VIEWS & REVIEWS Acute and preventive pharmacologic treatment of cluster headache

Transcription

1 VIEWS & REVIEWS Acute and preventive pharmacologic treatment of cluster headache George J. Francis, BSc Werner J. Becker, MD Tamara M. Pringsheim, MD Address correspondence and reprint requests to Dr. Tamara Pringsheim, Department of Clinical Neurosciences, 2888 Shaganappi Trail NW, Calgary, Alberta, T3B 6A8, Canada ABSTRACT Cluster headache (CH) is a rare and disabling primary headache disorder. CH attacks are unilateral, short, severe headaches associated with ipsilateral autonomic symptoms that occur in a periodic fashion. We provide a systematic review and meta-analysis of existing trials of pharmacotherapy for CH and evidence-based suggestions for acute abortive treatment and preventive therapy for cluster headache. Prospective, double-blind, randomized controlled trials of any pharmacologic agent for the symptomatic relief or prevention of CH were included in this evidence-based review. The main outcomes considered were headache response and pain-free response at 15 and 30 minutes for acute treatment trials, and the cessation of CH attacks within a specific time period or the number of days on which CH attacks occurred for preventive trials. Twenty-seven trials were included in the analysis. The American Academy of Neurology quality criteria were used to assess trial quality and to grade advisements. Based on the evidence, for acute treatment of CH, Level A advice can be given for subcutaneous sumatriptan 6 mg, zolmitriptan nasal spray 5 mg and 10 mg, and 100% oxygen 6 12 L/min. Level B advice can be given for sumatriptan nasal spray 20 mg and oral zolmitriptan 5 mg and 10 mg. For the prevention of CH, Level B advice can be given for intranasal civamide 100 g daily and suboccipital steroid injections, and Level C advice can be given for verapamil 360 mg, lithium 900 mg, and melatonin 10 mg. Neurology 2010;75: GLOSSARY AAN American Academy of Neurology; AE adverse event; CH cluster headache; CI confidence interval; NS nasal spray; OR odds ratio; RCT randomized controlled trial; SC subcutaneous; SCHSG Sumatriptan Cluster Headache Study Group. Supplemental data at The International Classification of Headache Disorders, 2nd edition, 1 defines CH as at least 5 severe to very severe unilateral headache attacks, lasting 15 to180 minutes untreated. Attacks are accompanied by ipsilateral autonomic symptoms, restlessness, or agitation. Attack frequency ranges from 1 every other day to 8 per day. There are 2 types of CH: episodic CH and chronic CH. Patients with episodic CH have attacks that occur in series lasting weeks or months separated by remission periods usually lasting months or years. Patients with chronic CH have attacks occurring for more than 1 year without remission or with remissions lasting less than 1 month. A total of 10% to 15% of patients with CH have chronic CH. 1 The pathogenesis of CH is inly understood, although hypothalamic dysfunction is suspected. 2 There is clinical evidence for altered biologic rhythms in patients with CH, and the timing of attacks relates to the sleep-wake cycle in many patients. 3 The uncertainty regarding the pathogenesis of CH renders it difficult to aim treatments toward a specific target. Treatment of CH has 2 strategies: symptomatic therapy taken at the time of an attack, and preventive therapy taken when the cluster bout begins to prevent further attacks. As CH causes significant disability, the investigation of evidence-based medical treatments is warranted. We present a systematic review and meta-analysis of treatment trials for CH and evidence-based advice for acute and preventive treatments of CH. Editor s Note: This article reflects the opinions of the authors and, unlike AAN evidence-based reviews, has not been endorsed by the AAN. From the Department of Clinical Neurosciences, University of Calgary, Canada. Disclosure: Author disclosures are provided at the end of the editorial. Copyright 2010 by AAN Enterprises, Inc. 463

2 OBJECTIVES/CLINICAL QUESTIONS Which acute treatments for CH are effective in reducing headache severity? Which preventive treatments are effective in reducing CH attacks? CRITERIA FOR CONSIDERING STUDIES Studies were required to be prospective, double-blind, parallel-group or crossover, randomized controlled trials (RCTs) of any medication vs placebo or another drug, for the symptomatic relief or prevention of CH. Study participants were 18 or older with either episodic CH or chronic CH. For efficacy analysis, the following main outcomes were considered: 1. Headache response at 15 or 30 minutes, defined as a reduction in headache from moderate, severe, or very severe to mild or no pain (symptomatic trials) 2. Pain-free response at 15 or 30 minutes (symptomatic trials) 3. Cessation of CH attacks within a specific time period (preventive trials) 4. Number of days on which a CH attack occurred (preventive trials) For the analysis of data on adverse events (AEs), we considered the overall number of patients reporting AEs. SEARCH METHODS FOR IDENTIFICATION OF STUDIES MEDLINE (1950 to June 2009) and EMBASE (1980 to June 2009) databases were searched for double-blind RCTs (see appendix e-1 on the Neurology Web site at for detailed search strategy). METHODS OF REVIEW Two reviewers (T.P. and G.J.F.) independently screened titles and abstracts for trials fulfilling inclusion criteria. Data were abstracted independently by the reviewers and confirmed for