Study questions causal link between cannabis and brain changes

Lifetime cannabis use is associated with several changes in brain structure and function in later life, suggests an observational study, but these associations may not be causal, finds a genetic analysis of the same data, published in the open access journal BMJ Mental Health.

Some other unidentified factors may explain the differences found, say the researchers, who nevertheless emphasise that further research is needed to fully understand the effects of heavy use and cannabis potency on the brain.

Cannabis use has increased worldwide following its legalisation for medical and recreational purposes. But this has happened in the absence of a thorough understanding of its potential long term effects on the brain, say the researchers.

Its use is associated with deleterious effects on neurocognitive performance, brain structure, and function. And it's not known if there is a safe threshold, they add.

But previously published observational studies have not been able to confirm that cannabis caused the changes seen among regular users of the drug, and few of these studies have included older people.

In a bid to strengthen the evidence base, the researchers used Mendelian randomisation, a technique that uses genetic variants as proxies for a particular risk factor-;in this case cannabis use or dependence-;to obtain genetic evidence in support of a particular outcome-;in this study brain structure among older lifetime users.

They drew on 15,896 cannabis users participating in the UK Biobank for whom genetic profiling (2012-13) and brain imaging (MRI scans) (2014-19) data were available.

Participants reported if they had 'ever taken cannabis', with response options ranging from once or twice to more than 100 times. 

All participants who responded 'yes' were categorised as lifetime cannabis users (3641; average age 61), with 'no' responders categorised as the comparison group (12,225; average age 64).

Cannabis users were further divided into low frequency users, defined as lifetime cannabis use of up to 10 times, and high frequency users defined as lifetime use of 11–100+ times. Participants also reported how old they were when they last used the drug. 

The researchers accounted for a range of potentially influential factors, including: age at first scan (in years), sex, and age-by-sex interaction; level of deprivation; employment status; qualifications; smoking and alcohol consumption; blood pressure; weight (BMI); mental state; and 613 brain imaging-related variables.

After adjusting for these factors, cannabis use was associated with multiple measures of changes in brain structure and function. 

Participants who had ever used cannabis had poorer white matter integrity, a constituent of the brain that is important for cognitive function. And this was particularly evident in the corpus callosum, the main route of communication between the left and right sides of the brain. 

Cannabis users also had weaker resting state neuronal connectivity in brain regions which make up underlying the default mode network, thought to be active during mind wandering or daydreaming. 

These areas of the brain are densely packed with cannabinoid receptors, the researchers point out.

Neither the duration of cannabis abstinence nor the frequency of cannabis use were strongly associated with any of the observed findings in brain structure and function.

But there were some notable differences between the sexes: while significant associations were observed in six specific brain regions among the men, the observations seen in women were spread across 24 brain structures and functional regions. 

Most associations were observed in functional connectivity among the men; in women, associations were primarily seen in white matter integrity, suggesting that cannabis use affects the sexes differently, say the researchers.

When Mendelian randomisation was applied to the findings, no significant associations emerged between genetically predicted cannabis dependence/abuse or lifetime cannabis use.

"To the best of our knowledge, this is the largest observational study of relationships between cannabis use and brain structure and function to date, and the first Mendelian randomisation investigation," say the researchers. 

"Cannabis users had significant differences in brain structure and function, most markedly for markers of lower white matter microstructure integrity. Genetic analyses found no support for causal relationships underlying these observed associations," they add.

There are several possible explanations for the differences between the observational and Mendelian randomisation findings, they suggest.

An unmeasured variable, such as family history, dietary intake, or use of certain medications, might have influenced the observational findings. And the Mendelian randomisation analyses might have had less statistical power to detect small effects.

Mendelian randomisation also assesses the lifelong impact of cannabis use, while changes in observational studies might be due to factors occurring at different points in a person's life rather than cumulative impact across the life course, suggest the researchers.

They also acknowledge the limitations of using the UK Biobank, which is made up of predominantly healthy White participants. And few of those included in this study had cannabis use disorders.

The researchers were also unable to look at the time points during life that might be critical for cannabis effects, and the study relied on participants' recall concerning the amount or frequency of cannabis use during their lifetime, which may have been inaccurate.

They caution: "Our results need to be interpreted with careful consideration. Additional research is needed to understand the effects of heavy cannabis use in this population, including considerations of potency and related information to inform public policy."