Cannabis, cocaine and amphetamines linked to higher stroke risk

By Dr. Liji Thomas, MDReviewed by Lauren HardakerMar 11 2026

A massive analysis of more than 100 million participants reveals that several commonly used illicit substances may raise the risk of stroke, with genetic evidence strengthening the case that substance use disorders could play a causal role.

Study: Does illicit drug use increase stroke risk? A systematic review, meta-analyses, and Mendelian randomization analysis. Image credit: cunaplus/Shutterstock.com

Multiple epidemiological studies have demonstrated a link between stroke risk and the use of illicit substances, but the causality remains unknown. A recent paper published in the Journal of Stroke reported the results of a meta-analysis of current observational evidence coupled with Mendelian randomization to shed light on whether these associations may reflect underlying causal relationships.

Linking substance misuse to the global rise in stroke

Global stroke risk is rising, and the condition is now the third leading cause of death and disability worldwide. Many risk factors are modifiable, including substance misuse, which is also linked to cardiovascular disease.

Substance misuse, whether of legal or illicit drugs, is a major threat to public health because of its association with multiple adverse health outcomes.

Meta-analysis and genetic data explore substance–stroke link 

The researchers performed a systematic review and meta-analysis. The latter included 32 studies with a total of over 100 million participants; 14 cohort, 10 case-control, and 8 cross-sectional studies. These were drawn from administrative, hospital-patient, and population-based datasets.

The pooled odds ratios for cannabis, cocaine, amphetamine, and opioid use were estimated for both stroke subtypes: ischemic and hemorrhagic. Summary statistics from genome-wide association studies (GWAS) were then used for Mendelian randomization (MR) analysis.

MR uses genetic variants linked to a given exposure to test the causality of the observed association between the exposure and the outcomes of interest. Since genetic variants are not caused by the exposures or outcomes, but rather predate them, these associations are more likely to be causal. Seven proxy exposures were used here (genetic variants strongly associated with each of the seven listed below):

• substance use disorder (SUD)
• problematic alcohol use (PAU)
• alcohol use disorder (AUD)
• problematic opioid use (POU)
• cannabis use disorder (CUD)
• cocaine dependence (CD)
• nicotine dependence (ND)

The outcomes assessed in the study included GWAS summary statistics for the following stroke phenotypes: any stroke, any ischemic stroke, large artery stroke, cardioembolic stroke, small vessel stroke, and intracerebral hemorrhage. Similar techniques have been used to identify how the consumption of legal substances like alcohol and tobacco is linked to stroke, but genetic liability to several substance use disorders has not yet been assessed.

The study included only research conducted on people of European ancestry to avoid bias introduced by variation in population characteristics.

Study findings

Meta-analysis

Multiple illicit drugs were linked to a higher risk of stroke. Cannabis use was linked to an increase in stroke risk by about 37 % in pooled observational analyses, while the risk increased to nearly double or even higher with cocaine and amphetamines.

Cannabis use was associated with a 39 % increase in the risk of ischemic stroke and 16 % for any stroke, although the authors noted substantial heterogeneity and evidence of small-study effects across cannabis studies. Similarly, cocaine use was correlated with roughly double the risk of non-specific stroke, ischemic stroke, and hemorrhagic stroke, though variability between studies limits the precision of these estimates.

Amphetamine use was associated with markedly higher stroke risk, with pooled odds ratios of about 2.37 for ischemic stroke and 2.83 for hemorrhagic stroke. Whilst opioids did not show any association overall, except with recent use. However, these were based on a small sample and should be interpreted with caution.

In people under 55 years of age, cannabis, cocaine, and amphetamine use showed similar risk patterns. However, opioid use was unexpectedly associated with lower stroke risk in this subgroup, which the authors suggest likely reflects selection bias and small sample sizes rather than a protective effect.

Most stroke studies involving dependency or misuse of these substances were small and showed significant heterogeneity between them, limiting the validity of the observed associations.

MR analysis

MR showed that variants linked to SUD predicted a 33 % higher risk of any stroke, but a nearly eight-fold risk of intracerebral hemorrhage, reflecting genetic liability to broad substance use disorder rather than any single drug exposure.

Variants associated with cannabis use disorder increased the risk of any stroke by 11 %, but of large-artery stroke by 35 %. With cocaine dependence, MR showed an 8 % higher risk of cardioembolic stroke (blockage of a cerebral artery by a traveling clot) and a 38 % increase in the risk of intracerebral hemorrhage. MR analysis was not performed for amphetamines, as genetic variants linked to this disorder have not been identified yet.

Genetic liability to problematic alcohol use was associated with a roughly 50 % increase in cardioembolic stroke risk and about a doubling of large-artery stroke risk, while alcohol use disorder itself was linked to modestly increased risks of any stroke and cardioembolic stroke. Genetic liability to problematic opioid use was linked to a modest increase in any stroke and ischemic stroke risk. Nicotine was not associated with a change in risk.

Drug-related vascular effects may help explain stroke risk

The findings of this broad analysis consistently demonstrate an association between increased stroke risk and the use of several substances, particularly cocaine, cannabis, and amphetamines, while MR analyses suggest that genetic variants associated with increased liability to several substance misuse exposures are also linked to stroke subtypes.

Some drugs, like cannabis, were primarily associated with a higher risk of ischemic stroke. Several of these substances, including cannabis, cocaine, and amphetamine, produce hypertension and cerebral vasospasm, possibly via sympathetic activation. Cannabis also stimulates platelet aggregation.

Future genetic studies are essential to provide stronger causal evidence for amphetamine, given these biologically plausible mechanisms of risk.

Strengths and limitations

The meta-analysis is claimed to provide “the most comprehensive evaluation of multiple substance misuse types and stroke subtypes” so far, while the use of MR yielded supportive genetic evidence consistent with possible causal effects. However, most studies overall, and all cocaine studies, came from the USA, limiting the generalizability of the findings. Most studies were based on hospital records, introducing selection bias. Inconsistent exposure reporting limited the ability to adjust for residual confounding factors. Finally, only people of European origin were represented.

Substance use disorders emerge as overlooked stroke risk

This study indicates that several forms of substance misuse are associated with a higher stroke risk. Such exposures must be accounted for when evaluating the stroke risk. Moreover, “these findings suggest important public health implications for prevention strategies targeting substance use disorders to mitigate stroke risk.”

Download your PDF copy by clicking here.