Multiple Sclerosis, Cannabinoids, and Cognition

Abstract

Back to table of contents Previous article Next article SPECIALFull AccessMultiple Sclerosis, Cannabinoids, and CognitionPanagiotis Papathanasopoulos M.D., Ph.D.Lambros Messinis Ph.D.Epameinondas Lyros M.D.Andreas Kastellakis Ph.D.George Panagis Ph.D.Panagiotis Papathanasopoulos M.D., Ph.D.Search for more papers by this authorLambros Messinis Ph.D.Search for more papers by this authorEpameinondas Lyros M.D.Search for more papers by this authorAndreas Kastellakis Ph.D.Search for more papers by this authorGeorge Panagis Ph.D.Search for more papers by this authorPublished Online:1 Jan 2008AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinked InEmail A lthough cannabis has been used for various medical purposes over centuries, 1 it is currently not being officially prescribed due to law restrictions and lack of solid evidence regarding its objective clinical effects and safety. Nevertheless, there has recently been a growing interest about the potential therapeutic value of cannabis in several medical conditions. 1 – 4 Beneficial effects of cannabis in multiple sclerosis have focused much of the attention of scientists in the past few years. 2 , 4 Illegal self-medication with cannabis both in smoked and in oral forms has, however, been reported for years. 5 Anecdotal reports suggest that cannabis can relieve not only muscle spasticity and pain but, in some cases, can also improve bladder control. Many randomized placebo-controlled clinical trials have attempted to investigate the potential efficacy of various cannabinoid treatments in managing spasticity, pain, bladder problems, sleep disturbances, and tremor due to multiple sclerosis. 4 , 6 – 16 Results of clinical trials have been mixed, but together with insights from basic research and animal models of multiple sclerosis, they provide reasonable evidence for the potential therapeutic use of cannabinoids in the treatment of multiple sclerosis related symptoms. 1 , 17 Findings of biological experiments have also revealed the possible immunomodulatory and neuroprotective properties of cannabinoids, 4 , 18 , 19 thus further supporting their potential utility in multiple sclerosis therapeutics. However, many issues still remain unresolved. 20 Accordingly, synthetic cannabis-based preparations [dronabinol (Marinol®), nabilone (Cesamet®), Δ 9 –tetrahydrocannabinol + cannabidiol (Sativex®)] have been used in the U.S., Canada and some other countries as an authorized treatment for nausea and vomiting in cancer chemotherapy, appetite loss in AIDS wasting syndromes and symptomatic relief of neuropathic pain in multiple sclerosis. 3 , 21 Recent research developments regarding the therapeutic use of cannabis analogues in several medical conditions and the concurrent possibility that cannabis might be fully legalized for patients with severe and chronic medical conditions such as multiple sclerosis provided the motive for this article. If cannabis products were to be used legally in the future for therapeutic purposes, this would naturally raise the concern of possible adverse effects associated with long-term use and especially with respect to the central nervous system (CNS) and cognition. It has been suggested that these effects may be greater in vulnerable groups such as patients suffering from diseases of the CNS. Non-acute cognitive dysfunction resulting from accumulative cannabis exposure has been a matter of investigation mainly with recreational cannabis users. 22 – 24 Psychiatric effects of long-term exposure to cannabis, such as comorbid depression and predisposal to psychosis, are issues that have also been reported in the literature. 25 – 27 Nevertheless, the focus of this article is the potential adverse effects of long-term exposure to cannabis-based medicinal extracts on cognition. The article provides a review of the current literature on this issue and critically considers the potential that cognitive deficits attributed to long-term heavy recreational exposure might be extended to controlled pharmaceutical use in multiple sclerosis patients. We performed a search of the PubMed database for articles dated from 1986 to March 2007 in order to locate studies of cannabinoid use in multiple sclerosis, cannabis and cognitive dysfunction, long-term effects of exposure to cannabis on cognition, and neuropsychological/neuropsychiatric aspects of multiple sclerosis. Key words used in the search strategy were: “multiple sclerosis,” “cannabinoids,” “cognition,” combined with the terms cannabis-based medicinal extracts, tetrahydrocannabinol (THC), marijuana, neuropsychology and neuropsychiatry. These searches were limited to English-language articles and augmented by bibliographical searches of the obtained references. Further, we hand searched the following journals dated from 2003 to March 2007: Archives of Clinical Neuropsychology, Journal of the International Neuropsychological Society, Neurology, and The Clinical Neuropsychologist . We included only relevant articles published in peer-reviewed journals. Neurobiology and Pharmacology of Cannabis and the Endocannabinoid System The main psychoactive component of cannabis is Δ 9 –tetrahydrocannabinol (Δ 9 –THC). The drug is most commonly delivered by smoking a plant-derived cigarette. Acute effects of cannabis have been well recognized, including euphoria and relaxation, intoxication, short-term memory impairment, disruption of psychomotor control, poor executive functions, distorted sense of time, increased appetite, and analgesia. 