Use of Medical Cannabis to Treat Traumatic Brain Injury

Abstract

There is not a single pharmacological agent with demonstrated therapeutic efficacy for traumatic brain injury (TBI). With recent legalization efforts and the growing popularity of medical cannabis, patients with TBI will inevitably consider medical cannabis as a treatment option. Pre-clinical TBI research suggests that cannabinoids have neuroprotective and psychotherapeutic properties. In contrast, recreational cannabis use has consistently shown to have detrimental effects. Our review identified a paucity of high-quality studies examining the beneficial and adverse effects of medical cannabis on TBI, with only a single phase III randomized control trial. However, observational studies demonstrate that TBI patients are using medical and recreational cannabis to treat their symptoms, highlighting inconsistencies between public policy, perception of potential efficacy, and the dearth of empirical evidence. We conclude that randomized controlled trials and prospective studies with appropriate control groups are necessary to fully understand the efficacy and potential adverse effects of medical cannabis for TBI.

Introduction

Traumatic brain injury (TBI) is a major public health problem, causing a significant number of hospitalizations, morbidity, mortality, and increased healthcare utilization and per capita costs attributable to disability in the United States each year.^1,2 Chronic symptomatology from TBI is associated with poor outcomes in vocational activities, interpersonal relationships, and independent living,^3,4 as well as a plethora of psychiatric and cognitive deficits.⁵ Importantly, despite over 30 phase III clinical trials, there is not a single pharmacological agent that has been demonstrated to have therapeutic efficacy for TBI,^6,7 potentially attributable to the heterogeneity of etiology, pathology, severity, and clinical manifestation of TBI. Thus, there is a great public need for interventions that are either neuroprotective or have the ability to reduce the symptom burden associated with TBI.

One relatively unexplored possibility is medical cannabis, which has gained recognition for its potential neuroprotective and psychotropic qualities in pre-clinical studies. The recent legalization efforts across much of the United States and Canada have increased accessibility, such that more patients with TBI may consider medical cannabis as a possible treatment. Though TBI is not a qualifying condition in the United States, a Canadian study found that 1.8% of medical cannabis users in their sample had indicated that brain injury was their primary qualifying condition.⁸ Observational and case studies also show that TBI patients are using cannabis to help mitigate their symptoms.^9–11 However, the efficacy of medical cannabis for neuroprotection or TBI-related symptoms, the potential effects on cognitive, psychiatric, and physical functioning, and alterations in underlying neural functioning remain poorly understood in patients with TBI.

The current review addresses the gap in the literature by providing the reader with the theoretical rationale for medical cannabis treatment for TBI, including the pathophysiology of TBI and pharmacology of cannabis. Given that human studies of medical cannabis and TBI are limited (five published studies, three published abstracts), a discussion is provided from supporting literature on the potential therapeutic and adverse effects of medical cannabis on TBI in the domains of neuroimaging and cognitive, psychiatric, and physical functioning. This is followed by a critical review of the literature on the effects of cannabis on TBI in humans, as well as current challenges and future directions for research.

Traumatic Brain Injury

There is currently no consensus on the definition or set criteria for the diagnosis of TBI; however, most define TBI as an injury resulting from an external force to the head followed by an alteration in brain function.¹² TBI can be clinically categorized as mild, moderate, or severe, which is typically based on additional criteria such as a post-injury Glasgow Coma Score (GCS), duration of loss of consciousness/amnesia, focal neurological signs, and post-concussive symptoms.^12,13

TBI is associated with a complex cascade of neurometabolic events. These include primary mechanisms of acute injury to brain tissue, such as contusion and diffuse axonal injury, abrupt changes in neuronal depolarization, release of excitatory neurotransmitters, disruption of ionic balance, changes in glucose metabolism, vascular trauma, altered cerebral blood flow/neurovascular coupling, and impaired axonal function.^14,15 Secondary mechanisms include increased intracranial pressure, cerebral hypoxia, inflammation, and disruption of the blood–brain barrier, all of which can occur up to several months after the injury.¹⁴ Thus, it is unlikely that there will ever be a single therapeutic agent that will be neuroprotective for the entire cascade of injuries that occur after trauma. The endocannabinoid system (ECS) plays a critical role in the support and protection of the central nervous system (CNS), as evidenced by increased endocannabinoid brain levels in response to injury,¹⁶ making the ECS a target for potential intervention. Moreover, as detailed in the sections below, not only does medical cannabis influence the biological cascade that follows an acute TBI, but it also has the potential to reduce several common psychiatric sequelae associated with TBI.^17,18

Pharmacology of Cannabis

Cannabinoids include plant-derived phytocannabinoids, synthetically produced cannabis, and endogenous cannabinoids such as anandamide (N-arachidonoylethanolamine; AEA) and 2-arachidonoylglycerol (2-AG).¹⁴ Cannabinoids act on G-protein-coupled cannabinoid receptors type 1 and type 2 (CB₁ and CB₂) and other receptors such as transient receptor potential cation channel subfamily V member 1 and G-protein-coupled receptor 55.¹⁶ CB₁ receptors are located in the CNS at high concentrations, particularly in the medial temporal lobe, striatum, and cingulate cortex.^14,17 CB₁ receptor binding mediates the psychoactive effects of cannabis by directly inhibiting the ongoing release of neurotransmitters, including gamma aminobutyric acid, glutamate, and acetylcholine, and indirectly affects opioid, dopamine, and serotonin signaling.^14,19,20

