Sage Journals: Discover world-class research

Abstract

Aims

A case report suggested the efficacy of cannabis to treat cluster headache (CH) attacks. Our aims were to study the frequency of cannabis use in CH patients, and the reported effects on attacks.

Methods

A total of 139 patients with CH attending two French headache centers filled out questionnaires.

Results

Sixty-three of the 139 patients (45.3%) had a history of cannabis use. As compared to nonusers, cannabis users were more likely to be younger (p < 0.001), male (p = 0.002) and tobacco smokers (p < 0.001). Among the 27 patients (19.4% of the total cohort) who had tried cannabis to treat CH attacks, 25.9% reported some efficacy, 51.8% variable or uncertain effects, and 22.3% negative effects.

Conclusions

Cannabis use is very frequent in CH patients, but its efficacy for the treatment of the attacks is limited. Less than one third of self-reported users mention a relief of their attacks following inhalation. Cannabis should not be recommended for CH unless controlled trials with synthetic selective cannabinoids show a more convincing therapeutic benefit.

Keywords

Cluster headache cannabis tobacco smoking lifestyle

Introduction

Cluster headache (CH) is a primary headache condition characterized by attacks of unilateral pain accompanied by autonomic symptoms and restlessness (1). Usual acute treatments are subcutaneous sumatriptan and normal pressure oxygenotherapy (2). Marijuana, extracted from the cannabis sativa plant, is a recreational drug used illegally in France in two forms: resin or leaves, taken by inhalation. The main active substance in cannabis is delta-9-tetrahydrocannabinol (THC), which is a non-selective agonist of cannabinoid receptors CB1 and CB2 (3), but other cannabinoid compounds such as cannabidiol may also be present (4). Following the publication of a case report stating the efficacy of cannabis and dronabinol to treat CH attacks in a patient refractory to usual treatments (5), we decided to investigate the frequency of regular use among CH sufferers, and the effect on attacks as reported by our patients.

Methods

From July to October 2009, all patients diagnosed with CH presenting to two centers, namely the Emergency Headache Centre at Lariboisière Hospital in Paris and the Neurology Department of La Timone Hospital in Marseille, were asked to fill out an anonymous questionnaire. All participants were informed that the information was collected for research purposes, would be kept confidential and gave consent to fill out the form. The diagnosis and subtype (episodic or chronic) of CH was confirmed according to the second edition of the International Classification of Headache Disorders (ICHD-II) criteria. Usual acute and prophylactic treatments were noted. Questions about substance use are reproduced in Table 1. Patients were asked about their habits concerning tobacco and cannabis. For cannabis, we documented if the use was ancient (stopped more than six months ago), or recent (during the preceding six months). Patients with experience of using cannabis were asked if they ever used cannabis specifically to treat CH attacks and, if that was the case, what effect they had observed. Smokers were asked if they tended to avoid tobacco and cannabis during the active periods or exacerbations of CH, and if cannabis could trigger attacks.

Table 1.

Questionnaire provided to patients, section about use of neuroactive substances.

Questions	Multiple choices
Are you currently smoking tobacco?	Yes/no
Did you ever use cannabis?	Yes/no Specified that an isolated trial during adolescence was not considered as yes
Did you use cannabis during the last six months?	Yes/no
Did you ever try cannabis to treat your cluster headache attack?	Yes/no
When did you use cannabis to treat your CH attack?	>5 years 1–5 years During the last year Current
How many attacks have you treated with cannabis?	1–2 3–10 >10
What are your conclusions about the effect of cannabis on the CH attack?	Very efficient Efficient >50% of the time Uncertain, unpredictable No effect Deteriorates the symptoms
If you are a smoker, do you tend to avoid tobacco during the CH period?	Yes/no
If you use cannabis, do you tend to avoid cannabis during the CH period?	Yes/no

CH: cluster headache.

Statistical analysis

Description of the cohort was performed with means and 95% two-tailed confidence intervals (CIs). Comparison between cannabis users and nonusers were done with the Student t test. The analysis was performed by University of Montreal Statistical Consulting Service using SPSS version 15 and a significance level of 0.05.