accuracy. Discrepancies between reviewers were resolved by discussion. The studies were evaluated using quality criteria developed by the American Academy of Neurology (AAN) 4 (see appendix e-2). Suggestions were made according to the AAN grades for classification of recommendations (see appendix e-3). SUMMARIZING THE RESULTS Meta-analysis was performed by treatment type. Odds ratios (OR) were calculated with 95% confidence intervals (CI). ORs from multiple studies were tested for homogeneity using the 2 test and by calculating the I 2 statistic. If study estimates were homogenous, they were combined using a fixed-effects model. When studies with heterogeneous results were clinically similar, the study estimates were combined using a randomeffects model. Clinical heterogeneity was by looking at trial and patient characteristics, and outcome measures. Clinically heterogeneous studies were not statistically combined. DESCRIPTION OF STUDIES A total of 1,547 abstracts were found by the combined searches (see the figure for flow diagram). A total of 1,499 abstracts did not meet inclusion criteria. Forty-nine full-text articles were reviewed. Twenty-three did not meet inclusion criteria (see appendix e-4). The remaining 26 studies were included in the analysis: 11 on acute treatment of CH, on prevention of CH, and 1 on both acute treatment and prevention of CH. 30 The MEDLINE search was re-executed prior to final revision of the manuscript (February 1, 2010). This revealed 22 more abstracts published since the original search. One abstract met inclusion criteria and was included in the study, 31 giving a total of 27 studies in the analysis. RCTs for acute treatment of CH had similar inclusion criteria. Studies included patients 18 to 65 years of age, with a history of episodic CH or chronic CH, and attacks of 45 minutes minimum duration untreated. Trials using triptans excluded patients with heart disease, uncontrolled hypertension, or concomitant use of ergotamine or monoamine oxidase inhibitors. The most common endpoints used were the headache response or pain-free response at 30 minutes. A summary of the trial characteristics, methodologic flaws, and results can be seen in table 1. Inclusion criteria for RCTs studying prevention of CH included patients with at least 1 cluster period lasting at least 1 month prior to the study. Patients were in a cluster period for no more than 10 days, with an expected remainder of cluster period of no less than 20 days. Specifically excluded were patients with cardiovascular, pulmonary, hepatic, or renal disease, as well as patients who were pregnant or currently taking prophylactic CH medication. The primary endpoint for studies was a reduction in the number of headaches during the period of study medication use, in comparison to a run-in period with no CH prophylaxis. A summary of the trial characteristics, methodologic flaws, and results can be seen in table 2. RESULTS Acute treatment. Which acute treatments for CH are effective in reducing pain? (See table 3 for an advice summary.) Sumatriptan. Three RCTs compare sumatriptan to placebo, all meeting AAN Class I criteria. 5-7 Ekbom et al. 5 and The Sumatriptan Cluster Head- 464 Neurology 75 August 3, 2010

3 Figure PRISMA flow diagram ache Study Group 6 (SCHSG) performed crossover studies of sumatriptan 6 mg subcutaneous injection (SC) vs placebo for 2 attacks. The study by Ekbom et al. also sumatriptan 12 mg. van Vliet et al. 7 performed a 2-group crossover study of sumatriptan 20 mg nasal spray (NS) vs placebo for 2 CH attacks. Only the Ekbom et al. and the SCHSG studies were combined statistically, as both these trials used sumatriptan SC. Sumatriptan SC 6 mg or 12 mg. Headache response at 15 minutes: the 2 studies included 258 CH attacks. 5,6 Meta-analysis using a random effects model gave a summary OR of 6.22 in favor of sumatriptan (95% CI , p ). No difference was found between the 2 sumatriptan doses. The most common AEs reported were injection site reactions, nausea and vomiting, dizziness, fatigue, and paraesthesias. Sumatriptan NS 20 mg. Headache response at 30 minutes: this crossover study included 83 patients. Two attacks were treated with sumatriptan 20 mg NS or placebo. 7 Sumatriptan was superior to placebo, as 57% of patients reported headache relief with sumatriptan compared to 26% with placebo (p 0.002). The most frequent AE reported was a bitter taste following use of the nasal spray in 21% of patients. Sumatriptan SC should be offered to patients for acute treatment of CH (Level A). Sumatriptan NS should be considered for acute treatment of CH (Level B). Zolmitriptan. Three RCTs compare zolmitriptan to placebo, all meeting AAN Class I criteria Two are crossover studies 8,9 comparing zolmitriptan NS 5 mg and 10 mg to placebo. The other 10 compares oral zolmitriptan 5 mg and 10 mg to placebo. Zolmitriptan NS, 5 and 10 mg. Headache response at 30 minutes: Both studies used a primary efficacy analysis of headache response at 30 minutes, including 451 CH attacks. 8,9 Meta-analysis using a random effects model found a summary OR of 5.03 (95% CI , p ) for zolmitriptan 10 mg Neurology 75 August 3,

