American Headache Society Guideline

Transcription

1 ISSN Headache doi: /head VC 2016 American Headache Society Published by Wiley Periodicals, Inc. American Headache Society Guideline Treatment of Cluster Headache: The American Headache Society Evidence-Based Guidelines Matthew S. Robbins, MD; Amaal J. Starling, MD; Tamara M. Pringsheim, MD; Werner J. Becker, MD; Todd J. Schwedt, MD Background. Cluster headache (CH), the most common trigeminal autonomic cephalalgia, is an extremely debilitating primary headache disorder that is often not optimally treated. New evidence-based treatment guidelines for CH will assist clinicians with identifying and choosing among current treatment options. Objectives. In this systematic review we appraise the available evidence for the acute and prophylactic treatment of CH, and provide an update of the 2010 American Academy of Neurology (AAN) endorsed systematic review. Methods. Medline, PubMed, and EMBASE databases were searched for double-blind, randomized s that investigated treatments of CH in adults. Exclusion and inclusion criteria were identical to those utilized in the 2010 AAN systematic review. Results and Recommendations. For acute treatment, sumatriptan subcutaneous, zolmitriptan nasal spray, and high flow oxygen remain the treatments with a Level A recommendation. Since the 2010 review, a study of sphenopalatine ganglion stimulation was added to the current guideline and has been administered a Level B recommendation for acute treatment. For prophylactic therapy, previously there were no treatments that were administered a Level A recommendation. For the current guidelines, suboccipital steroid injections have emerged as the only treatment to receive a Level A recommendation with the addition of a second Class I study. Other newly evaluated treatments since the 2010 guidelines have been given a Level B recommendation (negative study: deep brain stimulation), a Level C recommendation (positive study: warfarin; negative studies: cimetidine/chlorpheniramine, candesartan), or a Level U recommendation (frovatriptan). Conclusions. This AHS guideline can be utilized for understanding which therapies have superiority to placebo or sham treatment in the management of CH. In clinical practice, these recommendations should be considered in concert with other variables including safety, side effects, patient preferences, clinician experience, cost, and the invasiveness of the intervention. Given the lack of Class I evidence and Level A recommendations, particularly for a number of commonly used preventive therapies, further studies are warranted to demonstrate safety and efficacy for established and emerging therapies. Key words: cluster, headache, episodic, chronic, guidelines, evidence-based medicine, neurostimulation, sphenopalatine, calcium-channel blockers, verapamil, oxygen, lithium (Headache 2016;56: ) From the Montefiore Headache Center, Albert Einstein College of Medicine, Bronx, NY, USA (M.S. Robbins); Mayo Clinic, Phoenix, AZ, USA (A.J. Starling and T.J. Schwedt); University of Calgary, Calgary, Alberta, Canada (T.M. Pringsheim and W.J. Becker). Address all correspondence to T.J. Schwedt, MD, Mayo Clinic, 5777 East Mayo Blvd, Phoenix, AZ 85054, USA. Accepted for publication May 10, Conflict of Interest: Dr. Robbins has received honoraria for educational activities with the American Headache Society, Springer, and book royalties from Wiley. Dr. Schwedt has received consulting fees from Amgen, Dr. Reddy s, GBS, Supernus, and Teva. Dr. Schwedt owns stock in GBS and Second Opinion. Dr. Schwedt also has received royalties from UpToDate and Cambridge University Press. Drs. Starling, Pringsheim, and Becker report no conflicts of interest. 1093

2 1094 July/August 2016 Cluster headache (CH), the most common of the family of headache disorders known as the trigeminal autonomic cephalalgias, 1 has a lifetime prevalence exceeding one in CH is considered among the most severe of the primary headache disorders because of extreme pain, associated autonomic symptoms, and high attack. In addition, a sizeable proportion of patients with CH have, 2 which features only brief or no remission periods 1 and may be particularly refractory to medical therapies. Evidence suggests that CH is not easily recognized, leading to diagnostic delay. Furthermore, patients with CH are often suboptimally treated, an especially disappointing fact considering there are treatments available for aborting and preventing attacks. 3,4 Treatment of CH in accordance with guidelines is associated with better outcomes. 5 The objective of this systematic review of the literature is to provide evidence-based guidelines for the abortive and preventive treatment of CH. To do so, we updated the 2010 American Academy of Neurology (AAN) endorsed systematic review. 6 The interval of time that has elapsed since completion of the 2010 literature review necessitates these updated recommendations. 7-9 These American Headache Society (AHS) guidelines provide an updated assessment of available data and also include non-medication treatments. Akin to the AAN endorsed systematic review, these guidelines aim to answer the following clinical questions: 1. Which treatments are for the acute treatment of CH attacks? 