In contrast, CB₂ receptors are localized in the periphery and immune system, which may explain cannabinoids' potential impact on pain and inflammatory responses.^19,21 Δ⁹-tetrahydrocannabinol (THC) and cannabidiol (CBD) are the most commonly occurring phytocannabinoids in both recreational and medical cannabis.¹⁶ THC partially agonizes the CB₁ and CB₂ receptors, which produces cannabis's antinociceptive, antiemetic, anxiogenic, and cognitive blunting effects.²¹ CBD antagonizes the action of CB₁ and CB₂ receptor agonists²¹ and has a low affinity for CB₁ and CB₂.²² CBD results in increased anandamide levels and facilitates neurotransmission mediated by the serotonin receptor, 5-HT_1A.²³ CBD is associated with decreased levels of proinflammatory cytokines and increased levels of anti-inflammatory cytokines and is thus associated with neuroprotection.²³ It also has antioxidant properties.²³ CBD is not psychoactive,¹⁹ and some studies suggest that CBD may mitigate the undesirable effects of THC.^16,19,24 Thus, the treatment efficacy or side effects of medical cannabis may depend on the ratio of THC to CBD.^17,19 However, it is worth noting that the Cannabis sativa plant contains over 100 phytocannabinoids and that optimizing the therapeutic potential of medical cannabis will rely on a better understanding of the complex interaction among these compounds.²⁵

There are many routes of cannabis administration, each with their own pharmacokinetics that can influence potential beneficial and harmful effects.^26,27 Routes of administration are consistent among medical and recreational users, with inhalation by smoking and vaporization being the most common and producing the most rapid bioavailability.^27,28 Whereas vaporization may reduce the risk of respiratory symptoms associated with smoked cannabis, there is still concern that vaporizers can produce volatile compounds and cause lung injury.^28–30 Inhalation of cannabis concentrates (e.g., “dab,” “wax,” or “shatter”) is associated with adverse effects because of their high concentration of THC and is not advised for medical purposes.²⁶ Edibles, which include cannabis-infused food and drink, are more commonly used in U.S. states in which medical or recreational cannabis use is legalized.²⁷ Edibles eliminate the concerns for respiratory symptoms associated with inhalation methods and may have a longer duration of action than other methods.³¹ However, the peak therapeutic effects of edibles are also delayed because of slower bioavailability and different pharmacokinetics compared to inhaled cannabis.

THC is rapidly absorbed through the lungs, with peak plasma occurring within 10 min.³² After oral ingestion, systemic absorption is slower, with peak plasma occurring within 1–2 h of use.³² Further, whereas hepatic metabolism occurs for both inhaled and edible cannabis, the ingestion of cannabis is associated with a more extensive metabolism of THC.^32,33 In combination with potentially inaccurate product labeling, patients may be unfamiliar or have difficulty with appropriate dosing or titration of edibles, which can lead to overconsumption and overintoxication.³¹ Other routes include oromucosal and sublingual administration, such as tinctures and lozenges and topical, intravenous, and rectal treatments.²⁷ Tinctures and lozenges may have a more immediate onset of effects than edibles. However, research on the effects of these routes of administration is limited.^26,27

Pre-Clinical Studies Examining Treatment of Traumatic Brain Injury with Cannabinoids

Pre-clinical trials in rodent models have investigated several effects on TBI-induced pathophysiology with cannabinoid administration, including the prevention of neuronal cell death, activation of cell structure and remodeling processes, reduction of excitotoxicity, increased blood–brain barrier integrity, and reduction of neuroinflammation through increased anti-inflammatory cytokines and growth factors or reduced proinflammatory cytokines.^14,34 Selective targeting of CB₂ with agonists like JWH133 or GP1a improves the adverse outcomes of experimentally induced brain injury, including inflammation, edema formation, and neuronal cell death.³⁵ Exogenous administration of the CB receptor ligands, AEA and 2-AG, have also been shown to alleviate the neurometabolic cascade of events associated with TBI.^14,36,37 Similarly, the experimental administration of CBD prevented cell death, oxidative stress, inflammation, and excitotoxicity after hypoxic-ischemic brain injury in newborn swine.³⁸ The ECS also mediates several other physiological processes, making it a potential therapeutic target for different types of medical and psychiatric conditions after TBI. For example, one study showed that the oral administration of CBD oil to mice with mild TBI resulted in alleviation of behavioral alterations, such as pain behaviors, anxiety, aggressiveness, and depression.³⁹

Potential Therapeutic Effects of Medical Cannabis on Traumatic Brain Injury

TBI is heterogeneous in nature, manifesting in different clinical features depending on TBI severity, nature of the lesion(s), mechanism of the injury, underlying pathology, and pre-morbid factors.⁴⁰ Thus, there is likely to be a complex interaction between the various pathologies and clinical manifestations of TBI with cannabis treatment. Further, it is possible that cannabinoids may improve some symptoms of TBI, while simultaneously worsening or causing other symptoms. As demonstrated in our systematic review, there is a dearth of literature (only eight published studies) on the effects of cannabis on TBI in humans. The following sections therefore briefly review the extant literature of medical and recreational cannabis use broadly to hypothesize its potential beneficial and adverse effects on TBI.