4 Table 1 Summary of results: The effect of acute treatment of cluster headache Reference no. No. Quality assessment Intervention Primary outcome Dropouts Results p Value Quality criteria unfulfilled AAN Class I Sumatriptan 6 or 12 mg SC vs placebo; 2 attacks Headache response at 10 min 134 patients treated 2 attacks Headache response: sumatriptan 12 mg 63%; sumatriptan 6 mg 49%; placebo 25% AAN Class I Sumatriptan 6 mg SC vs placebo; 2 attacks Headache response at 15 min 39 patients evaluated Headache response: sumatriptan 74%; placebo 26% AAN Class I Sumatriptan 20 mg nasal spray vs placebo; 2 attacks Headache response at 30 min 85 patients treated 2 attacks Headache response: sumatriptan 57%; placebo 26% AAN Class I Zolmitriptan 5 or 10 mg nasal spray vs placebo; 3 attacks Headache response at 30 min 52 patients treated attack Headache response: zolmitriptan 10 mg 63.3%; zolmitriptan 5 mg 50%; placebo 30% 0.01; AAN Class I Zolmitriptan 5 or 10 mg nasal spray vs placebo; 3 attacks Headache response at 30 min 69 patients treated attack Headache response: zolmitriptan 10 mg 61%; zolmitriptan 5 mg 42%; placebo 23% AAN Class I Zolmitriptan 5 or 10 mg tab vs placebo; 3 attacks Headache response at 30 min 124 patients took trial med,123 evaluated Headache response: zolmitriptan 10 mg 47%; zolmitriptan 5 mg 40%; placebo 29% AAN Class I 100% oxygen 6 L/min vs room air; 1 attack Headache relief during attack All patients treated attacks Headache relief: oxygen 56%; room air 7% AAN Class II Cocaine 1 ml intranasal vs lidocaine 1 ml intranasal vs placebo; 1 attack Complete cessation of pain 9 patients had nitroglycerin induced attacks and were treated Complete cessation of pain: cocaine 31.3 min; lidocaine 37.0 min; placebo 59.3 min 0.01 No statement on allocation AAN Class II Octreotide SC 100 gvs placebo; 1 attack Headache response at 30 min 48 patients treated attack Headache response: octreotide 52%; placebo 36% 0.01 No statement on allocation AAN Class III Dihydroergotamine 1 mg nasal spray vs placebo; maximum 8 attacks Reduction in headache severity during the attack (not welldefined) 22 treated attacks Number of attacks strongly reduced: dihydroergotamine 41; placebo No statement on allocation ; no primary outcome stated 15 8 AAN Class III Somatostatin 25 givvs ergotamine 250 gimvs placebo IV; 1 attack Reduction of maximal pain intensity (not well-defined) Not stated Maximal pain reduction: somatostatin 18%; ergotamine 14%; placebo 0% No statement on allocation ; no primary outcome stated AAN Class III Prednisone 30 mg tablet vs placebo Pain relief (not well-defined) All patients treated attacks 17/19 patients improved 0.03 No statement on allocation ; no primary outcome stated AAN Class I 100% oxygen 12 L/min vs room air; 4 attacks Pain-free at 15 min 76 received treatment Pain-free at 15 minutes: 78% oxygen; 20% air Abbreviations: AAN American Academy of Neurology; SC subcutaneous. 466 Neurology 75 August 3, 2010

5 Table 2 Summary of results: The effect of short-term prevention on cluster headache Reference no. No. Quality assessment Intervention Primary outcome Dropouts Results p Value Quality criteria unfulfilled AAN Class I Civamide 100 L intranasal vs placebo; 3 wk Change in attack frequency per wk 4 did not Change in attack frequency at 1 wk posttreatment: civamide 55.5%; placebo 25.9% AAN Class I Rapid- and long-acting steroids (Verum) 2.5 ml vs placebo SC; 2 wk Disappearance of attacks at 72 h 1 wk All Headache-free: Verum 85%; placebo 0% AAN Class I Sodium valproate 500 mg tab vs placebo; 2 wk % of patient improvement ( 50% reduction in attacks/wk) 1 did not % patient improvement: sodium valproate 50%; placebo 62% NS AAN Class I Sumatriptan 100 mg tab tid vs placebo; 1 wk 50% reduction in attack frequency All Reduction in attack frequency: sumatriptan 23%; placebo 22% NS AAN Class II Lithium 800 mg tab vs placebo; 1 wk Cessation of attacks within 1 wk All Cessation of attacks: lithium 15%; placebo 14% NS No statement on allocation AAN Class II Melatonin 10 mg tab vs placebo; 2 wk Reduction in daily headache frequency compared to run-in period All Reduction in headache frequency: melatonin 1.79; placebo No statement on allocation 22 8 AAN Class II Misoprostol (prostaglandin E analogue) 300 g tab bid vs placebo; 2 wk Attack frequency 2 wk posttreatment All Attack frequency: misoprostol 25.1; placebo 26.1 NS No statement on allocation AAN Class II 100% oxygen (hyperbaric) vs sham; 2 attacks 50% reduction in headache index following 1 wk of treatment All Patients with 50% reduction in headache index: hyperbaric oxygen 35.7%; sham 37.5% NS No statement on allocation AAN Class II Verapamil 120 mg tab tid vs lithium 300 mg tab tid; 8 wk Headache index relative to placebo period 6 did not Reduction in headache index: verapamil 50%; lithium 37% 0.01 No statement on allocation AAN Class III Verapamil 120 mg tab tid vs placebo; 2 wk Daily attack frequency per wk All Attack frequency: verapamil 0.6; placebo No statement on allocation, primary outcome not clearly stated AAN Class III Cimetidine 400 mg tid chlorpheniramine 4 mg tid vs placebo; 6 wk Mean number of attacks per wk 3 did not Mean number of attacks at wk 12: cimetidine chlorpheniramine 8.7; placebo 10.7 NS No statement on allocation, intervention not clearly stated AAN Class III Cimetidine 400 mg qid vs placebo; 6 wk Headache incidence and duration 1 did not Incidence improvement: cimetidine 4; placebo 0 NS No detailed results; no statement on allocation AAN Class III Capsaicin intranasal cream vs placebo; 2 wk Headache severity rating over 2 wk 3 excluded Headache severity ratings: capsaicin 2.46; placebo Assembly of incomparable groups; intervention not clearly stated; no statement on allocation 29 9 AAN Class III Nitrate tolerance 5-ISMN 30 mg tab tid vs placebo; 4 wk Summed maximal interval headache score 4 did not Median summed maximal headache score: nitrate tolerance 42; placebo 0 NS No statement on allocation, inadequate accounting for dropouts AAN Class III Prednisone 20 mg every other day for 1 wk vs placebo Attack frequency (not defined) All Greater attack frequency in placebo group, data not given 0.03 No statement on allocation ; no primary outcome stated Abbreviations: AAN American Academy of Neurology; NS not significant; SC subcutaneous. Neurology 75 August 3,