2. Which treatments are for reducing the of CH attacks? These AHS guidelines rate the strength of evidence for CH treatment efficacy. Of course, efficacy must be considered in conjunction with side effects, risks, patient preferences, cost, and other factors when making treatment recommendations. METHODS Authorship Committee. For development of these guidelines, the AHS Guidelines Committee assembled a panel of AHS members with expertise in CH and guideline development. Consistent with AHS policy, the guideline development process adhered to the recommendations included in the AAN Clinical Practice Guideline Process Manual, 10 though for this review the evidence ratings and recommendations reflected the same methodology employed in the previous systematic review. Other than providing a meeting room for panel members to meet during the 2015 Scottsdale Headache Symposium, no material support was provided by the AHS. In addition, no outside funding was used for guideline development. The AHS Guideline Committee and AHS Board of Directors reviewed and approved this guideline. Search Strategy. With the assistance of a medical research librarian, Medline, PubMed, and EMBASE databases were searched for doubleblind, randomized s that investigated treatments of CH. The search terms and strategy are shown in Supplemental Appendix 1. Inclusion and Exclusion Criteria. To be included, studies must have been double-blind, randomized trials of any treatment vs placebo or sham or vs another treatment studying adult participants (at least 18 years of age). Studies of episodic CH,, or a combination were included. Studies were excluded if they were not doubleblinded, if there was no randomization, if a treatment or placebo/sham comparator was not used, and if children were studied. Research published between 1950 and April 15, 2015 was considered for inclusion. Since these AHS guidelines are an update of the 2010 AAN systematic review, data from publications previously considered for inclusion in 2010 (whether or not they were ultimately included or excluded) were not reanalyzed for efficacy during development of these guidelines. However, included studies in those guidelines were reanalyzed for safety and tolerability to align with the current data abstraction tool as the 2010 AAN systematic review focused on assessing efficacy. To ensure assessment and potential inclusion of contemporary publications, the database search was reexecuted prior to manuscript finalization (on November 24, 2015), accompanied by headache subspecialty journal (Headache, Cephalalgia,

3 Headache 1095 Journal of Headache and Pain) hand-searching of electronically published manuscripts ahead of print by one reviewer (MSR). Methods of Review. Thetitlesandabstractsof all publications resulting from the search were independently screened by two reviewers (MSR and AJS) who indicated articles for exclusion or for further review. When these two reviewers provided conflicting recommendations, the article was included for full review. The same two reviewers each independently reviewed full manuscripts; these reviews were utilized for further determination of inclusion or exclusion. For those publications included for further review, the full manuscript was utilized for abstracting the study methods and results, using standardized data collection forms (Supplemental Appendix 2). A third panel member (TJS) compared and contrasted the data abstracted by each reviewer and identified discrepancies. When discrepancies existed, consensus was reached after discussion among the three reviewers (MSR, AJS, and TJS). Rating the Evidence. The quality of the evidence provided from each study was determined according to theaanqualitycriteriautilizedinthe2010systematic review. 6 Studies are rated Class I if all of the following criteria are met: (a) allocation concealment, (b) primary outcome(s) clearly defined, (c) inclusion/ exclusion criteria clearly defined, (d) adequate accounting for drop-outs and cross-overs with numbers sufficient to have minimal potential for bias, and (e) relevant baseline characteristics are presented and substantially equivalent among treatment groups or there is appropriate adjustment for differences. A Class II study is a prospective matched group cohort study in a representative population with masked outcome assessment that meets (a) to (e) above, or a randomized (RCT) in a representative population that lacks one criteria of (a) to (d). A Class III study is a including well-defined natural history controls or patients serving as own controls in a representative population, where outcome is independently assessed, or independently derived by objective outcome measurement, or an RCT that lacks two criteria of (a) to (d). Class IV studies include uncontrolled studies, case series, case reports, or expert opinion. Making Recommendations. Recommendations for the treatment of CH were made according to AAN grades for classification utilized in the 2010 systematic review 6 : A 5 Established as, in, or harmful (or established as useful/predictive or not useful/predictive) for the given condition in the specified population. (Level A rating requires at least two consistent Class I studies.)