Neuroimaging findings

Findings on neuroimaging are proxy measurements for underlying primary and secondary brain-related biological mechanisms associated with both TBI and cannabis use and therefore may elucidate how cannabis interacts with TBI in the CNS. Moderate and severe TBI is commonly associated with frank pathology on structural imaging and eventual atrophy.⁴¹ Notably, the medial temporal and frontal lobes are also most susceptible to trauma because of bony ridges in the cranial fossa of the skull.^42,43 Because these regions contain a high density of CB₁ receptors,^18,44 they may be particularly vulnerable to the interacting effects of TBI and cannabis.

Traditional structural imaging techniques typically show no abnormalities for persons with mild TBI,⁴¹ although more advanced imaging sequences frequently suggest other types of abnormalities such as reduced cerebral blood flow (CBF) or microstructural injuries.^12,45,46 Thus, more advanced imaging sequences that measure CBF or microstructural injuries may also be critical for consideration of the effects of medical cannabis on the entire spectrum of TBI. Studies across the full range of TBI severity typically have found local and global decreases in CBF and inverse neurovascular coupling.^47,48 However, increased CBF in the left dorsal anterior cingulate cortex is also shown to associate with worse post-concussive symptoms.⁴⁹ Reduced resting-state functional connectivity has been observed in the default mode, attention, and sensorimotor networks with increased connectivity within the ventromedial pre-frontal cortex, thalamus, and cerebellum.^50,51 During tasks requiring cognitive control and inhibition, persons with TBI show amplified activation of task-related networks (e.g., salience, executive) and brain structures (e.g., dorsolateral pre-frontal cortex, anterior cingulate) with increased deactivation of the default mode network.^52–54

Notably, there has been one study of medical cannabis's effects on functional magnetic resonance imaging that suggests medical cannabis may provide beneficial alterations in neural functioning. Gruber and collegues⁵⁵ conducted a 3-month longitudinal study in persons using medical cannabis for a variety of conditions (e.g., pain, psychiatric). The results suggested an improvement in executive function task performance with increased activation in the cingulate cortex and frontal regions relative to baseline, similar regions that impact TBI.⁵⁶ Although no control group was used in this study, this pattern of activation was similar to that of healthy controls from the authors' previous research, suggesting a potential therapeutic effect of medical cannabis. The authors suggested that besides potential effects from cannabis directly, other factors, including a reduction in conventional medications and improvement in psychiatric and physical symptoms, may have influenced the changes. However, as has been observed in the TBI literature, increased neural activity may also be the result of deleterious changes to the brain.

Studies on recreational cannabis use show alterations in CBF, and resting-state and task-evoked brain activations relative to healthy controls. Recreational cannabis is generally associated with increased CBF after acute exposure and decreased CBF with chronic use, which is particularly prominent in frontal regions.^18,57 Therefore, TBI and cannabis may have different effects on CBF. Recreational cannabis use is also associated with decreased resting-state connectivity in the default mode network, which typically includes medial and lateral parietal, medial pre-frontal, and medial and lateral temporal cortices.⁵⁸ Both recreational cannabis and experimental THC are associated with increased resting-state functional connectivity in the frontal lobe regions, precentral gyrus, superior frontal gyrus, posterior cingulate cortex, cerebellum, and sensorimotor areas.^59,60 During executive functioning tasks, recreational cannabis use has been consistently associated with decreased task activation in frontal lobe regions (e.g., dorsolateral pre-frontal cortex) and the anterior cingulate cortex.^61–63 However, other cannabis studies have found increased activity in the frontal cortex and occipitoparietal regions, with greater internetwork functional connectivity between the right frontal control network to the substantia nigra/subthalamic nucleus network.^64,65

Findings from neuroimaging studies suggest that both TBI and cannabis commonly affect pre-frontal and limbic neuronal circuitry,^66,67 providing the rationale for the potential synergistic effects on neural functioning. Whereas some effects on neuroimaging in both TBI and cannabis studies are associated with poorer task performances compared with healthy controls,^63,68,69 some studies, particularly ones that show increased activity relative to healthy controls, show no differences in performance in both TBI^52–54 and cannabis users.^64,65 This suggests the presence of compensatory neural mechanisms in response to TBI and cannabis use.

Cognitive effects

Cognitive changes are one of the most common sequelae after TBI.⁷⁰ Some of the frequent cognitive changes associated with TBI across the range of severity levels include executive dysfunction and decline in attention and processing speed.^5,41 For most persons with mild TBI, cognitive impairments typically diminish within ∼1 week to 90 days post-injury, with the ability to return to their usual activities within a year of the injury.^71–73 In moderate-to-severe injuries, global cognitive decline can be observed, including impaired verbal learning and metacognition, which may be attributable to the vulnerability of the ventral frontal and medial temporal lobes during closed TBI.^74,75 Therefore, potential cognitive effects of medical cannabis are especially a concern for patients who have compromised cognitive functioning related to TBI history.