6 Table 3 Summary of advisements for acute abortive treatment of cluster headache Maneuver Effectiveness Levels of evidence Advice Sumatriptan (subcutaneous) Direct evidence that sumatriptan 6 mg is effective in improving headache events: injection site reactions, nausea and vomiting, dizziness, fatigue, paresthesias. clinical trials; 2 AAN Class I Level A: should be offered for acute treatment of CH Zolmitriptan (nasal spray) Direct evidence that zolmitriptan 5 mg and 10 mg are effective in improving headache response in CH. Nonserious adverse events: unpleasant taste, nasal cavity discomfort, somnolence, dizziness, nausea, throat/neck tightness. clinical trials; 2 AAN Class I Level A: should be offered for acute treatment of CH Sumatriptan (nasal spray) Direct evidence that sumatriptan 20 mg is effective in improving headache event: bitter taste. Level B: should be considered for acute treatment of CH Zolmitriptan (oral) Direct evidence that zolmitriptan 5 mg and 10 mg are effective in improving headache response in CH. Nonserious adverse events: paresthesias, heaviness, asthenia, nausea, dizziness, nonchest tightness. Level B: should be considered for acute treatment of CH Oxygen Direct evidence that 100% oxygen 6 12 L/min is effective in improving headache response in CH. Adverse events not reported. clinical trials; AAN Class I Level A: should be offered for acute treatment of CH Cocaine/lidocaine Evidence that 10% cocaine hydrochloride and 10% lidocaine are effective in improving headache events: nasal congestion, unpleasant lidocaine taste. I Level C: may be considered for acute treatment of CH Octreotide Evidence that octreotide 100 gis effective in improving headache events: injection site reactions, diarrhea, abdominal bloating, nausea, dull background headache, lethargy. I Level C: may be considered for acute treatment of CH Dihydroergotamine (nasal spray) Insufficient evidence that dihydroergotamine 1 mg intranasally is effective in improving headache response in CH. Adverse events not reported. to advise for the acute treatment of CH Somatostatin Insufficient evidence that somatostatin 25 g is effective in improving headache event: nausea without vomiting. to advise for the acute treatment of CH Prednisone Insufficient evidence that prednisone 30 mg is effective in improving headache response in CH. to advise for the acute treatment of CH Abbreviations: AAN American Academy of Neurology; CH cluster headache. and 2.61 (95% CI , p 0.001) for zolmitriptan 5 mg. The most common AEs reported were bad taste (22%), nasal cavity discomfort (12%), and somnolence (8%). Zolmitriptan oral, 5 and 10 mg. Headache response at 30 minutes: a single crossover study investigated oral zolmitriptan 10 mg and 5 mg vs placebo for 3 attacks in 102 patients. 10 Both doses of zolmitriptan doses were superior to placebo in patients with episodic CH (p 0.05), but not those with chronic CH. The most common AEs reported were paraesthesia, heaviness, asthenia, nausea, dizziness, and nonchest tightness. Zolmitriptan NS should be offered to patients for acute treatment of CH (Level A). Oral zolmitriptan should be considered for acute treatment of episodic CH (Level B). Oxygen. Two Class I crossover studies have investigated 100% oxygen vs placebo for the acute treatment of CH. Fogan s study 11 evaluated 100% oxygen vs placebo (regular air), at a rate of 6 L/min for up to 15 minutes during an acute attack. Nineteen patients treated up to 6 attacks. The primary endpoint was headache relief following treatment. A total of 56% of patients breathing oxygen perceived substantial headache relief, compared with 7% of those treated with placebo (p 0.01). AEs were not reported. A more recent study by Cohen et al % oxygen at 12 L/min for 15 minutes vs regular 468 Neurology 75 August 3, 2010

7 air for the acute treatment of CH. Patients treated 4 attacks, 2 with each treatment, in a crossover fashion. The primary endpoint was pain-free response at 15 minutes. Patients were pain-free at 15 minutes in 78% of attacks treated with oxygen, compared to 20% of attacks treated with air (p 0.001). No serious adverse events occurred. One hundred percent oxygen should be offered for acute treatment of CH (Level A). Cocaine/lidocaine. One Class II RCT cocaine intranasal vs lidocaine intranasal and placebo. 12 The study was a parallel-group design with 9 patients receiving mg of either10% cocaine hydrochloride, 10% lidocaine, or placebo intranasally for 1 nitroglycerin-induced headache. The primary endpoint was the time elapsed until obtaining pain relief. On average, cessation of pain occurred after 31.3 minutes for cocaine, 37.0 minutes for lidocaine, and 59.3 minutes for saline (p 0.01 for both study medications). The most common side effects were nasal congestion following study medication administration as well as unpleasant taste of lidocaine. Intranasal cocaine and lidocaine may be considered for acute treatment of CH (Level C). Octreotide. One Class II crossover study evaluated subcutaneous octreotide in 57 patients for the treatment of 2 CH attacks. 