* B 5 Probably, in, or harmful (or probably useful/predictive or not useful/predictive) for the given condition in the specified population. (Level B rating requires at least one Class I study or at least two consistent Class II studies.) C 5 Possibly, in, or harmful (or possibly useful/predictive or not useful/predictive) for the given condition in the specified population. (Level C rating requires at least one Class II study or two consistent Class III studies.) U 5 Data inadequate or conflicting; given current knowledge, treatment (test, predictor) is unproven. (Studies not meeting criteria for Class I- Class III.) *In exceptional cases, one convincing Class I study may suffice for an A recommendation if (1) all criteria are met, (2) the magnitude of effect is large (relative rate improved outcome >5 and the lower limit of the confidence interval is <2). Medications for which no new evidence became available since the publication of the 2010 AAN systematic review received the same ratings as those given in the 2010 AAN systematic review. Medications for which new evidence was available received updated ratings. Since nonmedication treatments were not included in the 2010 AAN document, all ratings and recommendations regarding these treatments were based entirely upon the results of this review. RESULTS Figure 1 demonstrates the flowchart for the search results. The initial search yielded 1527 publications. After review of the titles and abstracts of

4 1096 July/August 2016 Fig. 1. PRISMA flow chart demonstrating the identification, screening, and review of studies for the treatment of cluster headache. these publications, 10 were ultimately included for full manuscript review. Once fully reviewed, seven manuscripts met all inclusion and exclusion criteria and the data from these seven manuscripts were utilized for updating the 2010 AAN systematic review for creation of these AHS guidelines. These publications described treatment with: sphenopalatine ganglion stimulation (n 5 1), warfarin (n 5 1), frovatriptan (n 5 1), candesartan (n 5 1), cimetidine (n 5 1), suboccipital steroid injection/occipital nerve blocks (n 5 1), and deep brain stimulation (n 5 1). Which Treatments Are Effective for Aborting CH Attacks and Reducing CH Attack Severity? Since thepublicationofthe2010aanreview,thereareno new data from double-blind, controlled trials that contribute to determining the efficacy or safety of the following acute treatments: sumatriptan, zolmitriptan, oxygen, octreotide, dihydroergotamine, prednisone, cocaine/lidocaine, and somatostatin. Thus, the assessment of the data and recommendations regarding these treatments have been carried forward from the 2010 AAN assessment. New data from a double-blind, are available for the acute treatment of cluster with sphenopalatine ganglion stimulation (Table 1). Sphenopalatine Ganglion Stimulation. In a Class I study, patients with were all randomized to receive full, subperception, and sham stimulation after surgical implantation of a sphenopalatine ganglion stimulator (Autonomic Technologies, Inc. [ATI] Neurostimulator). Full stimulation parameters were titrated for each individual patient. Acute pain relief at 15 min was achieved in 67.1% of full stimulation-treated attacks at 15 min vs 7.4% of sham stimulation treated attacks (P <.0001). Device/procedural adverse effects occurred in 15.6%, including three lead revisions and two device explants. Most patients (81%) experienced loss of sensation in the distribution of the maxillary nerve, with one patient experiencing persistent paresthesias in the same distribution. Infection (6.3%), mild focal facial paresis (6.3%), and maxillary sinus puncture (6.3%) were also reported. Which Treatments Are Effective for Reducing the Frequency of CH Attacks? Since the publication of the 2010 AAN evidence assessment, there are no

5 Headache 1097 new data from double-blind, controlled trials that contribute to determining the efficacy or safety of the following preventive treatments: civamide, sodium valproate, sumatriptan, melatonin, verapamil, lithium, misoprostol, hyperbaric oxygen, capsaicin, nitrate tolerance, and prednisone. Thus, Table 1. Efficacy, Adverse Events, Level of Evidence and Recommendations for the Acute Therapy of Cluster Headache Treatment Study Population Efficacy Adverse Events Level of Evidence Recommendation Positive evidence Sumatriptan (subcutaneous) Zolmitriptan (nasal spray) Oxygen Sumatriptan (nasal spray) Zolmitriptan (oral) Sphenopalatine ganglion stimulation* Cocaine /lidocaine (nasal spray) Octreotide (subcutaneous) Chronic CH dissatisfied with current treatment Sumatriptan 6 mg is in Zolmitriptan 5 mg and 10 mg are in 100% oxygen 6-12 L/min is in improving headache Sumatriptan 20 mg is in Zolmitriptan 5 mg and 10 mg are in Sphenopalalatine ganglion stimulation is in 10% cocaine hydrochloride and 10% lidocaine are in Octreotide 100 mg is in Nonserious: injection site reactions, nausea and vomiting, dizziness, fatigue, paresthesias Nonserious: unpleasant taste, nasal cavity discomfort, somnolence, dizziness, nausea, throat/neck tightness Nonserious: bitter taste Nonserious: paresthesias, heaviness, asthenia, nausea, dizziness, chest pain Serious: lead revisions, device explants Nonserious: loss of sensation in the distribution of the maxillary nerve, infection, mild focal facial paresis, maxillary sinus puncture Nonserious: nasal congestion, unpleasant lidocaine taste Nonserious: injection site reactions, diarrhea, abdominal bloating, nausea, dull background headache, lethargy Two class I controlled trials Two class I controlled trials Two class I controlled trials One class I controlled trial One class I controlled trial One class I controlled trial One class II One class II Level A: established as Level A: established as Level A: established as Level B: probably Level B: probably Level B: probably

