates were the Charlson comorbidity index (CCI)), material (MDI) and social (SDI) deprivation indices, anatomical therapeutic chemical (ATC) subgroup count, mental health comorbidities, and healthcare utilization. The 2012 mid-year population of Alberta was used to estimate the point prevalence of CUD. Survival analysis was performed using the Kaplan-Meier method.

The log-rank test computed the associations between CUD and CVD, accounting for censoring by mortality. In stratified analyses, crude rate ratios (RRs) and 95% confidence intervals were estimated. The Mantel-Haenszel technique was used to pool estimates when stratum-specific estimates were similar, yielding adjusted RRs. Dose-response relationships were also evaluated.

Findings

The study included 59,528 individuals or 29,764 pairs. The overall prevalence of CUD was around 0.8%. The median CCI was zero in cases and controls. Cases and controls had identical MDI (4) and SDI (3) scores. CUD individuals were more likely to be in the most deprived quartiles and less likely to be in the least deprived quartiles.

The median number of prescriptions six months before the index date was two in cases and one in controls. CUD individuals had a median of five practitioner, emergency department (ED), and inpatient visits six months before the index date, while controls had only one visit. About 2.4% of cases and 1.5% of controls had at least one incident CVD event. The overall association between CUD and CVD outcomes was significant.

The strength of the association increased in a dose-dependent manner with increasing CUD severity (more CUD diagnosis codes). The RRs were 1.32, 2.47, and 2.67 for individuals with one, two to four, and five or more CUD diagnosis codes, respectively. CUD was significantly associated with a lower time to incident CVD events. Individuals with higher RRs were those without prescriptions, comorbidities, healthcare utilization in the past six months, and mental health comorbidities.

Conclusions

In summary, adults with CUD exhibited a 60% increased risk of an adverse CVD event than their age- and sex-matched controls. The findings suggest an elevated risk of CVD events among healthy subjects without medical/medication history if they have CUD. Notably, the study could not causally link the higher risk to CUD. Moreover, the researchers could not account for confounding by tobacco smoking due to the unreliability of the available data.

Further, the analysis did not include a direct measure of the amount of cannabis used. Instead, it relied on the number of CUD diagnosis codes as a proxy for CUD severity.

Overall, the study highlights the higher risk of CVD events in CUD individuals, despite not establishing a causal link, and that cannabis use may elevate CVD risks in otherwise healthy subjects. Therefore, it is crucial to educate patients about the potential risks associated with cannabis and CUD.