Studies have been limited in examining the effects of medical cannabis on cognition. However, the few that exist suggest that cognition may remain stable or improve with use. One small pilot study (N = 11) tested participants before starting medical cannabis treatment for a variety of conditions and 3 months later. Participants reported a reduction in conventional pharmacological interventions use, reduction in depressive symptoms and impulsivity, and improved quality of life after 3 months of use. There was also improvement in performance on tasks of processing speed, set-shifting, and inhibition,⁷⁶ which may have also been attributed to improvements in psychiatric functioning. In a 3-month prospective study of cancer patients undergoing chemotherapy treatment who were either using medical cannabis (N = 17) or not (N = 17), medical cannabis users had more fatigue, pain, insomnia, and appetite loss and lower functional status at baseline compared to non-users. However, there were no differences between groups in performance on a cognitive screening measure and tests of psychomotor speed at baseline or after 3 months, suggesting that medical cannabis did not cause cognitive changes.⁷⁷

In contrast, recreational cannabis and experimental administration of cannabinoids to healthy controls have shown to adversely affect the domains of memory, attention, executive functioning, psychomotor speed, and motivation,^78–80 which overlap with the cognitive impairments observed in TBI.^5,41 Verbal learning and memory appear to be the most affected by cannabis use across studies⁷⁹ and is thought to be induced by THC rather than CBD.^81–83 Exposure to CBD before THC exposure or greater CBD to THC ratios may protect against the negative effect of THC on memory.^24,83,84 This result has been produced in an experimental study with healthy controls²⁴ and in naturalistic studies of recreational users.^82,83 Impairments in memory are thought to be associated with cannabis's effects on the hippocampus, a brain region that has a high density of CB₁ receptors and is essential for the formation of new memories.⁸⁵ Because of the overlap in potential cognitive effects, there may be challenges in discriminating between TBI-related cognitive impairment and the cognitive effects of medical cannabis, which can cause difficulties in tracking TBI recovery over time. Further, impaired cognition resulting from medical cannabis can potentially impact everyday functioning, such as driving,^86,87 which may already be impacted as a result of TBI.⁷⁴

Psychiatric effects

In the chronic stages of TBI, psychiatric comorbidities are among the most disabling sequelae, with long-term prevalence reaching 54%.^88–90 Psychiatric comorbidities in TBI are associated with increased cognitive, social, and functional impairment, as well as caregiver burden.^91,92 Psychiatric conditions can result from continued chronicity or exacerbation of pre-morbid psychiatric conditions or novel onset related to physiological or psychosocial consequences of the injury (e.g., employment changes).^12,93 Mesocorticolimbic and frontal-subcortical networks and/or their white matter connections may be directly (e.g., limbic cortex impacting upon bony protuberances)⁴³ or indirectly (e.g., secondary processes such as neuroinflammation)⁹⁴ affected by TBI, leading to organically induced neuropsychiatric syndromes.^56,95 These networks are also sensitive to the effects of cannabinoids,¹⁸ creating the potential for cannabinoids to modulate TBI-related psychiatric symptoms.

Across the range of TBI severity, depression is the most commonly diagnosed psychiatric condition^12,90 with rates of depression estimated up to 50% within the first year of injury.⁹² Anxiety disorders are also quite common, with an estimated prevalence of 1–28%.⁹¹ Post-traumatic stress disorder (PTSD) and TBI are highly comorbid, particularly in military personnel, with estimated pooled prevalence rates of up to 43.9%.⁹⁶ Pre-clinical research suggests that cannabinoids may have antidepressant effects as a result of direct (e.g., CB₁ receptor agonists) and indirect (e.g., endocannabinoid reuptake inhibitors, AEA degradation inhibitors) activation of CB₁ receptors. CB₁ receptor activation increases 5-HT neuron firing and reduces hypothalamus-pituitary-adrenal axis activity.⁹⁷ Anxiolytic effects of cannabis may be associated with the high density of CB₁ receptors found in the amygdala-hippocampal-cortico-striatal circuits, which contribute to anxiety, fear-related behavior, hyper-responsiveness to emotional stimuli, and retrieval of fear-based memories.⁹⁸

However, while PTSD is an approved condition for medical cannabis, depression, and anxiety are not in the United States.^99,100 It is therefore notable that mood and anxiety disorders rank among the most common conditions for which medical cannabis is used in the United States and Canada.¹⁰⁰ The perceived efficacy of medical cannabis for mood and anxiety disorders is high among medical cannabis users.^28,101 There is also evidence for self-medication without supervision from a physician for mood and anxiety disorders, particularly in locations in which medical cannabis is legal.^28,102,103 In open-label, self-report studies, patients report improvement of depression subsequent to medical cannabis use,^17,103,104 yet there is minimal evidence for cannabis's use for depression from prospective studies and randomized controlled trials (RCTs).^100,105 Similarly, only a few studies involving observation and patient report with open-labeled use exists showing the efficacy of cannabis for PTSD.^99,106–108 Improvement in anxiety symptoms has been demonstrated in RCTs; however, the trials typically consist of participants with medical conditions other than a primary anxiety disorder.¹⁰⁵ Therefore, public perception of cannabis' efficacy for mood and anxiety disorders is inconsistent with the limited scientific evidence for its efficacy.