6 1098 July/August 2016 Table 1. Continued Treatment Study Population Efficacy Adverse Events Level of Evidence Recommendation Insufficient evidence Dihydroergotamine (nasal spray) Somatostatin Episodic CH ( typical periodic ) Insufficient evidence that dihydroergotamine 1 mg is in Insufficient evidence that somatostatin 25 mg is in improving headache Prednisone Not specified Insufficient evidence that prednisone 30 mg is in improving headache Nonserious: nausea without vomiting One class III One class III controlled trial One class III controlled trial Level U: insufficient evidence to make recommendation Level U: insufficient evidence to make recommendation Level U: insufficient evidence to make recommendation Recommendations are based only upon assessing the level of evidence for treatment efficacy. *Treatments for which new data were available since the 2010 AAN systematic review for the treatment of cluster headache and thus their ness, level of evidence, and recommendation were updated. Not routinely available in the United States. CH 5 cluster headache. the assessment of the data and recommendations regarding these treatments have been carried forward from the 2010 AAN assessment. New data from double-blind, controlled trials regarding the preventive treatment of cluster with the following treatments are available: warfarin, frovatriptan, candesartan, cimetidine/ chlorpheniramine, suboccipital steroid injections, and deep brain stimulation (Table 2). Warfarin. A Class II crossover study 12 of 34 patients with evaluated the use of warfarin with an International Normalized Ratio goal of 1.5 to 1.9 over a 12-week treatment period. In the intention to treat analysis, 50% underwent remission for 4 weeks during the warfarin period compared with 11.8% patients during the placebo period (P 5.004), associated with a relative risk of 0.57, absolute risk reduction of 0.38 (95% CI ), and a number needed to treat of 2.6 (95% CI ). Adverse effects were not serious; 5.9% developed epistaxis and 5.9% developed skin bruising, all while on warfarin, though without significance (P 5.113). Frovatriptan. A Class III parallel group study 13 of frovatriptan 5 mg daily for episodic CH prophylaxis fell short of its recruitment goal with only 10 patients completing the study. The mean attack per week during treatment was higher in the frovatriptan group than in the placebo group (14.1 vs 10.1) but failed to reach significance (P ). There were no adverse events in the active treatment group, and 6 adverse events in the placebo group, but the details were not reported. Candesartan. A Class II parallel group study 14 evaluated candesartan 16 mg daily for 1 week followed by 32 mg daily for 2 weeks in episodic CH prophylaxis with 32 patients completing the study. The study failed to meet the primary endpoint, with a mean reduction in attacks per week from week 1 to week 3 of for candesartan vs for placebo (61% vs 38% attack reduction, P 5.38). The subject dropout rate was 20%, and no intention to treat analysis was reported. Testing accounting for the temporal properties of the data (exact Poisson test), yielded significance

7 Headache 1099 Table 2. Efficacy, Adverse Events, Level of Evidence and Recommendations for the Prophylactic Therapy of Cluster Headache Treatment Study Population Efficacy Adverse Events Level of Evidence Recommendation Positive evidence Suboccipital steroid injection* Civamide (nasal spray) Lithium Verapamil Warfarin* Melatonin Episodic CH Refractory Suboccipital single injection or injection series with corticosteroids is in reducing attack 100 ml of 0.025% civamide in each nostril daily is in Lithium 900 mg is in Nonserious: transient injection site pain, headache Nonserious: nasal burning, lacrimation, pharyngitis, rhinorrhea Two class I s One class I Nonserious: polyuria Two class II s (one positive, one negative) Verapamil 360 mg daily Nonserious: constipation, reduced is in reducing attack blood pressure, reduced heart rate Warfarin daily with International Normalized Ratio goal of 1.5 to 1.9 is in Melatonin 10 mg every evening is in Insufficient evidence Frovatriptan* Episodic CH Insufficient evidence frovatriptan 5 mg daily is in reducing attack Capsaicin (intranasal) Nitrate tolerance Chronic cluster headache Insufficient evidence that capsaicin 0.025% cream applied twice daily is in Insufficient evidence that nitrate tolerance with 5-isosorbide mononitrate 30 mg three times daily is in reducing attack Prednisone Not specified Insufficient evidence that prednisone 20 mg every other day is in reducing attack Nonserious: epistaxis, skin bruising Nonserious: hypotension, headache One class II and one class III s One class II One class II One class III randomized One class III randomized One class III randomized One class III randomized Level A: established as Level B: probably Level U: insufficient evidence to make recommendation Level U: insufficient evidence to make recommendation Level U: insufficient evidence to make recommendation Level U: insufficient evidence to make recommendation

