1. No category

Effects of Cannabis Use on Gray Matter Volume and
Cognition in Patients with Multiple Sclerosis
Kristoffer Romero, Ph.D & Anthony Feinstein, M.D., Ph.D
Sunnybrook Health Sciences Centre, Toronto, Canada
[email protected]
[email protected]
ANALYSES
INTRODUCTION
• T1-weighted and FLAIR MRI images
• White matter lesions appearing hyperintense on FLAIR and hypointense on T1 images were filled
using the Lesion Segmentation Toolbox (Schmidt et al., 2012)
• Filled T1 scans were segmented using VM8 and non-linear DARTEL normalization
• T1 scans were also segmented using FreeSurfer, to obtain subcortical volumes for follow-up analyses
• Hippocampal volumes were differentially correlated
with verbal memory across groups (Zdiff = 1.73, p
<.05, 1-tailed)
• Conversely, thalamic volumes were similarly
associated with processing speed in both groups
80
Statistical analyses
• Behavioural partial least squares analysis (PLS) (Krishnan et al., 2011): data-driven, multivariate
technique that extracts latent variables that maximize the covariance between brain data and test
performance
• PLS determines whether a) there is an association between gray matter, white matter and
neuropsychological test scores, and b) whether this association differs across groups?
SRT Total Learning Score
• Cognitive impairment is common in multiple
sclerosis (MS), affecting 40 -60% of patients
• A subset of MS patients smoke cannabis for relief
of pain, muscle spasticity, or anxiety
• However, recent evidence suggests that cannabis
use has a deleterious effect on cognitive
functioning in MS patients (Honarmand et al.,
2011).
• Given the high prevalence of cognitive
impairment, and the effects of MS on gray
matter and white matter, MS patients who
smoke cannabis may show further declines in
cognition, which may be tied to structural
changes in gray and/or white matter
Pre-processing
ROI analysis (FreeSurfer)
RESULTS
PURPOSE
Gray matter – cognition correlations
70
60
50
30
20
Non-cannabis
10
Cannabis
0
3000
White matter – cognition correlations
SDMT Total Score
60
Patients
• MS patients who smoke cannabis (n = 20) and
patients who are cannabis-naïve (n = 19)
• All patients had a confirmed diagnosis of relapsingremitting MS according to McDonald et al (2001)
criteria.
• Patient groups matched for age, sex, level of
education, and disease characteristics (see Pavisian
et al., 2014)
5000
r = 0.48
r = 0.58
50
40
30
20
Non-cannabis
10
Cannabis
0
5000
6000
7000
8000
Left Thalamus Volume (mm3)
9000
CONCLUSIONS
Neuropsychological tests
• Patients screened for cannabis use via urine sample
and salivary assay (NarcoCheck) for THC metabolite
3500
4000
4500
Left Hippocampus Volume (mm3)
70
METHODS
Drug screening
r = -0.03
40
To determine the association between cannabis use,
cognitive functioning, and structural brain changes in
MS patients
• All patients were given the Brief Repeatable
Neuropsychological Battery :
1) Buschke Selective Reminding Test (SRT)
2) Rao 10/36 Spatial Recall Test (10/36)
3) Controlled Oral Word Association Test (COWAT)
4) Symbol Digit Modalities Test (SDMT)
5) Paced Auditory Serial Addition Test (PASAT) (2 s)
r = 0.52
PLS analysis
•
•
•
•
LV 1(33% variance): significant association between gray matter and cognitive test scores
LV 2 (17% variance): significant association between white matter and cognitive test scores
Cannabis group: Lower tissue volume correlated with processing speed and memory
Cannabis-naïve group: Lower tissue volume correlated only with processing speed
• MS patients who smoke cannabis showed a stronger
and more widespread association between brain
volume loss and cognitive performance, suggesting
cannabis has a moderating effect on the link
between brain tissue and cognition.
• Hippocampal gray and white matter were
particularly sensitive to this effect.
• One possibility is that cannabis-naïve MS patients
can recruit other regions to compensate for regional
volume loss, whereas cannabis-smoking patients
cannot mount a compensatory response.
REFERENCES
• Clinical variables:
• Cannabis-smoking MS patients showed poorer performance on tests of spatial memory (10/36) and
processing speed (PASAT) (See Pavisian et al., 2014)
•
•
•
•
Honarmand et al. (2012). Neurology; 76(13):1153-60.
Krishnan et al. (2011). Neuroimage, ;56(2):455-75.
Pavisian et al. (2014). Neurology , 82(21):1879-87
Schmidt et al. (2012). Neuroimage, 59(4):3774-83