Results

Clinical characteristics

During the study period, 139 patients (116 males (M)/23 females (F), sex ratio 5.0) agreed to fill out the questionnaire (Table 2). One hundred and thirty-four patients were recruited in Paris and 35 in Marseille. There were 93 episodic cases (66.9%) and 42 chronic cases (30.2%). The subtype of CH was not defined in four patients (2.9%) who had CH for less than one year. Usual acute treatments were subcutaneous sumatriptan in 77% of the patients, inhaled oxygen in 5.7%, and other or not specified in 17.3%. The usual prophylactic treatments were verapamil alone in 53%, combination of verapamil and suboccipital steroid injections (SSI) in 13%, combinations of other drugs including lithium, topiramate, steroids, amitryptiline, gabapentin, indoramine or lamotrigine in 8%, and SSI alone in 3%. The remaining patients (23%) had no usual prophylaxis.

Table 2.

Characteristics of patients.

	All patients n = 139 (%)	Males n = 116 (%)	Females n = 23 (%)
Mean age (years ± SD)	40.4 ± 11.3	40 ± 10.2	43 ± 12.7
CH subtype
Episodic, n (%)	93 (66.9)	78 (67.2)	15 (65.2)
Chronic, n (%)	42 (30.2)	35 (30.2)	7 (30.4)
Not available, n (%)	4 (2.9)	3 (2.6)	1 (4.3)
Cannabis
Users, n (%)	63 (45.3)	59 (50.9)	4 (17.4)
Nonusers, n (%)	73 (52.5)	54 (46.5)	19 (82.6)
Not available, n (%)	3 (2.2)	3 (2.6)	0
Cannabis for CH attack
Yes, n (%)	27 (19.4)	27 (23.3)	0
No, n (%)	112 (80.6)	89 (76.7)	23 (100)
Not available, n (%)	0	0	0
Tobacco
Smokers, n (%)	103 (74.1)	87 (75)	16 (69.6)
Nonsmokers, n (%)	35 (25.2)	28 (24.1)	7 (30.4)
Not available, n (%)	1 (0.7)	1 (0.9)	0

CH: cluster headache.

Cannabis use and profile of users

Overall, 63 patients (59 M/4 F, sex ratio 15) reported a history of cannabis use (45.3%, 95% CI 37% to 54%) and 45 patients (32%, 95% CI 25% to 40%) had used cannabis during the past six months. A total of 103 patients were current tobacco smokers (74.1%, 95% CI 67% to 82%). Three patients did not answer about their cannabis use (Table 2).

Cannabis users had a younger age (36.2 vs 44.0 years, p < 0.001, 95% CI of the difference 4.1 to 11.4) and were more frequently males (93.6% vs 74%, p = 0.002, 95% CI of the difference −32% to −7.3%) or active tobacco smokers (92% vs 59%, p < 0.001, 95% CI of the difference −47% to 19%) than the nonusers (Table 3). There was a nonsignificant trend for an association between cannabis use and chronic cluster (37% vs 24%, p = 0.11, 95% CI for the difference −26% to 3%).

Table 3.

Comparison of cannabis users to nonusers among CH patients.

	Cannabis users (n = 63)	Nonusers (n = 73)	p value^a (95% CI)
Mean age (years ± SD)	36.2 ± 7.9	44.0 ± 12.5	<0.001 (4.1–11.4)
Percentage of male patients	94%	74%	0.002 (−32% to −7.3%)
Percentage of chronic patients	37%	24%	0.11 (−28% to 3%)
Percentage of tobacco smokers	92%	59%	<0.001 (−47% to −19%)

CH: cluster headache; CI: confidence interval for a t test on independent samples; ^aBased on Student’s t test for independent samples.

Effect of cannabis on CH attacks

A total of 27 of the 139 patients (19.4%) reported having tried cannabis for the acute treatment of cluster attacks (Table 2). Among these 27 patients, 25 also described themselves as cannabis users whereas two did not, but reported an isolated use for a single attack. Seven patients had tried cannabis as an acute CH treatment only once or twice whereas 20 patients had tried three times or more. The last trial dated back to more than five years in nine patients, one to five years in four, and was recent (one year or current) in 13. One patient did not specify the timing of the trial. Cannabis was judged very efficient by one patient, more than 50% efficient by six patients, variable or uncertain by 14 patients, and inefficient or aggravating by six patients (Figure 1).