Little is known about the adverse psychiatric effects associated with medical cannabis. In a systematic review of medical cannabis in neurological patient populations, the risk of serious adverse psychiatric effects, including mood and behavior changes, hallucinations, or suicidal ideation, was ∼1%.¹⁰⁹ Recreational use has been consistently associated with a risk of developing depression,¹¹⁰ anxiety disorders,¹¹¹ and worse outcomes for persons with PTSD.¹¹² The relationship between cannabis use and adverse psychiatric outcomes may be moderated by genetic factors.¹¹³ Reduction in recreational cannabis use has also been associated with improvement in psychiatric status.^114,115 The risk of psychosis after administration of THC has been well established in both observational and experimental studies with healthy volunteers.^116,117 Case studies in adolescents who sustained TBI show an increased risk of psychosis with cannabis use, which may be related to both TBI and cannabis's ability to affect CB₁ receptors.¹¹⁸ Cannabis has been associated with amotivation or apathy,⁷⁸ though the association is not as well established as with psychosis, with some studies showing association with apathy whereas others have not.¹¹⁹ This may be particularly important for persons with TBI using medical cannabis, given that apathy and amotivation are frequent (46–71%) after TBI.⁹⁰

TBI is also associated with comorbid substance use, with up to 50% of cases of TBI occurring under the influence of a substance.^12,120,121 Preliminary studies have shown that medical cannabis may improve outcomes for those in substance use treatment¹²² or play a role in reducing opioid withdrawal symptoms.⁸ However, the prevalence of a cannabis use disorder among medical cannabis users ranges from 2% to 38%, with up to 67% of medical users reporting moderate-to-severe withdrawal symptoms, which include sleep disturbance, anxiety, and irritability.^123,124 The prevalence of cannabis use disorder among medical users may be an underestimate, however, given that medical cannabis users perceive their use as beneficial rather than harmful and may deny its adverse impacts¹²⁴ or difficulty controlling use.¹⁰¹ Given the potential risk of dependency, patients with TBI using medical cannabis should be monitored for a comorbid cannabis use disorder given that this appears to be a risk for exacerbation of psychiatric symptoms.

Physical effects

Persons with TBI may seek medical cannabis to treat physical symptoms associated with their injury, such as pain or sleep disturbance.⁹ These physical symptoms can be related to the TBI directly or from other injuries sustained from the associated incident,¹²⁵ and have shown to be present when controlling for comorbid psychiatric/medical conditions as well as demographic characteristics.¹²⁶ Also noteworthy are the potentially hazardous physical side effects from medical cannabis use, including dizziness, vomiting, and fatigue, feelings of intoxication, respiratory issues associated with inhaled cannabis, and e-cigarette or vaping product use-associated lung injury.^19,109 These adverse physical effects may ultimately depend on the method, frequency, and quantity of use.

The prevalence of chronic pain in TBI has been estimated to be between 43% and 75% and is more commonly associated with mild rather than moderate or severe TBI.¹²⁷ Chronic pain is also one of the most common conditions for which persons use medical cannabis.⁸ Studies have shown increased substitution of medical cannabis for opioid medications, including in a sample of patients with TBI, suggesting medical cannabis's potential for harm reduction in persons with chronic pain.^8,128,129 Whereas some literature reviews have reported that there is good-quality evidence for the treatment of chronic and neuropathic pain with medical cannabis,^130,131 others are more cautious in this interpretation and call for more research.^132,133 A primary concern is a potential reduction in analgesic benefits associated with regular, heavy medical cannabis over time, leading to increased use.¹⁰⁰

Sleep disturbance is also common in TBI, with prevalence estimated to be ∼50%.¹³⁴ Relaxation is one of the most frequently reported effects of cannabis.¹³⁵ Therefore, sleep disturbance is unsurprisingly among the most commonly reported motivations for medical cannabis use.¹²³ The ECS has been hypothesized to contribute to the circadian sleep-wake cycle,¹³⁶ making medical cannabis a potential therapeutic agent for sleep. THC is associated with sedation, which may be related to increasing adenosine or inhibition of the arousal system.¹³⁷ CBD at low doses has stimulating effects, but has been associated with sedating effects at high doses.¹³⁶ In addition to the direct impact on the biological mechanisms associated with sleep, cannabis may help improve other symptoms, such as chronic pain, which can result in improved sleep.^138,139 Observational, open-label, and single-blind studies show that patients report improved sleep resulting from medical cannabis use.^138,139

Placebo-controlled studies examining cannabis's effects on sleep are limited, however.¹³⁹ One study found improved non-REM sleep during 8 days of treatment with dronabinol, a synthetic cannabinoid, in HIV-positive cannabis users.¹⁴⁰ However, another study found no effect on an objective measure of sleep time,^138,141 with small sample sizes (N = 7) in both studies. Cannabis may only have a short-term beneficial impact on sleep that diminishes over time attributable to increased tolerance or habituation to its sleep-related effects.^137,142,143 Cannabis's adverse effects on sleep are particularly prominent during withdrawal.¹⁴³ A study of medical cannabis users showed that sleep-related cannabis withdrawal symptoms were reported in 65% of the study participants.¹⁴² Similarly, chronic recreational users who discontinue cannabis experience increased sleep difficulties and adverse changes on polysomnography.¹³⁹

Human Studies of Medical Cannabis and Traumatic Brain Injury

Methods: Criteria for literature review

Despite the promising results from pre-clinical studies, human studies of medical cannabis and TBI have been limited. An exhaustive search of the literature was completed with consultation with an expert medical librarian. Importantly, a formal systematic or scoping review was attempted, but could not be completed because of to the paucity of literature. Articles were included if they met the following criteria: 1) peer-reviewed original research articles (i.e., no case studies or reviews), 2) published or e-published in English, and 3) focused on neuroprotection, treatment, or adverse effects of cannabis in humans with TBI.