13 The primary endpoint was headache response at 30 minutes. A total of 52% of patients treated with octreotide reported headache relief in 30 minutes, compared to 36% of patients treated with placebo (p 0.007). Eight patients taking octreotide reported diarrhea, abdominal bloating, or nausea. Other AEs included a dull background headache, lethargy, and injection-site reactions. Octreotide may be considered for acute treatment of CH (Level C). Dihydroergotamine, ergotamine, somatostatin, and prednisone. There are single Class III RCTs investigating dihydroergotamine NS, 14 IM ergotamine, 15 IV somatostatin, 15 and prednisone 30 for the acute treatment of CH. While all of these trials reported symptomatic benefit, these studies had important methodologic limitations, preventing evidence-based advice to be given. There is insufficient evidence to advise the use of dihydroergotamine, ergotamine, somatostatin, and prednisone for the acute treatment of CH (Level U). Preventive treatment. Which preventive treatments for CH are effective in reducing headache frequency, duration, and pain? See table 4 for a summary of suggestions. Civamide. One Class I RCT investigated civamide intranasal vs placebo 16 for cluster prophylaxis. A total of 28 patients received either 100 L of 0.025% civamide into each nostril daily or placebo for 7 days. Patients were observed for 20 days posttreatment. The primary endpoint was change in frequency of CH attacks per week during the observation period. For the first 7 days of observation, patients taking civamide had a greater decrease in the number of headaches compared to the placebo group ( 55.5% vs 25.9%; p 0.03) and a greater decrease for the entire posttreatment period that approached significance ( 61.4% vs 30.9%; p 0.054). The most common reported AEs were nasal burning (78%), lacrimation (50%), pharyngitis (44%), and rhinorrhea (33%). Intranasal civamide should be considered for the prevention of CH (Level B). Suboccipital steroid injection. There is 1 Class I RCT investigating suboccipital steroid injections vs placebo. 17 The study medication consisted of mg betamethasone dipropionate and 5.26 mg betamethasone disodium phosphate mixed with 0.5 ml Xylocaine 2%. A total of 23 patients participated in the study. The primary endpoint was the disappearance of CH attacks within 72 hours. A total of 85% of patients in the study group had relief of attacks by 72 hours, compared to 0% of patients in the placebo group (p ). Two patients reported transient pain at the injection site. Suboccipital steroid injection should be considered for the prevention of CH (Level B). Sumatriptan, sodium valproate. There are single Class I RCTs on the use of oral sumatriptan 100 mg 3 times 18 daily and sodium valproate 500 mg 19 vs placebo for the prevention of CH. Neither of these treatments resulted in a significant decrease in CH attacks. Sumatriptan and sodium valproate are not advised for the prevention of CH (Level B). Lithium. There is 1 Class II RCT comparing lithium 800 mg daily to placebo 20 in patients with episodic CH, and one Class II RCT comparing lithium 900 mg daily to verapamil 24 in patients with chronic CH. The parallel-group study comparing lithium to placebo the percentage of patients in whom attacks ceased within 1 week as the primary outcome. A secondary endpoint was the percentage of patients who were substantially better subjectively within 1 week. There was no difference in the percentage of patients having cessation of attacks in the lithium group (15%) compared to the placebo group (14%). More patients, however, felt substantially better in the lithium group (70%) than in the placebo group (43%), though significance was not calculated. The common AE reported by patients was polyuria. Bussone et al. 24 compared verapamil 360 mg daily for 8 weeks to lithium 900 mg daily in a Class II study. Thirty patients with chronic CH Neurology 75 August 3,

8 Table 4 Summary of advisements for preventive treatment of cluster headache Maneuver Effectiveness Levels of evidence Advice Civamide Direct evidence that civamide 100 Lis effective in improving headache response in CH. Nonserious adverse events: nasal burning, lacrimation, pharyngitis, rhinorrhea. Level B: should be considered for the prevention of CH Suboccipital steroid injection Direct evidence that long- and rapid-acting steroids 2.5 ml are effective in improving headache response in CH. Nonserious adverse event: transient injection site pain. Level B: should be considered for the prevention of CH Sodium valproate Direct evidence that sodium valproate 500 mg is not effective in improving headache response in CH. Adverse events not reported. Level B: not advised for the prevention of CH Sumatriptan Direct evidence that sumatriptan 100 mg is not effective in improving headache events: nausea, vomiting, headache, malaise. Level B: not advised for the prevention of CH Melatonin Evidence that melatonin 10 mg is effective in improving headache response in CH. Nonserious adverse events: none reported. I Level C: may be considered for the prevention of CH Verapamil Evidence that verapamil 360 mg is effective in improving headache response in CH. Nonserious adverse events: constipation, reduced blood pressure, reduced heart rate. clinical trials; 1 AAN Class II, 1 AAN Class III Level C: may be considered for the prevention of CH Cimetidine/ chlorpheniramine Evidence that cimetidine 