Figure 1.

Self-reported efficacy of cannabis for cluster headache attacks in 27 patients.

During the active periods of episodic CH or during exacerbations of chronic CH, 46% of the 103 tobacco smokers tended to avoid tobacco while 50% did not, and 4% did not provide an answer; 43% of the 63 cannabis users tended to avoid cannabis while 38% did not, and 19% did not provide an answer. In addition, 25.3% of the 63 cannabis users thought that cannabis could provoke attacks while 57.2% did not, and 17.5% did not provide an answer. Three patients spontaneously reported that cannabis was beneficial by bringing relaxation and quietness, not by acting on the pain. Four patients wrote that cannabis could either help or provoke an attack.

Discussion

Our bi-centric study shows for the first time a strikingly high prevalence of cannabis use among French episodic and chronic CH patients, with 45.3% of ever users and 32% of recent or current users. These figures seem much higher than in the general population, although cannabis is the most frequently used recreational illicit drug in France, with a prevalence of 7% in the preceding year among the 18- to 64-year-old age group, according to a survey conducted in 2005 (6). In a previous multi-centric French study, Donnet et al. reported that 26% of 113 patients with chronic CH did use cannabis regularly, which is in accordance with the 32% found in the present study (7). A recent study from the United States described a lower prevalence of cannabis use (10.2%) in a cohort of 49 CH patients (8). This lower figure could be explained by methodological differences, under-reporting of use, cultural differences, and a higher mean age (47.8 vs 36.2 years old) of the patients in the study from the United States than in our French series (8). We found that cannabis use in CH patients was associated with younger age, male sex, and tobacco use, similar to what is observed in the adult French general population (6). However, the male-female sex ratio in cannabis users seems higher in CH patients compared to the French general population (15 vs 2.4) (6).

Previous studies have underlined that CH patients, regardless of gender, were heavy smokers (9,10). Our survey confirms the high prevalence of tobacco smokers in CH patients with 74.1% as compared to 29% in the 18- to 64-year-old age group from the French general population (6). The nature of the relationship between tobacco and CH is still a matter of debate. Smoking may be a consequence of the pain condition. Alternatively, some authors have suggested that tobacco may play a role in the appearance of CH, because its use usually predates the onset of CH (11). Rozen has suggested that early exposure to tobacco smoke was a risk factor for CH (12). The relationship between cannabis use and CH raises the same questions. As we did not record the respective age of onset of CH, tobacco use and cannabis use, we cannot provide a temporal sequence in our panel. Another hypothesis could be that the molecular mechanisms underlying CH also increase the susceptibility toward addictive behaviors such as tobacco and cannabis use. Some studies have described specific personality traits in CH patients (13 –15), but to our knowledge, no study has used addictive personality questionnaires. Although the role of the dopaminergic system, a key player in addiction mechanisms (16), is still incompletely understood in the pathophysiology of CH, increased levels of dopamine in platelets have been described in some patients (17), and some cases respond to dopamine agonists (18).

The endogenous cannabis system has effects on cardiovascular function, nociception, cognitive function, appetite and body temperature regulation, memory, anxiety, motor behavior and mood (19). Synthetic cannabinoids have been tried for different diseases, and their use is currently approved in North America for the treatment of multiple sclerosis-associated spasticity, nausea induced by chemotherapy and treatment of intractable cancer pain (20). In a study of epilepsy, 21% of 36 users claimed cannabis had a prominent efficacy on their condition (21). In addition, some patients use cannabis for headache and migraine (22,23).

Our study provides interesting new data about the effects of cannabis on CH attacks. Contrary to what we expected, CH patients mostly rated the effects of cannabis as moderate, variable and unpredictable. Cannabis was even reported as a trigger or a deteriorating agent for CH pain by a significant proportion of patients, conducting them to reduce their use during active CH periods.