The first search was conducted in PubMed (1946 to May 19, 2020) with the following Medical Subject Headings (MeSH): TBI terms: “Brain Injuries,” “Brain Injury, Chronic,” “Brain Injuries, Traumatic,” “Head Injuries, Penetrating,” “Brain Injuries, Diffuse,” OR “Brain Concussion.” Cannabis terms: “Cannabis,” “Medical Marijuana,” “Marijuana Smoking,” and “Cannabinoids.” Forty articles were retrieved. The abstracts were reviewed for each of these articles. One phase II clinical trial on TBI and medical cannabis was found through one of the articles in this search because it was a commentary on the clinical trial. Given the few articles that were found, we sought to broaden our criteria for search by using non-MeSH terms (cannabis AND Brain AND injur* [tw]). Eighty records were retrieved. The Latin American & Caribbean Health Sciences Literature Database (LILACS; 1982 to May 19, 2020) with the same search terms retrieved zero results. The following search terms were used in PsycINFO (1887 to May 19, 2020): (DE “Cannabis” OR DE “Hashish” OR DE “Marijuana”) AND (DE “Traumatic Brain Injury”), which retrieved nine records. The following terms were used to search EMBASE (1974 to May 19, 2020) and Medline (1946 to May 19, 2020): (“Brain injur*” OR “head injur*” OR concuss*) AND (“cannabis” OR “medical marijuana” OR “marijuana” OR “cannabinoids” OR “hashish”). Thirty-eight records were retrieved. See Figure 1 for Preferred Reporting Items for Systematic Reviews and Meta-Analyses Flow Chart (PRISMA).

FIG. 1.

PRISMA flow diagram detailing identification, screening, evaluation, and inclusion of articles in this review.

Three of the authors (D.H., V.S., and A.M.) independently evaluated each study according to the Strength of Recommendation Taxonomy (SORT) criteria for determining levels of evidence.¹⁴⁴ Level 1 studies indicate the highest quality of evidence, whereas level 3 studies are considered to be the lowest quality. Disagreements between the reviewers were reconciled with consensus. Published abstracts by McVige and colleagues^129,145 and Shetty and colleagues¹⁴⁶ were included in the review because of the paucity of published research articles. Because of this limitation, they were not given SORT ratings because of limited information.

Literature review results

A total of eight studies were included in this review, including four case-control studies, two RCTs, one prospective cohort study, and one focus-group survey-based study. Four of eight studies examined TBI outcomes related to medical cannabis, whereas the remaining studies focused on recreational use. The RCTs reported specific TBI diagnostic criteria (e.g., GCS 4-8, positive findings on computed tomography [CT]) and specified dosage information for the medical cannabis (e.g., Dexananbinol injection). One of the case-control studies used International Classification of Diseases codes related to TBI and positive toxicology screen of THC for participant inclusion criteria. However, the remaining studies were less specific with their TBI criteria and used questionnaires or medical record review to assess cannabis use information (see Table 1 for full diagnostic criteria for each study).

Table 1.

Articles that Examine Cannabis Use and TBI Outcomes

BI, Barthel Index; BP, blood pressure; CBD, cannabidiol; CNR, cannot rate; CPP, cerebral perfusion pressure; +CT, positive findings on computed tomography scan of the head; GCS, Glasgow Coma Scale; GOAT, Galveston Orientation and Amnesia Test; GOS, Glasgow Outcome Scale; GOSE, Glasgow Outcome Scale Extended; ICD, International Classification of Diseases; ICP, intracranial pressure; inj., injection; i.v., intravenous; mL, milliliters; PGI-I, Patient Global Impression of Improvement Scale; RCT, randomized controlled trial; SCAT5, Sport Concussion Assessment Tool 5: SCI, spinal cord injury; SF36, 36-Item Short Form Survey; SORT, Strength of Recommendation Taxonomy; TBI, traumatic brain injury; TBI-MC, traumatic brain injury–medical cannabis users; TBI-NMC, traumatic brain injury–non-medical cannabis users; THC, Δ⁹-tetrahydrocannabinol; TID, three times a day; QoLIBRI, quality of life after brain injury.

There have been only two RCTs testing the efficacy of medical cannabis for the treatment of TBI. There was a phase II RCT of dexanabinol (HU-211), a synthetic non-psychotropic analog of THC, which has been shown to have neuroprotective properties in animal models of TBI.^147,148 The phase II study demonstrated that dexanabinol was safe and well tolerated in a severe TBI patient group. Treatment effects included a reduction in intracranial pressure, cerebral perfusion pressure, and systolic blood pressure, with trending effects on neurological outcome measures, including the Glasgow Outcome Scale and Galveston Orientation and Amnesia Test.¹⁴⁸ In the follow-up, placebo-controlled phase III RCT, however, at 6 months post-injury, there was no difference in outcomes between placebo and intervention groups.¹⁴⁹

The next group of studies was largely uncontrolled and suggests that patients with TBI may use recreational or medical cannabis to treat their symptoms. A recent published abstract that involved a retrospective chart review comparing patients with mild TBI who were either using medical cannabis (N = 70) or not using medical cannabis (N = 33) found an association with improvement in sleep, headache, mood, and quality of life for the medical cannabis group. Of the medical cannabis users who also used opioid medications in this study, 38% decreased or discontinued opioid use.¹²⁹ An earlier abstract by the same group reported similar results.¹⁴⁵ These findings are consistent with other studies, which have shown that medical cannabis use is associated with less opioid use to treat pain and reduction of other medications for mood/anxiety, sleep, and migraines.^150,151 This suggests that lessening reliance on opioid medications may be a benefit of medical cannabis for TBI.