2,000 mg and chlorpheniramine 20 mg are not effective in improving headache response in CH. Nonserious adverse events: transient, erythematous skin rash. Other adverse events not reported. clinical trials; 2 AAN Class II Level C: not advised for the prevention of CH Lithium Evidence that lithium 900 mg is effective in improving headache response in CH. Nonserious adverse event: polyuria. clinical trials; 2 AAN Class II Level C: may be considered for the prevention of CH Misoprostol Evidence that misoprostol 300 gisnot effective in improving headache response in CH. Adverse events not reported. I Level C: not advised for the prevention of CH Oxygen Evidence that 100% hyperbaric oxygen is not effective in improving headache response in CH. Adverse events not reported. I Level C: not advised for the prevention of CH Capsaicin Insufficient evidence that capsaicin is effective in improving headache response in CH. Adverse events not reported. to advise for the prevention of CH Nitrate tolerance Insufficient evidence that nitrate tolerance via 5-ISMN 30 mg is effective in improving headache response in CH. Adverse events not reported. to advise for the prevention of CH Prednisone Insufficient evidence that prednisone 20 mg every other day is effective in improving headache response in CH. Adverse events not reported. to advise for the prevention of CH Abbreviations: AAN American Academy of Neurology; CH cluster headache. participated in this crossover study, and outcomes included the intensity, frequency, and duration of attacks during the trial period. A total of 50% of patients in the verapamil group and 37% of those in the lithium group experienced an improvement in the headache index, compared to the run-in period (p 0.01). Lithium may be considered for the prevention of CH (Level C). Melatonin. There is 1 Class II RCT on melatonin for cluster prevention. 21 This placebo-controlled, parallel-group trial of 20 patients investigated melatonin 10 mg daily vs placebo for 2 weeks. In comparison to the run-in period, there was a reduction in daily headache frequency in the melatonin group (p 0.03), but not the placebo group. Adverse events were not reported. Melatonin may be considered for the prevention of CH (Level C). Misoprostol, 100% hyperbaric oxygen. There are single Class II RCTs evaluating misoprostol 22 and 100% hyperbaric oxygen 23 for the prevention of CH. Neither treatment resulted in a significant decrease in CH attacks. 470 Neurology 75 August 3, 2010

9 Misoprostol and 100% hyperbaric oxygen are not advised for the prevention of CH (Level C). Verapamil. There are 2 RCTs investigating the use of verapamil as a preventive medication for CH. 24,25 Bussone et al. 24 compared verapamil 360 mg daily for 8 weeks to lithium 900 mg daily in an AAN Class II study (results reported in Lithium section). Leone et al. 25 studied verapamil 360 mg daily vs placebo for 2 weeks in 30 patients in an AAN Class III study. The primary endpoint was the reduction in attack frequency per week. Verapamil was found to be superior to placebo, with patients taking verapamil experiencing 0.6 attacks per day, compared to 1.65 per day in the placebo group (p 0.001). Reported AEs included constipation, reduced blood pressure, and reduced heart rate. Verapamil may be considered for the prevention of CH (Level C). Cimetidine/chlorpheniramine. Two Class III RCTs have investigated the effect of cimetidine, an H2 antagonist, vs placebo. 26,27 One of these studies additionally looked at chlorpheniramine, an H1 antagonist, vs placebo. Neither of these studies found any benefit from the use of these treatments for the prevention of cluster attacks. Cimetidine and chlorpheniramine are not advised for the prevention of CH (Level C). Capsaicin, nitrate tolerance, prednisone. There are single Class III RCTs on the use of capsaicin intranasal cream, 28 nitrate tolerance, 29 and oral prednisone 30 for the prevention of CH. While the capsaicin study reported less severe headaches in the treatment group and the prednisone study reported more frequent attacks in the placebo group, all studies had important methodologic limitations. There is insufficient evidence to advise the use of intranasal capsaicin, nitrate tolerance, and prednisone for the prevention of CH (Level U). DISCUSSION The treatment of CH is largely pharmacologic, although behavioral modification can play a role in some patients, and surgical treatment options are available for patients who do not respond satisfactorily to drug treatment. Not all medications which are used widely for CH based on clinical experience have been adequately studied using modern clinical trial methodology. This review focused exclusively on double-blind RCTs as such trials provide the best evidence for therapeutic efficacy. Based on the quality of published trials, verapamil receives a Grade C rating. It is important to note, however, that verapamil is generally considered to be the mainstay of CH preventive therapy. 32 Clinical experience with both verapamil and lithium as preventive treatments for CH is extensive. While both agents only received Grade C ratings due to trial quality, they are used much more routinely in clinical practice than civamide, which, as a newer treatment, has only been evaluated in 1 trial. The evidence for the efficacy of lithium for the prevention of CH is stronger for patients with chronic CH than those with episodic CH. While side effects reported in both trials were minimal, 20,24 long-term use of lithium has been associated with tremor, hypothyroidism, and nephrogenic diabetic insipidus. Monitoring of serum lithium levels and thyroid and renal function is recommended in patients on long-term therapy. 