8 1100 July/August 2016 Table 2. Continued Treatment Study Population Efficacy Adverse Events Level of Evidence Recommendation Negative evidence Sodium valproate Sumatriptan Deep brain stimulation* Refractory Cimetidine/ Not specified chlorpheniramine* Misoprostol Oxygen (hyperbaric) Refractory Sodium valproate mg daily is not in reducing attack Sumatriptan 100 mg three times daily is not in reducing attack Unilateral hypothalamic deep brain stimulation is not in Cimetidine mg and chlorpheniramine mg are not in reducing attack Misoprostol 300 mg daily is not in 100% hyperbaric oxygen is not in Candesartan* Episodic CH Candesartan 32 mg daily is not in Nonserious: nausea and vomiting, somnolence Nonserious: nausea, vomiting, headache, malaise Serious: subcutaneous infection, preoperative loss of consciousness with hemiparesis after test stimulation, severe micturition syncope Nonserious: transient diplopia, change in hunger, change in libido Serious: prolonged rash Nonserious: transient, erythematous skin rash One class I One class I One class I Two class II and one class III s One class II randomized One class II randomized Nonserious: orthostatic One class II reactions, randomized dizziness, tiredness Level B: probably in Level B: probably in Level B: probably in in in in in Recommendations are based only upon assessing the level of evidence for treatment efficacy. *Treatments for which new data were available since the 2010 AAN systematic review for the treatment of cluster headache and thus their ness, level of evidence, and recommendation were updated. Steroid regimens included either a single suboccipital injection of short and long-acting betamethasone or a series of 3 suboccipital injections with cortivazol each performed 48 to 72 hours apart as an add-on therapy to verapamil. Not routinely available in the United States. CH 5 cluster headache. (P <.0001) but this was demonstrated in a post-hoc analysis. Only mild adverse events were reported with a similar in candesartan vs placebo groups (47.4% vs 38.5%, P 5.73). The most common adverse effects were orthostatic reactions or dizziness (four in candesartan group vs none in placebo group) and tiredness (eight in candesartan group vs three in placebo group). Cimetidine/Chlorpheniramine. In a Class III crossover study, 15 cimetidine 400 mg twice daily

9 Headache 1101 with chlorpheniramine 4 mg four times daily vs cimetidine vs placebo was evaluated in 22 patients with CH. Patients received each of the 3 treatments for 1 month. There was a 50% reduction of attacks in 20/40 treated with cimetidine with chlorpheniramine, 19/40 treated with cimetidine alone, and 19/40 treated with placebo. No tests of statistical significance were performed. Seven patients experienced adverse events (four during active treatment, three during placebo), one of which was potentially severe (prolonged rash). Suboccipital Steroid Injections. In a Class I parallel group study, 16 suboccipital injections with cortivazol 3.75 mg ipsilateral to the CH attack side were performed three times, each h apart for the shortterm prophylaxis of episodic and in comparison to the same injection series with the same volume using normal saline. The injections were performed as an add-on therapy to the contemporaneous initiation or escalation of verapamil in episodic CH patients, and prophylactic therapy of patients was maintained. Further addition of prophylactic therapies was permitted only after 15 days post-initial injection. Thirty-seven subjects completed the study, and 95% of the patients in the cortivazol group in comparison to 55% of controls had a reduction to 2 attacksperday during the second, third, and fourth days after the third injection (odds ratio 14.5, 95% CI , P 5.012). At day 15, the proportion of patients with a 50% attack reduction did not differ between patients receiving cortivazol group and controls. Remission rates at day 30 were much the same between groups, but cortivazol induced a 7-day remission at a median of 7 days earlier than did placebo. At day 30, the proportion of patients having 2 attacks a day was the same between groups. There were no serious adverse events; nonserious adverse events occurred similarly in the cortivazol vs control groups (86% vs 64%, P 5.162), most commonly featuring neck pain at the injection site and headache other than CH. Deep Brain Stimulation. Eleven patients with refractory completed a Class I crossover study 17 evaluating unilateral hypothalamic deep brain stimulation with a of 185 Hz and a pulse duration of 60 ms in comparison to sham stimulation. The voltage was individually adjusted by increasing the voltage by 3 V or 80% of the threshold producing side effects. Treatment periods lasted for 1 month and were separated by a 1-week washout period. The attacks/week treatment effect difference was similar between active and sham stimulation in the on-off group: 0.2 (95% CI ), as well as the difference between active and sham in the off-on group: 22.7 (95% CI ); P Three serious adverse events occurred: subcutaneous infection, preoperative loss of consciousness with hemiparesis after test stimulation, and severe micturition syncope. Nonserious adverse events (n 5 26) were all mild per the authors, some of which may have occurred during the open-label extension phase of the study. Adverse event rates were similar in both the active and sham stimulation groups. Review of Studies From the 2010 AAN Systematic