The unpredictability of cannabis’ effect on CH attacks may be explained either by variation in THC content between users and between uses, or by different actions of THC depending on other factors in users. Up to 100 different cannabinoids can be found in different preparations (24), each one with a specific pharmacological profile. For example, THC is anxiogenic, whereas cannabidiol is anxiolytic, and both substances have different effects on functional magnetic resonance imaging (fMRI) studies (25). Plants may also be grown with other substances like stimulants. In theory, cannabis may exert an effect on the pain of CH attack either through activity on the cerebral and dural arteries, release of inflammatory substances like calcitonin gene-related peptide (CGRP) and vasoactive intestinal polypeptide (VIP), or via the pain modulation system. Cannabinoids mainly cause vasodilation of cerebral arteries (26,27). It seems unlikely that a vasodilator effect could treat a CH attack, but it could explain why cannabis triggers or worsens CH attacks in some patients. According to multiple animal experiments, the effect of cannabinoids on pain (28) is multifocal, involving peripheral afferents, brainstem modulation areas, and cortical areas of the pain matrix. Cannabinoids have been shown to have more influence on tonic pain and allodynia than on acute pain, which may explain the limited effects on acute CH attacks (29). Whether cannabis could or could not have a prophylactic effect on CH is an unresolved issue. Our study aimed to study the effects of cannabis on acute CH attacks, and not the effects of regular use on the duration of CH bouts, or the frequency of attacks.

Our study has several limitations. It was not case controlled, and it was not exhaustive. Due to the setup of the emergency department, it is possible that some CH patients were not asked to fill out the questionnaire, and we did not keep track of the number of patients who declined, but this number is estimated to be low. Moreover, our study did not evaluate psychiatric comorbidities. Several epidemiological studies conducted in the general population have evidenced an association between tobacco smoking or nicotine dependence and depression (30). Future studies on large cohorts of CH patients should assess substance use, addiction scores, depression, anxiety and their relationships.

Conclusion

From our observation, we conclude that substance use should always be addressed when caring for CH patients. Cannabis use is very frequent, especially in young men, and has medical and legal implications. The reason for the high prevalence of cannabis use in CH patients remains to be determined. We advise our patients to stop using cannabis and provide them with adequate medications to manage sleep difficulties and anxiety symptoms. In contrast with its wide use among CH patients, cannabis seems to have limited efficacy when used to treat attacks, and can even worsen them in some patients. The therapeutic potential of some cannabinoids is not discarded by our results, but considering the legal issues surrounding cannabis, structured trials with adequately controlled doses of selective synthetic cannabinoid agonists should be the first way to investigate it further.

Clinical implications

The prevalence of cannabis use appears to be higher in cluster headache (CH) patients than in the general population.

When used during a CH attack, cannabis has unpredictable effects in half the patients, has a modest effect in one fourth, and deteriorates the attack in another fourth. Cannabis use can also trigger an attack.

Cannabis should not be considered as a therapeutic option for the CH attack.

CH patients should be screened for cannabis use, and encouraged to abstain.

Footnotes

Funding

This research received no specific grant from any agency in the public, commercial or not-for-profit sectors.

Conflict of interest

None declared.

References

Capobianco

Dodick

. Diagnosis and treatment of cluster headache. Semin Neurol 2006; 26: 242–259.

Cohen

Matharu

Goadsby

. Trigeminal autonomic cephalalgias: Current and future treatments. Headache 2007; 47: 969–980.

Goulle

Saussereau

Lacroix

. Delta-9-tetrahydrocannabinol pharmacokinetics [in French]. Ann Pharm Fr 2008; 66: 232–244.

Elsohly

Slade

. Chemical constituents of marijuana: The complex mixture of natural cannabinoids. Life Sci 2005; 78: 539–548.

Robbins

Tarshish

Solomon

. Cluster attacks responsive to recreational cannabis and dronabinol. Headache 2009; 49: 914–916.

Beck

Legleye

Spilka

. Les niveaux d'usage des drogues en France. Tendances 2006; May.

Donnet

Lanteri-Minet

Guegan-Massardier

. Chronic cluster headache: A French clinical descriptive study. J Neurol Neurosurg Psychiatry 2007; 78: 1354–1358.

Robbins

Bronheim

Lipton

. Depression and anxiety in episodic and chronic cluster headache: A pilot study. Headache 2012; 52: 600–611.