3 Acute panic, paranoia, and psychosis have been reported in some subjects. 26 , 28 Although some researchers claim that marijuana is not particularly addictive, there is evidence that heavy and chronic users are likely to develop tolerance and dependence on the drug. 3 , 29 – 30 On the basis of current research, cannabis does not appear to induce a clear withdrawal syndrome, as do other drugs of abuse, such as opiates. 31 However, there is no doubt that individuals suffer unpleasant effects when abstaining from chronic cannabis use. 31 Thus, although addictive behaviors such as compulsive drug seeking due to craving are rarely induced by marijuana use, preparations containing higher Δ 9 –THC concentrations, such as hashish, have been shown to induce addictive behaviors, especially in populations at risk. 32 Furthermore, some users develop problems related to the drug, and many request professional assistance in limiting their consumption. 33 Indeed, cannabis dependence is a DSM-IV diagnostic category. However, it has been proposed that dependence on cannabis should deal only with psychological dependence. 34 Whether a certain individual will eventually succumb to the addictive potential of cannabis will obviously be the outcome of a combination of factors. Thus, it has been suggested that the dependence syndrome on cannabis via international classification systems (e.g., DSM-IV) should be revised. 34 Recent investigations have shed light on the mechanisms of the pharmacological actions of Δ 9 –THC and other chemically related molecules known as cannabinoids. Exogenous cannabinoids include phytocannabinoids such as Δ 9 –THC, cannabinol, cannabidiol, and other cannabinoid compounds present in extracts of the plant Cannabis sativa , as well as synthetic cannabinoids produced in the laboratory. Two endogenous cannabinoid receptors (CB1 and CB2) have been identified 35 , 36 although additional cannabinoid receptors may exist but have not yet been cloned. 37 Furthermore, endogenous ligands for these receptors have been isolated. These endogenous cannabinoids comprise a series of arachidonic acid derivatives such as anandamide, 2-arachidonoylglycerol (2-AG), 2-arachidonoylglycerol ether, virodhamine and N -arachidonoyldopamine which are referred to as “endocannabinoids” (for a review, see Mechoulam et al.). 38 Δ 9 –THC and other synthetic analogues exert their effects by acting as agonists at the cannabinoids (CB1 and CB2) receptors. However, some of the other constituents of cannabis, such as cannabidiol, which have well-documented biological effects of potential therapeutic interest, such as antianxiety and anticonvulsive properties, do not significantly interact with CB1 or CB2 receptors. This fact possibly explains the reason why cannabidiol does not possess psychotropic actions. Its effects have been largely attributed to inhibition of anandamide degradation or its antioxidant properties. 39 CB1 receptors are thoroughly expressed in the CNS, most densely in the frontal cortex, basal ganglia, cerebellum, hypothalamus, anterior cingulate cortex, and hippocampus, and are localized to axons and nerve terminals while being absent from the neuronal soma and dendrites. 3 In consistency with their presynaptic localization, the activation of CB1 receptors is thought to modify synaptic function by inhibiting the release of various neurotransmitters (L-glutamate, GABA, dopamine, 5-HT, and acetylcholine). 40 CB2 receptors, in contrast, are mostly found in peripheral tissues, predominantly in the immune system. 3 Interestingly, very recent evidence suggests that the CB2 receptors are expressed in the mammalian brain and may influence behavior. 41 The endocannabinoid system controls signaling processes between neurons in a retrograde manner. Endocannabinoids are not stored but are rapidly generated by postsynaptic neurons in response to Ca 2+ influx resulting from depolarization induced opening of voltage-controlled Ca 2+ channels. 42 When endocannabinoids are released from the postsynaptic membrane, they act backward across the synapse, tonically activating presynaptic CB1 receptors to decrease release of either inhibitory or excitatory transmitters. 40 Following an assumed cellular uptake mechanism, endocannabinoids are then degraded by intracellular enzymes. Fatty acid amide hydrolase (FAAH) and a monoacylglycerol (MAG) lipase are the enzymes proposed to catalyze the hydrolysis of endocannabinoids. 4 Endocannabinoids have been implicated in a variety of physiological functions. These areas of central activity include pain reduction, motor regulation, learning/memory, and emotion. Interestingly, recent evidence has emerged that tissue concentration of endocannabinoids and/or cannabinoid receptor density-activity changes (increasing or decreasing depending on the specific disease or pathological state) in a range of disorders. 43The Neuroprotective Role of Cannabinoids Endocannabinoids might serve several functions in the brain. Of great interest is the potential neuroprotective role in the CNS. Experimental data indicate that endocannabinoids are released and accumulated in response to many different types of toxic insult, such as excitotoxicity, excitotoxic stress, traumatic injury and ischemia possibly representing a compensatory repair mechanism. 