In a survey study of cannabis use among patients with moderate-to-severe TBI in Colorado, a U.S. state that has legalized recreational and medical cannabis, participants reported reasons for current cannabis use post-injury.⁹ The most popular reasons included recreational purposes (72%), stress/anxiety reduction (62%), and sleep improvement (55%), suggesting that patients with TBI may use recreational cannabis to help improve psychiatric and sleep symptoms. The patients also reported that the adverse effects of cannabis include decreased motivation, feeling “hazy/dull,” and fatigue.⁹ Notably, these symptoms overlap with common symptoms of TBI, causing concern for compounding effects.

The last group of studies provides some support for the neuroprotective hypothesis of cannabinoids for TBI. A recent study examining the effects of recreational cannabis, alcohol, and cigarette use on concussion within 2 weeks post-injury found that of the 307 participants in the study, 24.4% reported regular cannabis use 1 month before their injury and 14% reported use post-injury. At 3- and 4-week follow-up, acute post-injury cannabis use was associated with reduced symptom severity scores on the Sport Concussion Assessment Tool 5 (SCAT5) compared with non-users. Cannabis use, however, was not associated with physician-determined recovery, given that it did not impact additional markers of recovery such as changes in everyday functioning and exertion tolerance.¹⁵² In contrast, another abstract indicated that recreational cannabis use was associated with prolonged recovery after TBI.¹⁴⁶

Another study compared mortality rates of persons who had a positive toxicology screen for THC at the time of injury. Compared to those who had a negative toxicology screen, those who had a positive screen had a statistically lower mortality rate (2.4% vs. 11.5%).¹⁵³ Similar results have been found in a study of stroke patients where positive toxicology screen results for cannabis were associated with better outcomes on a measure of functional independence and lower intracerebral hemorrhage volume at discharge, even when controlling for demographic variables such as age and sex, intracranial hemorrhage volume and location, GCS, anticoagulant use, and intraventricular hemorrhage.¹⁵⁴ However, another study that examined mortality after TBI in the presence of positive toxicology for several substances, including alcohol, amphetamine, cocaine, and cannabis, also found a lower mortality rate for substance users generally.¹⁵⁵ There are several hypotheses for this finding. THC has a long terminal half-life attributable to its pharmacokinetics,¹⁴ so in polysubstance users, THC may be the only substance present in the body in active amounts. Alternatively, polysubstance use, rather than THC alone, may be a protective factor in acquired trauma. One possibility for this is that acute substance use may be associated with less severe TBIs attributable to their association with lower impact events, such as falls or violence.¹⁵⁶ See Table 1 for a summary of studies.

Discussion

Challenges and future directions

There are considerable challenges in both the research and clinical care of patients using medical cannabis. Medical cannabis is not carefully controlled or regulated, with studies showing discrepancies between product labels and actual active ingredient in the medical cannabis product.^157,158 Although some U.S. states require laboratory testing, one study found inconsistencies in interlaboratory report of cannabinoid content.¹⁵⁹ Because there are no standards for quantity or frequency of use per day, patients may have difficulty quantifying their use, even with traditional methods used to quantify substance use, such as the Timeline Followback procedure.¹⁶⁰ Thus, the quantity of active ingredients persons are using may be difficult to determine.

Additionally, those who are using cannabis for medical purposes may not be registered to use medical cannabis and may self-medicate.²⁸ For instance, in a study of cannabis users who were in a medically assisted treatment for opioid addiction, 60% of the participants indicated that they were using cannabis to manage pain, but only 5% were registered to use medical cannabis.¹⁶¹ Patients may also use additional products, such as hemp-derived products or CBD topical ointments, that they obtain from non-dispensary sources, such as supermarkets, that the patient does not self-disclose as a medical cannabis product. Therefore, healthcare providers may be unaware of their patients' use. Further, medical cannabis users have also reported negative social and economic aspects of medical cannabis, which include high cost and stigma of use from the public and their doctors.^128,150,162

Finally, much of what we know about the effects of cannabis is from the recreational/illicit cannabis use literature, with contrasting findings frequently reported after medical cannabis use. Specifically, as summarized above, there are reports of reduced psychiatric symptoms,¹⁷ pain,¹³¹ sleep disturbance,^138,139 and improved cognition⁷⁶ in the medical cannabis literature whereas research from the recreational literature suggests opposite and deleterious effects.^{18,78,115,139} It is critical to note that studies of recreational cannabis may or may not generalize to medical cannabis populations for several different reasons. First, medical cannabis users are more likely to use cannabis almost daily, be in poorer medical health, and are less likely to meet criteria for a substance use disorder or use illicit substances.^163–165 Second, harmful associations with recreational cannabis may also be associated with psychosocial factors related with its illicit nature, such as lower educational attainment, unemployment, and crime, rather than biological effects.^166,167 Third, THC potency has been increasing in recreational or illicit cannabis, whereas CBD potency has been decreasing.

