Abstract
Cannabis is one of the most widely used illicit drugs in New Zealand and world- wide. By 21 years old, 80% of New Zealanders will have tried it at least once, with 10% going on to develop a pattern of heavy use or dependence. Those with cannabis dependence oftentimes have poorer socio-economic and functional outcomes, and can experience disorganised thinking, hallucinations, and short-term memory loss. Studies have suggested short-term cannabis-related structural and functional changes in the brain, but less is known regarding the long-term effects of cannabis use. Understanding long term brain changes associated with cannabis use is of particular importance as societal attitudes towards the medical and recreational use of cannabis become increasingly tolerant.
In this thesis, I used magnetic resonance imaging (MRI) to investigate the long-term effects of cannabis exposure on brain structure, cerebral perfusion, cerebrovascular health, and functional connectivity.
Sixty-nine participants, aged 43, were recruited as a subset from the Christchurch Health and Development Study (CHDS) – a longitudinal study of people born in Christchurch, New Zealand, in 1977, followed from birth. The CHDS has detailed prospective self-report data on cannabis use in late adolescence and early adulthood, which was used to classify participants as past cannabis users (n=35) or non-using controls (n=34), matched for sex and tobacco use. Cannabis-exposed participants were defined as having a history of heavy cannabis use or dependence at one or more timepoints from age 15 onward. All participants underwent 3T MRI scanning; acquisitions included (1) T1-weighted structural imaging, (2) pseudo continuous arterial spin labelling (pCASL) perfusion imaging, (3) high angular resolution diffusion imaging (HARDI), and (4) resting state functional connectivity imaging.
All MRI modalities were analysed using ANCOVA models for comparisons between the past cannabis using and the non-using control groups. I first investigated subregions of the hippocampus and amygdala, selected a priori for being rich in CB1 cannabinoid receptors. This was then followed by exploratory whole brain analyses using non-parametric, voxel-wise cross-subject statistics.
I found significant focal volumetric loss in the hippocampal and amygdala subregions of past cannabis users relative to non-users. However, no further evidence suggested long-term cannabis use was associated with widespread differences in grey matter volume, cerebral perfusion, white matter microstructure, or connectivity of resting state networks.
These findings may suggest that any potential brain changes associated with cannabis use may be primarily transient and only detectable during periods of use; the brain may recover from any acute structural or functional insult with extended periods of discontinued use. Future directions include imaging the full CHDS cohort for a complete MRI dataset. This would allow more potentially related covariates, such as alcohol use and childhood adversities, to be considered.