Review for Safety and Tolerability. To enable reporting of adverse events, the 26 manuscripts fully reviewed in the 2010 AAN assessment were reviewed again. Tables 1 and 2 contain serious and nonserious adverse event descriptions where available for all therapies included in the 2010 guidelines and interim studies. DISCUSSION In this systematic review, we appraise the available evidence for the acute and prophylactic treatment of CH, providing an update of the 2010 AAN systematic review. This AHS assessment can be utilized as a guide for understanding which therapies have superiority to placebo or sham treatment in the management of CH. For acute treatment, sumatriptan subcutaneous, zolmitriptan nasal spray, and high flow oxygen remain the treatments with a Level A recommendation. Since the 2010 guidelines, a study of sphenopalatine ganglion stimulation 11 was added to the current guideline and has been administered a Level B recommendation. Currently, sphenopalatine ganglion stimulation is not readily available in the United States though is currently under investigation ( For maintenance and transitional prophylactic therapy, previously there were no treatments that

10 1102 July/August 2016 were administered a Level A recommendation. For the current guidelines, suboccipital steroid injections have emerged as the only treatment to receive a Level A recommendation with the addition of a second Class I study. 16 Other newly evaluated treatments since the 2010 guidelines have been given a Level B recommendation (negative study: deep brain stimulation), a Level C recommendation (positive study: warfarin; negative studies: cimetidine/ chlorpheniramine, candesartan), or a Level U recommendation (frovatriptan). This review does not provide exclusive guidance for the treatment of CH in specific patients. In such circumstances, factors other than level of evidence for efficacy, such as clinical experience, comorbidities, potential adverse event profile, latency of treatment effect, invasiveness of the therapy considered, coexisting medications used, and s to previous medication trials, influence therapeutic decisions. In addition, cost and availability, particularly for procedural treatments, are important variables. Some of the assessed therapies are not readily available in the United States, including sphenopalatine ganglion stimulation, intranasal cocaine/lidocaine, somatostatin, cortivazol, civamide, and hypothalamic deep brain stimulation. Our review also did not address treatment of CH in children, pregnant or lactating women, or the elderly, though to our knowledge clinical trials in these populations are lacking. Verapamil is generally regarded as the maintenance prophylactic therapy of choice for CH despite only a Level C recommendation by this review and its predecessor, 6 which generated commentary highlighting the discordance between the evidence base and expert opinion in CH therapy. 18 However, since suboccipital corticosteroid injections are typically considered only for transitional prophylaxis, lithium (two class II RCTs) and verapamil (one class II and one class III RCTs) have the highest evidence base among the maintenance preventive therapies available in the US. In other guidelines, verapamil is recommended as a first-line preventive therapy for CH One study that was included in the 2010 systematic review compared the onset latency of verapamil vs lithium and observed a shorter latency period with verapamil, which may confer a major advantage. 22 This guideline does not directly provide guidance for one prophylactic treatment in comparison to another for its rapidity to suppress an episodic CH attack period or to induce a remission of. In clinical practice, some prophylactic therapies may have shorter onset latencies, a treatment effect on the order of 1-2 weeks, and are considered transitional prophylactic therapies (ie, short-term or bridge therapies). Transitional treatments are often utilized in concert with an oral maintenance preventive medication with the intent of achieving short-term symptom improvement or potentially inducing a remission of the CH attack period or exacerbation. 6 The use of corticosteroids may serve this purpose when administered either by suboccipital nerve injection or orally. In one of the suboccipital steroid injection studies reviewed, this therapy was administered together with the initiation or escalation of verapamil, echoing what is often performed in clinical practice. 16 Oral corticosteroids such as prednisone are used commonly for transitional prophylaxis and are recommended as first or second line treatment by other guidelines, but have Level U evidence for prophylaxis per our review. However, an RCT is currently underway that evaluates prednisone as an add-on therapy to verapamil in the treatment of episodic CH that may help reinforce the utility of contemporaneous treatment with transitional and maintenance prophylactic therapies. 23 Warfarin is the only medication that has been administered a new positive recommendation (Level C). The rationale for studying this treatment for CH 12 was derived from anecdotal reports of incidental CH benefit in patients taking warfarin for other indications. 24,25 The study author hypothesized that the potential benefit of warfarin in CH treatment relates to the impact of vitamin K antagonism on dendritic and neuronal metabolism, circadian rhythms of the hypothalamus, and nitric oxide mediated neurogenic inflammation. We agree with the study conclusions of the author, who advocated for larger RCTs that would further elucidate the possible efficacy of warfarin for treating CH, and if