44 – 47 The excitotoxicity hypothesis is used to explain the common biochemical basis behind many acute and chronic neurodegenerative disorders, including stroke, brain injury, amyotrophic lateral sclerosis, multiple sclerosis, Parkinson’s disease, and Huntington’s disease. 48 , 49 Thus, the cannabinoid system can serve to protect the brain against excitotoxicity and neurodegeneration and may play a primary role in limiting brain damage. Researchers’ current hypothesis of how cannabinoids provide their protective effects is perhaps that their general dampening of neural activity reduces excitotoxicity. 50 Cannabinergic activity through CB1 receptors is linked to signaling pathways that are protective against pathogenic insults and promote cell survival and maintenance. Such pathways include the activation of mitogen-activated protein kinases (MAPKs), such as the extracellular signal-regulated kinase. Activation of CB1 receptors has been shown to block presynaptic release of glutamate. 51 Interestingly, endocannabinoid levels were increased in a mouse model of multiple sclerosis, 52 although in a more recent study brain concentrations of endocannabinoids were not raised despite the cell damage induced by experimental allergic encephalomyelitis (EAE) in mice. 53 The disruption of the neuroprotective effect of endocannabinoids was more pronounced in purinergic P2×7 receptor knockout mice after induction of EAE. The authors showed that in this mouse model of multiple sclerosis the limited production of endocannabinoids could be attributed to an interferon gamma-mediated disruption of the function of P2×7 receptors, the activation of which leads to increased production of endocannabinoids by microglia. Collectively, these results strengthen the concept that the endogenous cannabinoid system may serve to establish a defense system for the brain. This system may be functional in several neurodegenerative disorders as well as in acute neuronal damage. Residual Neurocognitive Effects of Cannabis Use Although the acute effects of cannabis use have been well established, the question of whether chronic cannabis use causes residual neurocognitive deficits has produced conflicting data. Despite intensive research efforts in this regard, the literature remains divided and controversial, as no hard findings or consistent results have emerged. The cognitive deficits observed in chronic cannabis users may be temporary and reversible after a period of abstinence (attributable to late intoxication or withdrawal effects) or permanent (due to neurotoxicity). This question apart from its theoretical interest also harbors practical considerations regarding former drug user’s functional capacity after cessation of cannabis use. Even subtle cognitive impairments may prove a serious impediment especially for people whose occupation calls for intact agility and creative thinking. As far as current users are concerned, it is important to examine whether and how cognition is afflicted within unintoxicated intervals, in order to realistically assess functional operations in activities of daily living. Furthermore, it is essential to determine which variables of cannabis consumption (dosage, frequency, duration, or age at onset) independently correlate with cognitive deficits in cannabis users. The drug delivery is perhaps a critical parameter as well, since speculated neurotoxicity of cannabis may be dependent on the pharmacokinetic profile of the drug entering the organism. The smoked route is subject to wide variation. 54 Orally or sublingually delivered cannabis used in modern clinical trials may not provide standard dose absorption; however, the concentration curve of the drug is not as sharp as with smoked cannabis. 54 Intravenous administration retains control of dose but requires strict laboratory conditions. 54 Evidence of cognitive impairment in chronic cannabis users is available in the literature, mainly concerning memory, attention, and executive functions, but it is not clear how long such impairments persist after cannabis use is terminated or whether these deficits are accompanied by drug-induced neuropathology. Studies of nonacute effects of cannabis on brain function require a period of abstinence to eliminate the acute pharmacological actions of the drug. 55 Another general issue affecting most studies on the nonacute effects of cannabis exposure is that the premorbid neurocognitive abilities of these participants are largely unknown; some of the cognitive deficits noted possibly reflect preexisting cognitive impairments rather than consequences of cannabis exposure. 22 , 24 , 55 Potential neurocognitive effects of cannabis withdrawal syndrome might also need to be considered. 29 Pope et al. reported cognitive deficits in current heavy cannabis users after at least 7 days of abstinence. These, however, did not persist after 28 days of drug cessation and were related to recent, not cumulative, drug use. 55 On the contrary, Bolla et al. reported a dose-related effect of cannabis use on neurocognitive performance even after 28 days of abstinence. 56 Moreover, Solowij et al. have demonstrated cognitive impairments in long- versus short-term cannabis users after a median 17-hour abstinence period, indicating an important influence of duration of cannabis consumption on cognitive measures, predominantly verbal learning and memory. 