Long Term Effects of Marijuana Use on the Brain, Filbey, Francesca M. et al, 2014 PNAS 111:47
The title says it all—A lack of information regarding the long-term effects of marijuana use on the structure and function of the human brain was the driving force behind this study. In order to best elucidate the characteristics of the brain, the authors chose gray matter (GM) volume as their output variable, measured by structural MRI across the whole brain.
Functional connectivity MRI, and white matter integrity between abnormal GM regions was characterized by diffusional tensor imaging in a group containing 48 marijuana users and 62 age- and sex-matched non-using controls. Diffusional Tensor Imaging (DTI), and its related measurement, fractional anisotropy (FA), are a technique measuring diffusion rates along defined vectors in 3D space. Isotropic diffusion refers to free molecular movement in all directions; in the presence of membranes or other barriers, there will be a limit to molecular movement in a defined direction, causing anisotropy. This anisotropy is represented as an ellipsoid shape (a “tensor”) in 2D-space, for a defined plane within the structure (Fig.1). The map of all tensors in a given 2D plane offers insight into the structures present within that plane, and when put into the context of all neighboring planes will elucidate microstructures of the brain or other tissue in high 3-dimensional resolution. The degree of anisotropy offers an understanding of neural connectivity within the brain, and thus made DTI a logical choice for the design of this experiment, as the aim of the study was not to investigate changes of activity in the brain during the high associated with marijuana use.
Fig.1: Sample Diffusional Tensor Map- https://upload.wikimedia.org/wikipedia/commons/c/cc/DTI-axial-ellipsoids.jpg
Participants in the study were not given specific mixtures of lead compounds to be tested. Therefore, I cannot comment about any pharmaceutical implications of this article. I also cannot comment about the specific strains of marijuana consumed by the participants, or how often and how heavily they consumed it. Participants were included based on self-reporting of these factors, which is still at best difficult to control. In addition, the participants themselves probably don’t care what type of marijuana they consume from day to day over long periods of time. Though this may be a potential weakness of the paper, these factors are effectively impossible to control. One could envision a Nazi-esque regime in which defined groups of people are kept under strict supervision and ordered to smoke defined amounts of marijuana at certain times by specific methods over a period of years before a similar study—but such an experimental design is utterly ridiculous.
The design of the experiment was, in my opinion, sufficiently described. As I had expected, there would be terms and techniques I was previously unaware of, since my area of expertise is neither neuroscience nor MRI imaging. Following a few quick Google searches, I had a clear picture of what the authors were measuring, and why. In addition, the experimental design appears to meet the standards I was taught during my own education; a control group, and an experimental group were present, with sufficient numbers of members in each. Further, matching the control and experimental groups by age-sex stratification reduced the potential for confounding factors to influence the scientific integrity of the results.
The results showed a specific region of the brain to be responsive to the effects of marijuana use, namely, the bilateral orbitofrontal gyri. Voxelwise comparison of marijuana users’ brains versus control revealed a reduction in the GM volume of the right middle- and left superior- orbitofrontal gyri in marijuana users. This observation was complemented by the analysis of functional connectivity of the orbitofrontal network in control and user groups. This yielded a complex, and somewhat confusing result, because the initial response to marijuana use appears to be an increase in functional connectivity, as measured by FA, but over protracted periods of regular use, this connectivity decreases again. The forceps minor tract, a structure which connects the orbitofrontal regions, was also measured in both groups. This similarly revealed a higher degree of structural connectivity of the forceps minor in user groups than in the control. This was a conclusion drawn from measurements of fractional anisotropy (FA) in both groups (Table 2). Furthermore, it was revealed that the Radial and Axial Diffusivity (RD, AD) were significantly different between the two groups. The authors posit, the difference in FA between groups may have been driven by lower average RD, and suggests greater myelination in marijuana users. In addition, the authors comment that DTI imaging is sensitive to the relative amounts of water in the tissue; lower FA acts as a readout of higher rates of inflammation, suggesting that cannabis use leads to higher FA.
To conclude, the authors comment that chronic exposure to marijuana (i) reduces gray matter volume in the orbitofrontal cortex, (ii) increases structural and functional connectivity, and (iii) leads to complex neuroadaptive changes in the brain, which are modulated by age of onset and duration of use. Further experimentation is required to determine whether the changes caused by marijuana use are permanent, or able to revert to ‘normal’ after long periods of abstinence. It is necessary to focus future experiments on the nature of the complex changes in the neural response to marijuana, especially with regards to poor knowledge of the pathophysiology of marijuana use.
I believe the conclusions drawn by the authors are fair, but do not deliver the knock-out punch I was expecting. The effects of marijuana on the brain are indeed complex, but their functional relevance remains unclear—is there a negative impact on quality of life? As far as I am concerned, this is a good quality paper, but my opinion may be biased by the fact that I know it came from the journal of the PNAS. I cannot comment to the importance of the paper, because the results are, as in most biological contexts, to be reviewed with caution. Such results demand confirmation in subsequent experiments, and to really impact our understanding of marijuana’s long term effects, there is a need for a more telling technique to establish the behavioral nature of the observed changes in the brain and classify them as positive or negative. Still, this study presents interesting insights into the structural changes induced by marijuana use.
In answer to the core question—is marijuana a hallucinogenic, self-intoxicating abused drug which should not be used in medicine? Well, obviously this paper cannot be used to discuss the intoxicant or hallucinogenic properties of the drug, since that was neither addressed by the methodology, nor was it a stated goal of the article to characterize these properties. Still, there are those who would argue that any drug causing marked structural changes in the brain poses a potential health risk. However, without a clear readout of the consequences of the structural changes caused by marijuana use, I believe this paper cannot be used unequivocally to argue that there are positive or negative effects on the cannabis user. The patients using marijuana had significantly lower IQ, but this cannot be named specifically as a consequence marijuana use and furthermore, in my opinion, IQ is not an adequate measure of intelligence. In short—the jury is still out, and may be deliberating on this question for decades to come.