11 Headache 1103, would help to determine the optimal level and duration of anticoagulation. Collection of longer-term safety data would also be imperative since this therapy could feature major bleeding as a serious adverse effect. Therefore, such a treatment should only be studied in patients with refractory. Conversely, melatonin is a prophylactic treatment that features the same Level C recommendation, and clinicians may have a much lower threshold for its use because of its potentially favorable adverse effect profile. Suboccipital (or greater occipital) injections have long been employed in the treatment of various headache disorders 26 and are widely used by headache specialists. 27 However, the injection location, volume, components, and may be highly variable, even among a consensus of experts. 28 This guideline provides a framework for the use of suboccipital injections in the transitional prophylactic treatment of cluster headache. The two Class I studies 16,29 employed suboccipital injections (in the area of the greater occipital nerve) ipsilateral to the side of the CH pain and used corticosteroids (either cortivazol or betamethasone). They differed with diluent (saline 16 vs local anesthetics 29 ) and injection, as one study utilized a single injection, 29 whereas the other employed an injection series, 16 which was but may be cumbersome to incorporate into clinical practice. Further research is needed to elucidate the optimal corticosteroid, diluent and injection to be used to treat CH. Invasive and noninvasive neurostimulation have emerged as a therapeutic class to treat a variety of headache disorders. 30 We reviewed a class I study of sphenopalatine ganglion stimulation that demonstrated efficacy in the acute treatment of CH with a high order of magnitude over sham stimulation, though this therapy is not routinely available in the U.S. outside of a clinical trial. 11 We also reviewed a class I study of hypothalamic deep brain stimulation which did not demonstrate efficacy over sham stimulation. 17 Occipital nerve stimulation was not included in our systematic review because of the lack of RCTs, but over 100 patients have been reported to be treated with this modality, and a prospective RCT is underway. 39 As invasive neurostimulation carries inherent risks of serious adverse events, these therapies should largely be reserved for further clinical trials or where patients with have truly failed a multitude of other prophylactic therapies. 40 However, the threshold to utilize noninvasive neurostimulation would presumably be much lower, though evidence at this time is limited and comparative studies to medication therapies are not yet available. A study examining the use of a novel noninvasive vagus nerve stimulator was recently published. 41 It compared adjunctive stimulation as a prophylactic treatment with medical standard of care vs standard of care alone in a sample of patients with. The study was not blinded and, therefore, was excluded from our systematic review. However, during the randomized phase of the study there was a significant reduction of weekly attack in those treated with vagus nerve stimulation, with no serious adverse events attributed to the device. Future studies that are blinded with a sham control are warranted to elucidate the efficacy and safety of noninvasive vagus nerve stimulation for treatment of CH. Another potential advantage of neurostimulation devices in the treatment of CH is that a single therapy might have both acute and prophylactic treatment effects, 11,41 which has also been suggested in the treatment of migraine, 42 and represents a versatility that has failed many medical therapies for CH This therapeutic flexibility does not apply to all neurostimulation devices, as noninvasive vagus nerve stimulation does not appear to be in the acute treatment of CH, at least in a population. 41 Other therapies with reports of efficacy in the treatment of CH that did not satisfy the inclusion criteria for our systematic review have been either recommended by other guidelines, discussed as therapeutic options in expert opinions, or reported in case series or open label studies ,46-52 Such treatments include methysergide, methylergonovine, flunarizine, topiramate, gabapentin, baclofen, clonidine, pizotifen, histamine sulfate, kudzu, 2-bromo lysergic