57 In line with previous reports demonstrating cognitive impairment of long-term cannabis users in the unintoxicated state, Messinis et al. recently reported neuropsychological deficits in long-term heavy cannabis users after at least 24 hours (range=36 to 240) of abstinence, indicating duration, frequency, and possibly dosage of cannabis administration as determinants of this effect. 23 In the Messinis et al. study, verbal learning was the cognitive domain most substantially affected by chronic heavy cannabis use, followed by psychomotor speed, attention, and executive functioning. 23 Prior to this study, a meta-analysis of a small number of select studies on the long-term residual neurocognitive effects of cannabis use concluded that among various cognitive domains, impaired learning and retrieval of information were the only cognitive domains to demonstrate a significant effect size. 22 , 24 See Table 1 for a summary of recent studies evaluating the effects of cannabis use on neuropsychological functions. Therefore, deficits in verbal learning and memory tasks in long-term heavy cannabis users have variously been attributed to duration, 23 , 57 frequency of cannabis use, 23 , 55 or cumulative dosage effects. 56 The most pronounced evidence of impaired learning are effects on recall after interference or delay, flatter learning curves, fewer words recalled on each learning trial, and poorer recognition performance. 24TABLE 1. Summary of Recent Studies Providing Evidence for Neuropsychological Deficits in Cannabis UsersTABLE 1. Summary of Recent Studies Providing Evidence for Neuropsychological Deficits in Cannabis UsersEnlarge table The age of the cannabis users participating in neuropsychological studies also appears to influence their verbal learning abilities. More specifically, Fletcher et al. reported that only older (≥45 years) cannabis users differed from control subjects in list learning abilities, whereas users younger than 28 years remained unaffected. 58 Age at onset of cannabis use also seems to influence the occurrence of cognitive impairment, possibly attributable to a neurotoxic effect of cannabis on the developing brain. 59 Pope et al. recently reported that early onset of cannabis use (before the age of 17) is associated with poorer verbal IQ, even after adequate abstinence periods. 60 One important caveat in interpreting these results is that different behavioral trends of adolescents who use cannabis could account for these observations. In addition, visual search was found to be disturbed in early onset, long-term cannabis users. 61 The latency of auditory-evoked potentials has been reported to be significantly longer in cannabis users, especially in early onset users compared to healthy control subjects. In further investigating this issue, brain imaging studies report that adolescents who begin cannabis use before the age of 17 exhibit brain morphological changes, such as smaller brain volume, less cortical gray matter, increased cortical white matter, and increased cerebral blood flow, compared with late-onset users. 62 Increasing evidence further suggests that cannabis-based medicinal extracts may have toxic effects on the fetal nervous system. 19 Macleod et al., 63 in a systematic review of the psychological and social sequelae of cannabis and other illicit drug use by young people, reported that available data do not support a causal relationship between cannabis use by young people and psychosocial harm, but cannot exclude the possibility of such a relationship existing. There is laboratory evidence in support of neurotoxicity of THC. Results of both animal studies and in vitro cell experiments, however, have provided conflicting results. 3 , 44 Chronic heavy cannabis use is not associated with structural changes within the brain as a whole or the hippocampus in particular, 64 so functional alterations might rather account for the observed neurocognitive effects. Clinical neurophysiological studies have further contributed toward investigating this issue. Regular use of cannabis was associated with abnormalities in the P300 wave of event-related potentials during an auditory discrimination task 65 and electroencephalographic abnormalities in another study. 66 Neuroimaging studies have also provided results indicative of persistent effects of cannabis on brain functions. Block et al. 67 and Matochik et al. 68 observed structural alterations in the brain tissue of frequent and heavy users, respectively, while Wilson et al. 62 reported brain morphological and functional changes in early onset cannabis users. Using positron emission tomography (PET), Block et al. 69 detected altered memory-related regional blood flow in the brain of frequent marijuana users after a minimum 26 hours of abstinence. Two other PET studies revealed altered patterns of brain activity in abstinent heavy cannabis users while performing tasks depending on executive functioning either in lack 70 or in presence 71 of performance differences. A functional magnetic resonance imaging (fMRI) study showed increased and more widespread brain activation in heavy cannabis users compared to controls while performing a spatial working memory task. 72 Two additional fMRI studies have demonstrated alterations in brain activity in heavy cannabis users during working memory assessments. 