33 The serotonin inhibitor methysergide has been recommended for the prevention of CH in previous guidelines, 34 although no placebo-controlled, double-blind studies of this treatment have been. We have not advised methysergide for several reasons in addition to the lack of controlled studies. Methysergide is not available in the United States, and has been associated with the development of pulmonary and retroperitoneal fibrosis, and therefore treatment must be limited to 6 months. 35,36 In addition, methysergide should not be coadministered with triptans or ergots, 37 and sumatriptan and zolmitriptan are standard symptomatic treatments for CH. We recognize that there may be special clinical scenarios in which the use of methysergide may be appropriate but believe that general endorsements for use of this drug should be avoided. The use of a short course of steroids to stop a cluster bout is common and, based upon clinical experience, effective. It is recommended in previous guidelines. 34 However, aside from the trial that examined the efficacy of suboccipital steroid injection, 17 only 1 Class III RCT of steroids in CH has been performed. 30 Conceptually, one can consider some of the medications used in CH as transitional preventive treatments. Transitional treatments are medications that are started with longer-term preventive drugs. Their function is to stop the CH attacks almost immediately, and to maintain this pain relief until the dose of the longer-term preventive drug can be increased and becomes effective. 38 Most transitional therapies have not been subjected to rigorous clinical trials, but are commonly used in clinical practice. They include oral steroids (prednisone 60 mg daily for 3 days, then decreased by 10 mg every 3 days for a total of 18 days of therapy) 39 and dihydroergotamine (1 mg SC/IM twice a day for several days). 39 Ergotamine tartrate 1 or 2 mg given once daily or in divided doses has also been used for transitional therapy. 39 Suboccipital steroid injections might also be considered transitional therapy. 38,39 Transitional therapies are especially ap- Neurology 75 August 3,

10 propriate for patients who present with a high frequency of attacks. Finally, in refractory patients prophylactics are often used in combination. For example, verapamil prophylaxis may be combined with lithium in an attempt to control intractable cluster headaches 32 even though there is no evidence base for this practice. CONCLUSION For the acute treatment of CH attacks, Level A advice can be given for the use of sumatriptan SC 6 mg, intranasal zolmitriptan 5 and 10 mg, and 100% oxygen. Level B advice can be given for the use of intranasal sumatriptan 20 mg and oral zolmitriptan 5 and 10 mg. Level C advice can be given for intranasal 10% cocaine, intranasal 10% lidocaine, and subcutaneous octreotide 100 g. For the prevention of CH attacks during a cluster bout, Level B advice can be given for intranasal civamide 100 L and suboccipital steroid injection. Level C advice can be given for melatonin 10 mg, verapamil 360 mg, and lithium 900 mg. Several new treatments are emerging for the prevention of CH. Open-label and pilot studies have been conducted with topiramate 40 and gabapentin. 41 Deep brain stimulation and occipital nerve stimulation 45 have also been with small trials in medically refractory patients. In the future, there may be a greater number of evidence-based treatment options for the prevention of CH. When patients present with a new cluster bout, it is appropriate to initiate both acute symptomatic therapy and preventive therapy. Choosing between recommended treatments should be based on headache frequency, patient comorbidities, and side effects. AUTHOR CONTRIBUTIONS Statistical analysis was conducted by Dr. Tamara Pringsheim. DISCLOSURE Dr. Francis reports no disclosures. Dr. Becker has served on medical advisory boards for Merck Serono, Pfizer Inc., Allergan, Inc., AGA Medical Corporation, and Teva Pharmaceutical Industries Ltd.; has been a local PI for clinical trials for Merck Serono, Allergan, Inc., Medtronic, Inc., and AGA Medical Corporation; and has received speaker honoraria from Merck Serono, AGA Medical Corporation, and Pfizer Inc. Dr. Pringsheim has served on a scientific advisory board for Pfizer Inc. and has received speaker honoraria from Shire Canada, Pfizer Inc., and Teva Pharmaceutical Industries Ltd. Received December 29, Accepted in final form April 22, REFERENCES 1. Headache Classification Subcommittee of the International Headache Society. The International Classification of Headache Disorders, 2nd edition. Cephalalgia 2004; 24(suppl 1): Bussone G. Cluster headache: from treatment to pathophysiology. Neurol Sci 2008;29(suppl 1):S1 S6. 