Effect of Cannabis on Sleep and Glymphatic Function in People with Multiple Sclerosis

Teddy Salan, PhD
University of Miami

Co-authors: Varan Govind¹, Sandra Diadhiou¹, Micheline McCarthy¹, Kottil Rammohan¹, Ramon Flores Gonzalez¹, Flavia Nelson^1
1University of Miami

Background: Sleep disruption is a common symptom affecting half of people with multiple sclerosis (PwMS). Consequently, PwMS are increasingly turning to cannabis to alleviate sleep difficulties. However, the effect of cannabis on sleep is not well established, evidence from the literature suggests that cannabis may help reduce sleep disturbances in PwMS, while other studies report no benefit. Furthermore, sleep regulates the brain’s glymphatic system (GS), a clearance system responsible for the removal of toxic waste products and molecules. Impaired GS function is also common in neurodegenerative disorders including MS, yet the impact of cannabis on GS remains understudied

Objectives: To evaluate the effect of cannabis on sleep, and its association to GS function in PwMS.

Methods: In this ongoing study, we enrolled 6 PwMS using cannabis (MS+CB+; 3M/3F; age= 41.6±12.9) and 7 PwMS not using cannabis (MS+CB−; 2M/5F; age= 36.8±8.5). Sleep data were collected using subjective (Pittsburgh Sleep Quality Index; PSQI) and objective measurements (total sleep time, time in different stages: light, deep, REM) obtained from Fitbits worn over 2 weeks. We also collected brain MRI data including diffusion tensor imaging (DTI) and processed using established methods to evaluate DTI along the perivascular space (DTI-ALPS) index, a marker of GS function. We performed t-tests to find group differences in sleep measurements and DTI-ALPS index, and spearman correlations to associate DTI-ALPS with sleep outcomes.

Results: While none of the sleep measures were statistically significant, MS+CB+ subjects recorded more time in deep sleep and less in light and REM sleep, and had more sleep records per day compared to MS+CB−. They also had higher DTI-ALPS index (1.73) compared to MS+CB− (1.61), although this was not statistically significant either (p=0.45). In addition, DTI-ALPS index negatively correlated with REM sleep for all subjects (p=0.007), with stronger correlation among MS+CB+ (p=0.007).

Conclusions: In this preliminary analysis, we found that MS+CB+ subjects spent more time in deep sleep rather than light or REM sleep stages. This may explain the higher DTI-ALPS index in MS+CB+ as the GS is primarily active during deep or slow-wave sleep. However, MS+CB+ subjects also had more sleep records per day indicating more fragmented sleep. In addition, we found that higher REM sleep time associated with lower ALPS, as more REM time may induce less time in deep sleep. Although the current sample size is underpowered, ongoing enrollment of 16 participants per group is expected to provide more robust effect size estimates.