Impact of prenatal cannabis exposure on offspring brain structure and function: a systematic review of MRI studies.

Aim

Cannabis use during pregnancy has increased, raising concerns about potential effects on offspring brain development. The objective of this systematic review was to evaluate MRI-based evidence on the association between prenatal cannabis exposure (PCE) and offspring brain development across structural, diffusion, resting-state functional, and task-based functional MRI.

Methods

Following PRISMA guidelines, PubMed, Web of Science, and Embase were searched through September 4, 2026, for MRI studies comparing prenatally exposed offspring with non-exposed controls across any modality. Study quality was assessed using Joanna Briggs Institute tools, and certainty of evidence was evaluated using the GRADE framework.

Results

Eighteen studies (2004-2026) met inclusion criteria, with participants ranging from the fetal period to young adulthood. Structural MRI showed no consistent global brain-volume reductions, although some studies reported region-specific differences, particularly in frontal regions and intracranial-volume trajectories. Diffusion MRI identified alterations in selected white-matter pathways, including frontolimbic and corticostriatal tracts, but findings were inconsistent and not uniformly replicated. Resting-state functional MRI studies showed heterogeneous connectivity differences across hippocampal, salience, thalamic, and auditory networks, and task-based functional MRI studies reported variable reward- and executive-control activation, including some null findings. Overall certainty of evidence was low to very low because of risk of bias, inconsistency, imprecision, residual confounding, and methodological heterogeneity.

Interpretation

PCE may be associated with subtle, regionally specific differences in brain structure, microstructure, and function rather than consistent global abnormalities. However, structural MRI showed no consistent global abnormalities, diffusion MRI findings were inconsistent, and resting-state and task-based fMRI findings were heterogeneous. Causal interpretation remains limited by confounding, cohort overlap, exposure assessment, and methodological heterogeneity.