Levi

Edman

Ekbom

. Episodic cluster headache. II: High tobacco and alcohol consumption in males. Headache 1992; 32: 184–187.

10.

Hannerz

. Symptoms and diseases and smoking habits in female episodic cluster headache and migraine patients. Cephalalgia 1997; 17: 499–500.

11.

Torelli

Cologno

Manzoni

. Episodic and chronic cluster headache males: Differences in age at onset and habits. In: Olesen

Goadsby

(eds) Cluster headache and related conditions. Oxford: Oxford University Press, 1999, pp.42–47.

12.

Rozen

. Cluster headache as the result of secondhand cigarette smoke exposure during childhood. Headache 2010; 50: 130–132.

13.

Cuypers

Altenkirch

Bunge

. Personality profiles in cluster headache and migraine. Headache 1981; 21: 21–24.

14.

de Domini

Del Bene

Gori-Savellini

. Personality patterns of headache sufferers. Cephalalgia 1983; 3(Suppl 1): 195–214.

15.

Levi

Edman

Ekbom

. Episodic cluster headache. I: Personality and some neuropsychological characteristics in male patients. Headache 1992; 32: 119–125.

16.

Volkow

Fowler

Wang

. Dopamine in drug abuse and addiction: Results of imaging studies and treatment implications. Arch Neurol 2007; 64: 1575–1579.

17.

D'Andrea

Granella

Perini

. Platelet levels of dopamine are increased in migraine and cluster headache. Headache 2006; 46: 585–591.

18.

Palmieri

. Chronic cluster headache responsive to pramipexole. Cephalalgia 2006; 26: 761–762.

19.

Di Marzo

. The endocannabinoid system: Its general strategy of action, tools for its pharmacological manipulation and potential therapeutic exploitation. Pharmacol Res 2009; 60: 77–84.

20.

Aggarwal

Carter

Sullivan

. Medicinal use of cannabis in the United States: Historical perspectives, current trends, and future directions. J Opioid Manag 2009; 5: 153–168.

21.

Gross

Hamm

Ashworth

. Marijuana use and epilepsy: Prevalence in patients of a tertiary care epilepsy center. Neurology 2004; 62: 2095–2097.

22.

Russo

. Cannabis for migraine treatment: The once and future prescription? An historical and scientific review. Pain 1998; 76: 3–8.

23.

Ogborne

Smart

Weber

. Who is using cannabis as a medicine and why: An exploratory study. J Psychoactive Drugs 2000; 32: 435–443.

24.

Aggarwal

Carter

Sullivan

. Characteristics of patients with chronic pain accessing treatment with medical cannabis in Washington State. J Opioid Manag 2009; 5: 257–286.

25.

Bhattacharyya

Morrison

Fusar-Poli

. Opposite effects of delta-9-tetrahydrocannabinol and cannabidiol on human brain function and psychopathology. Neuropsychopharmacology 2010; 35: 764–774.

26.

Hillard

Thompson

. Inhibition of 2-arachidonoylglycerol catabolism modulates vasoconstriction of rat middle cerebral artery by the thromboxane mimetic, U-46619. Br J Pharmacol 2007; 152: 691–698.

27.

Akerman

Kaube

Goadsby

. Anandamide acts as a vasodilator of dural blood vessels in vivo by activating TRPV1 receptors. Br J Pharmacol 2004; 142: 1354–1360.

28.

Ibrahim

Rude

Stagg

. CB2 cannabinoid receptor mediation of antinociception. Pain 2006; 122: 36–42.

29.

Walker

Hohmann

. Cannabinoid mechanisms of pain suppression. Handb Exp Pharmacol 2005: 509–554.

30.

Tsuang

Francis

Minor

. Genetics of smoking and depression. Hum Genet 2012; 131: 905–915.

Use of cannabis among 139 cluster headache sufferers

Abstract

Aims

Methods

Results

Conclusions

Keywords

Introduction

Methods

Statistical analysis

Results

Clinical characteristics

Cannabis use and profile of users

Effect of cannabis on CH attacks

Discussion

Conclusion

Clinical implications

Footnotes

Funding

Conflict of interest

References