3. Pringsheim T. Cluster headache: evidence for a disorder of circadian rhythm and hypothalamic function. Can J Neurol Sci 2002;29: Edlund W, Gronseth G, So Y, Franklin G. Clinical Practice Guideline Process Manual. St. Paul: American Academy of Neurology; 2004: Ekbom K, Monstad I, Prusinski A, Cole JA, Pilgrim AJ, Noronha D. Subcutaneous sumatriptan in the acute treatment of cluster headache: a dose comparison study. Acta Neurol Scand 1993;88: The Sumatriptan Cluster Headache Study Group. Treatment of acute cluster headache with sumatriptan. N Engl J Med 1991;325: van Vliet JA, Bahra A, Martin V, et al. Intranasal sumatriptan in cluster headache. Neurology 2003;60: Rapoport AM, Mathew NT, Silberstein SD, et al. Zolmitriptan nasal spray in the acute treatment of cluster headache. Neurology 2007;69: Cittadini E, May A, Straube A, Evers S, Bussone G, Goadsby PJ. Effectiveness of intranasal zolmitriptan in acute cluster headache. Arch Neurol 2006;63: Bahra A, Gawel MJ, Hardebo JE, Millson D, Breen SA, Goadsby PJ. Oral zolmitriptan is effective in the acute treatment of cluster headache. Neurology 2000;54: Fogan L. Treatment of cluster headache: a double blind comparison of oxygen v. air inhalation. Arch Neurol 1985; 42: Costa A, Pucci E, Antonaci F, et al. The effect of intranasal cocaine and lidocaine on nitroglycerin-induced attacks in cluster headache. Cephalalgia 2000;20: Matharu MS, Levy MJ, Meeran K, Goadsby PJ. Subcutaneous octreotide in cluster headache: randomized placebocontrolled double-blind crossover study. Ann Neurol 2004;56: Andersson PG, Jespersen LT. Dihydroergotamine nasal spray in the treatment of attacks of cluster headache. Cephalalgia 1986;6: Sicuteri F, Geppetti P, Marabini S, Lembeck F. Pain relief by somatostatin in attacks of cluster headache. Pain 1984; 18: Saper JR, Klapper J, Mathew NT, Rapoport A, Phillips SB, Bernstein JE. Intranasal civamide for the treatment of episodic cluster headaches. Arch Neurol 2002;59: Ambrosini A, Vandenheede M, Rossi P, et al. Suboccipital injection with a mixture of rapid- and long-acting steroids in cluster headache: a double-blind placebo-controlled study. Pain 2005;118: Monstad I, Krabbe A, Micieli G, et al. Preemptive oral treatment with sumatriptan during a cluster period. Headache 1995;35: El Amrani M, Massiou H, Bousser MG. A negative trial of sodium valproate in cluster headache: methodological issues. Cephalalgia 2002;22: Steiner TJ, Hering R, Couturier EGM, Davies PTG, Whitmarsh TE. Double-blind placebo-controlled trial of lithium in episodic cluster headache. Cephalalgia 1997;17: Leone M, D Amico D, Moschiano F, Fraschini F, Bussone G. Melatonin versus placebo in the prophylaxis of cluster headache: a double-blind pilot study with parallel groups. Cephalalgia 1996;16: Neurology 75 August 3, 2010

11 22. Evers S, Masur H, Sörös P, Brilla R, Husstedt IW. Prostaglandin analog mechanisms are not effective in refractory chronic cluster headache. Headache 1998;38: Nilsson Remahl AIM, Ansjön R, Lind F, Waldenlind E. Hyperbaric oxygen treatment of active cluster headache: a double-blind placebo-controlled cross-over study. Cephalalgia 2002;22: Bussone G, Leone M, Peccarisi C, et al. Double blind comparison of lithium and verapamil in cluster headache prophylaxis. Headache 1990;30: Leone M, D Amico D, Frediani F, Moschiano F, Grazzi L, Attanasio A. Verapamil in the prophylaxis of episodic cluster headache: a double-blind study versus placebo. Neurology 2000;54: Russell D. Cluster headache: trial of a combined histamine H1 and H2 antagonist treatment. J Neurol Neurosurg Psychiatry 1979;42: Veger T, Russell D, Sjaastad O. Histamine H2 antagonists and cluster headache. BMJ 1976;4: Marks DR, Rapoport A, Padla D, et al. A double-blind placebo-controlled trial of intranasal capsaicin for cluster headache. Cephalalgia 1993;13: Christiansen I, Iversen HK, Olesen J. Induction of nitrate tolerance is not a useful treatment in cluster headache. Cephalalgia 2000;20: Jammes JL. The treatment of cluster headaches with prednisone. Dis Nerv Syst 1975;36: Cohen A, Burns B, Goadsby P. High flow oxygen for treatment of cluster headache. JAMA 2009;302: Tfelt-Hansen P, Tfelt-Hansen J. Verapamil for cluster headache: clinical pharmacology and possible mode of action. Headache 2009;49: Thomson Reuters. Lithium. In: Micromedex 2.0. Available at: Accessed March 2, May A, Leone M, Afra J, et al. EFNS guidelines on the treatment of cluster headache and other trigeminalautonomic cephalalgias. Eur J Neurol 2006;13: Graham J, Suby H, LeCompte P, Sadowsky N. Fibrotic disorders associated with methysergide therapy for headache. N Engl J Med 1966;270: Muller R, Weller P, Chemaissani A. Pleural fibrosis as a side effect of years long methysergide therapy. Dtsch Med Wochenschr 1991;116: Thomson Reuters. Methysergide. In: Micromedex 2.0. Available at: Accessed March 2, Peres MFP, Stiles MA, Siow HC, Rozen TD, Young WB, Silberstein SD. Greater occipital nerve block for cluster headache. Cephalalgia 2002;22: Gladstone JP, Dodick DW. Chronic daily Headache. In: Johnson RT, Griffin JW, McArthur JC, eds. Current Therapy in Neurologic Disease, 7th edition. Philadelphia: Mosby Elsevier; Wheeler SD, Carrazana EJ. Topiramate-treated cluster headache. Neurology 1999;53: Schuh-Hofer S, Israel H, Neeb L, Reuter U, Arnold G. The use of gabapentin in chronic cluster headache patients refractory to first-line therapy. Eur J Neurol 2007;14: Schoenen J, Di Clemente L, Vandenheede M, et al. Hypothalamic stimulation in chronic cluster headache: a pilot study of efficacy and mode of action. Brain 2005;128: Leone M, Franzini A, Broggi G, Bussone G. Hypothalamic stimulation for intractable cluster headache: long term experience. Neurology 2006;67: Leone M, Proiette Cecchini A, Franzini A, et al. Lessons from 8 years experience of hypothalamic stimulation in cluster headache. Cephalalgia 2008;28: Burns B, Watkins L, Goadsby PJ. Treatment of intractable cluster headache by occipital nerve stimulation in 14 patients. Neurology 2009;72: Neurology 75 August 3,