In contrast, medical cannabis products may contain higher concentrations of CBD than recreational cannabis, which may mitigate the effects of THC,^76,168 although contrary evidence for this concentration difference also exists.¹⁶⁹ Additionally, for the few medical cannabis studies that do exist, they likely do not generalize to TBI populations. That is, effects will likely differ based on distinct pathophysiology of different disorders and how that overlaps with alterations in CB signaling or in how the brain responds to cannabis intake.

Extrapolating the potential effects of medical cannabis for TBI treatment primarily based on recreational use is problematic because of potentially different publication biases. Specifically, recreational studies tend to focus on harmful outcomes whereas medical studies often focus on beneficial outcomes. Moreover, the uncontrolled nature of the majority of medical cannabis studies do not account for potential placebo effects, and recruitment generally consists of convenience or self-selected samples obtained in the course of regular clinical care. To date, there are no nationally representative samples.¹⁰⁰ Other sources of bias in medical cannabis research include a lack of psychometrically sound measures of cannabis use and inclusion of limited or no control groups.¹⁰⁰

Further, only specific locations have legalized medical and/or recreational cannabis, which typically results in greater accessibility to a variety of products for only some persons.²⁷ The stigma associated with use may result in bias attributable to less reliable prevalence estimates and self-reported responses.¹⁷⁰ Finally, in general, published studies often have higher effect sizes than non-published studies¹⁷¹; therefore, both medical and recreational studies that show minimal effects (whether harmful or beneficial) may not be published. This is also true for pre-clinical studies, given that negative studies may be less likely to be published and may explain some of the issues with translating pre-clinical studies to humans. Notably, the current systematic review was limited to medical cannabis treatment for TBI rather than also covering pre-clinical findings as well as recreational use.

Obviously, RCTs and prospective trials for medical cannabis treatment of TBI in humans are needed. In addition to exploring which cannabinoids and ratios of cannabinoids are most efficacious for TBI, an understudied area is the differences among routes of administration. Different routes of administration have vastly different pharmacokinetics (e.g., inhalation vs. transdermal), and different pathologies may be more amenable to certain routes of administration. The lack of these types of studies will likely remain a challenge given cannabis's illicit history and current status as a Schedule I controlled substance. This means that cannabis is considered to have no acceptable medical use and has a high potential for abuse (see dea.gov/drug-scheduling). This may be changing, given that the U.S. Drug Enforcement Administration has recently proposed new regulations to facilitate clinical research of medical cannabis (see https://www.dea.gov/press-releases/2019/08/26/dea-announces-steps-necessary-improve-access-marijuana-research). However, even with new studies, results may not generalize to all potential medical cannabis users, given that often in clinical trials, persons with complex medical, psychiatric, and substance use histories are excluded.¹³²

Several general key outstanding questions remain. Clinical trials are needed to determine appropriate dosage and THC-to-CBD ratios, method, and duration of use to maximize potential benefits and minimize risks. These studies should include indices associated with the pathophysiology of TBI (e.g., changes in intracranial pressure, blood–brain barrier integrity) as outcome measures. It is also currently unknown how genetic factors associated with TBI prognostication or effects of cannabis can affect outcomes of patients with TBI using medical cannabis. Studies will also benefit from incorporating neuroimaging techniques, cognitive testing, and monitoring of psychiatric and physical symptoms, given that past research has demonstrated both positive and negative associations with cannabis in these domains. Although some studies of medical cannabis in non-TBI populations show some promising findings on neuroimaging, cognitive testing, and psychiatric and physical symptoms, there also is the concern that the large body of research on recreational cannabis shows detrimental effects across these domains. Thus, additional information is needed about what variables moderate these potential effects.

Additionally, little is known about symptom or performance validity in medical cannabis users (although, in mild TBI, symptom and performance validity is quite well understood).¹⁷² The potential for secondary gain may be present for some, such as easier access to a psychoactive substance or the desire to present with a qualifying medical condition to legitimize recreational use. This view must be balanced with the fact that many recreational users of cannabis may, in fact, be using it through a self-medication regime.^28,102,103 Overall, given that patients with TBI are using cannabis for medicinal purposes, there is an urgent need for more empirical research to better educate the public about its potential efficacy and/or harmful effects.

In conclusion, medical cannabis may represent a viable treatment for TBI, but the research is very much in its infancy. The lack of high-quality experimental studies makes it difficult to draw concrete conclusions about the positive and negative effects of medical cannabis on TBI. Pre-clinical studies of medical cannabis demonstrating its neuroprotective qualities have yet to translate into human studies, mainly because very few studies on the topic exist. This review has identified several key research directions in TBI in the topics of neuroimaging, cognition, psychiatric, and physical symptoms that need to be addressed.

Funding Information

This research was supported by grants from the National Institutes of Health (NIH; https://www.nih.gov; grant numbers NIH 01 R01 NS098494-01A1 and -03S1A1) to Andrew R. Mayer.

Author Disclosure Statement

No competing financial interests exist.