73 , 74 Recently reported preliminary data from an fMRI study of verbal learning and memory in long-term cannabis users found altered activation of the frontal, medial, parietal, and cerebellar regions during the encoding and retrieval processes of words learned from the Rey Auditory Verbal Learning Test. 24 , 75Cannabis users seem to exhibit compensatory shifts and recruitment of additional brain regions to meet the demands of various cognitive tasks. These findings could mean that long-term heavy cannabis consumption causes neurophysiological deficits, even in the case of minimally impaired or even normal superficial performance on neuropsychological tests. The subtle cannabis-induced changes in brain function may, therefore, have a huge impact on behavioral and cognitive measures in multiple sclerosis patients, in whom cognitive reserve is assumedly limited and compensatory mechanisms insufficient, as will be discussed below.Cognition in Multiple Sclerosis Multiple sclerosis is a multifocal demyelinating disease of the CNS, in which neuroinflammatory as well as degenerative processes are involved. 76 It is the most common cause of nontraumatic neurological disability in young adults and has a major negative impact on quality of life indices. 76 Cognitive impairment is considered one of the clinical markers of the disease. 77 Neuropsychological deficits of varying degree are present in almost 50% of multiple sclerosis patients, interfering with occupational and social functioning even in the absence of significant physical disability. 77 , 78 Cognitive deficits in patients show no strong association to disease characteristics, such as disease duration, physical disability, disease subtype, or disease severity (lesion load in structural magnetic resonance imaging), with studies providing discrepant results. 77 No general consensus exists on the prevalence and special characteristics of cognitive dysfunction in multiple sclerosis, possibly reflecting disease heterogeneity and demyelination of different areas of the CNS. A review of the neuropsychological literature of multiple sclerosis using an effect size analysis, which gathered neuropsychological test results from a total of 1,845 multiple sclerosis patients and 1,265 healthy controls, confirmed that neurocognitive impairments are indeed evident in patients with multiple sclerosis on a number of cognitive measures. 79 This review also reported that chronic progressive multiple sclerosis patients display maximal deficits on frontal-executive tasks, whereas patients with relapsing-remitting multiple sclerosis present impairments predominantly on tasks of memory function. 79 Although the pattern of cognitive deficits identified in multiple sclerosis is not uniform, the domains most essentially affected are information processing speed and verbal memory. 77 , 80 , 81 A recent review of cognitive impairment in multiple sclerosis confirmed that measures of information processing speed appeared to be the most robust and sensitive markers of this impairment, and that single, predominantly speed-related cognitive measures may be superior to extensive and time-consuming test batteries in detecting cognitive decline. 82 Cognitive flexibility and executive functions, such as planning, have also been reported to be deficient. 77 A recent meta-analysis indicated that phonemic and semantic fluency, measures of executive functions, were among the most sensitive neuropsychological tests to detect cognitive impairment in multiple sclerosis patients. 83 In a large series of relapsing-remitting multiple sclerosis patients, information processing speed was the cognitive domain most frequently impaired, followed by memory. 81 On neuropsychological testing of early phase relapsing-remitting multiple sclerosis patients, information processing speed and memory task performances were impaired. 84 Deficits in the tasks of attention and information processing occurring early in the disease course have been attributed to working memory impairment and may, at least in part, account for subsequent decline in memory—mainly retrieval of information and conceptual reasoning. 85 Working memory deficits occurring early in multiple sclerosis were evident in two studies. 86 , 87 Working memory is dependent on a network of functionally interconnected brain regions, which subtends rehearsal loops handling information load, that exceeds short-term memory storage capacity. 88 , 89 Cognitive dysfunction in multiple sclerosis may result from an interruption of these loops, possibly due to white matter lesions. Compensatory mechanisms operate within the network of regions involved. Reserved brain regions may be recruited as a consequence of an adaptive response to neuronal dysfunction in one region in order to meet the demands of a cognitive task. Cognitive measures are not strongly and consistently correlated to structural damage (lesion load or brain atrophy) in conventional magnetic resonance imaging (MRI). 77 , 90 More advanced MRI methods, such as magnetization transfer ratio or diffusion tensor imaging, capable of detecting subtle abnormalities in the normal-appearing white matter have been used, and the findings were shown to better correlate with cognitive impairment in multiple sclerosis patients. 91 , 92 Explaining cognitive deficits in multiple sclerosis using multiple disconnections has recently focused interest. A dynamic assessment of brain activity during cognitive tasks has been attempted by three rec