TOPLINE:

Cannabis use was associated with smaller volumes in the amygdala, and tobacco smoking was linked to smaller volumes in the amygdala, insula, and pallidum and reduced total grey matter volume (TGMV). A systematic review and meta-analysis of 103 studies found consistent evidence across cross-sectional, longitudinal, and Mendelian randomisation (MR) studies for tobacco-related TGMV loss.

METHODOLOGY:

• Researchers conducted a systematic review and meta-analysis of 103 independent studies examining associations between cannabis use, tobacco use, co-use, and brain volume.
• The meta-analysis included a total of 77 studies and 72,798 participants: 44 studies (18,247 participants) examined cannabis use cross-sectionally, 30 studies (51,194 participants) examined tobacco use cross-sectionally, and three studies (3357 participants) examined tobacco use longitudinally.
• The analysis included cross-sectional, longitudinal, and MR study designs to triangulate evidence across different methodological approaches with varying sources of bias.
• Outcome measures focused on the brain volume of global, cortical, and subcortical regions assessed using T1-weighted structural MRI, with 33 brain regions of interest analysed.
• The researchers extracted both adjusted and unadjusted estimates and utilised random-effects meta-analyses stratified by exposure and study design.

TAKEAWAY:

• The meta-analysis of adjusted cross-sectional estimates showed that people who used cannabis had smaller volumes in the amygdala than control individuals, with a small effect size (17 studies; P = .016).
• People who smoked tobacco had smaller volumes in the amygdala (five studies; P = .025), insula (five studies; P = .011), and pallidum (five studies; P ≤ .0001) and smaller TGMV (seven studies; P = .020) than control individuals; however, there was weak evidence for smaller volumes in the hippocampus in this group (10 studies; P = .049).
• Longitudinal analysis indicated a greater decrease in TGMV among people who smoked tobacco than among control individuals (five studies; P = .037).
• MR studies provided weak evidence that smoking initiation might decrease amygdala volumes (P = .046) and TGMV (P = .122 after adjusting) while demonstrating strong evidence that smoking more cigarettes per day might significantly decrease hippocampal volumes (P = 1.8E-06).

IN PRACTICE:

“We found cross-sectional evidence that people who use cannabis had smaller volumes in the amygdala. There were smaller volumes in the amygdala, insula and pallidum associated with tobacco use. There was consistent evidence for reductions in TGMV associated with smoking across cross-sectional, longitudinal and MR studies,” the authors wrote.

“This review highlights significant gaps in the literature, including a lack of studies using longitudinal and causal inference designs, as well as a lack of research on cannabis and tobacco co-use,” they added.

SOURCE:

This study was led by Katherine Sawyer, University of Bath, Bath, England. It was published online on March 19, 2026, in Addiction.

LIMITATIONS:

Most included studies were cross-sectional, which prevented definitive causal inferences about effects of cannabis and tobacco use on brain structure. Adjusted estimates varied significantly between individual studies; some adjusted only for intracranial volume, which introduced heterogeneity into the analysis. Using cortical volume as the primary structural measure may have been less sensitive to differences driven by cortical thickness or surface area. Not all relevant regions could be assessed because there were insufficient independent studies for meta-analysis in some regions in which previous reviews had found reductions.

DISCLOSURES:

Several authors reported receiving funding, grants, investigator grants, senior research fellowships, PhD studentships, and postdoctoral fellowship awards from several organisations including but not limited to the Medical Research Council, UK Research and Innovation, the UK government’s Horizon Europe, Wellcome, the European Research Council, Cancer Research UK, and Pfizer. One author declared having previous employment at a consultancy that provided support for pharma companies.