12 1104 July/August 2016 acid diethylamide, hormonal therapies such as testosterone and clomiphene, as well as interventional therapies such as botulinum toxin injections, sphenopalatine ganglion blockade, and radio ablation. RCTs carefully studying the safety and efficacy of these therapies are required before any specific recommendations can be made, Other emerging therapies for the treatment of CH that are under active investigation include monoclonal antibodies to calcitonin gene-related peptide and other forms of neurostimulation with more refined methodology aimed at reducing procedurerelated adverse events ( results?term 5 cluster 1 headache&search 5 Search). In the meantime, this guideline, based upon the level of evidence that treatments are, can be considered along with other variables when making treatment recommendations for patients with episodic and. Acknowledgments: We thank Lisa Marks, MLS, AHIP, for her assistance with developing and conducting the literature search. STATEMENT OF AUTHORSHIP All authors provided substantial contributions to the conception and design of this review and the interpretation of data. Matthew Robbins and Todd Schwedt were involved in drafting the article and all authors were involved in revising it critically for intellectual content. All authors provided final approval of the version to be published. SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article at the publisher s website. REFERENCES 1. Headache Classification Committee of the International Headache Society. The international classification of headache disorders, 3rd edition (beta version). Cephalalgia. 2013;33: Fischera M, Marziniak M, Gralow I, Evers S. The incidence and prevalence of cluster headache: A meta-analysis of population-based studies. Cephalalgia. 2008;28: van Vliet JA, Eekers PJ, Haan J, Ferrari MD. Features involved in the diagnostic delay of cluster headache. J Neurol Neurosurg Psychiatry. 2003;74: Voiticovschi-Iosob C, Allena M, De Cillis I, Nappi G, Sjaastad O, Antonaci F. Diagnostic and therapeutic errors in cluster headache: A hospital-based study. J Headache Pain. 2014;15: Lademann V, Jansen JP, Evers S, Frese A. Evaluation of guideline-adherent treatment in cluster headache. Cephalalgia. 2015; Oct 19. pii: [Epub ahead of print]. 6. Francis GJ, Becker WJ, Pringsheim TM. Acute and preventive pharmacologic treatment of cluster headache. Neurology. 2010;75: Shekelle PG, Ortiz E, Rhodes S, et al. Validity of the agency for healthcare research and quality clinical practice guidelines: How quickly do guidelines become outdated? JAMA. 2001;286: Alderson LJ, Alderson P, Tan T. Median life span of a cohort of National Institute for Health and Care Excellence clinical guidelines was about 60 months. J Clin Epidemiol. 2014;67: Martinez Garcia L, Sanabria AJ, Garcia Alvarez E, et al. The validity of recommendations from clinical guidelines: A survival analysis. CMAJ. 2014;186: Gronseth GS, Woodroffe LM, Getchius TS. Clinical practice guideline process manual. In: American Academy of Neurology; Available at: tools.aan.com/globals/axon/assets/9023.pdf. (accessed November 30, 2015). 11. Schoenen J, Jensen RH, Lanteri-Minet M, et al. Stimulation of the sphenopalatine ganglion (SPG) for cluster headache treatment. Pathway CH-1: A sham-controlled study. Cephalalgia. 2013;33: Hakim SM. Warfarin for refractory chronic cluster headache: A randomized pilot study. Headache. 2011;51: Pageler L, Katsarava Z, Lampl C, et al. Frovatriptan for prophylactic treatment of cluster headache: Lessons for future trial design. Headache. 2011;51: Tronvik E, Wienecke T, Monstad I, et al. Randomised trial on episodic cluster headache with an

13 Headache 1105 angiotensin II receptor blocker. Cephalalgia. 2013; 33: Anthony M, Lord GD, Lance JW. Controlled trials of cimetidine in migraine and cluster headache. Headache. 1978;18: Leroux E, Valade D, Taifas I, et al. Suboccipital steroid injections for transitional treatment of patients with more than two cluster headache attacks per day: A randomised, double-blind, placebo-controlled trial. Lancet Neurol. 2011;10: Fontaine D, Lazorthes Y, Mertens P, et al. Safety and efficacy of deep brain stimulation in refractory cluster headache: A randomized placebocontrolled double-blind trial followed by a 1-year open extension. J Headache Pain. 2010;11: Peterlin BL, Purdy RA, Rapoport AM, et al. Acute and preventive pharmacologic treatment of cluster headache. Neurology. 2011;77: ; author reply Sarchielli P, Granella F, Prudenzano MP, et al. Italian guidelines for primary headaches: 2012 revised version. J Headache Pain. 2012;13:S31-S Treatment Guideline Subcommittee of the Taiwan Headache Society, Chen PK, Chen HM, et al. [Treatment guidelines for acute and preventive treatment of cluster headache]. Acta Neurol Taiwan. 2011;20: May A, Leone M, Afra J, et al. EFNS guidelines on the treatment of cluster headache and other trigeminal-autonomic cephalalgias. Eur J Neurol. 2006;13: Bussone G, Leone M, Peccarisi C, et al. Double blind comparison of lithium and verapamil in cluster headache prophylaxis. Headache. 1990;30: Holle D, Burmeister J, Scherag A, Ose C, Diener HC, Obermann M. Study protocol of Prednisone in episodic Cluster Headache (PredCH): A double-blind, placebo-controlled parallel group trial to evaluate the efficacy and safety of oral prednisone as an add-on therapy in the prophylactic treatment of episodic cluster headache with verapamil. BMC Neurol. 2013;13: Kowacs PA, Piovesan EJ, de Campos RW, Lange MC, Zetola VF, Werneck LC. Warfarin as a therapeutic option in the control of chronic cluster headache: A report of three cases. J Headache Pain. 2005;6: Souza JA, Moreira Filho PF, Jevoux Cda C, Martins GF, Pitombo AB. Remission of refractory chronic cluster headache after warfarin administrations: Case report. Arq Neuropsiquiatr. 2004;62: Ashkenazi A, Blumenfeld A, Napchan U, et al. Peripheral nerve blocks and trigger point injections in headache management A systematic review and suggestions for future research. Headache. 2010;50: Blumenfeld A, Ashkenazi A, Grosberg B, et al. Patterns of use of peripheral nerve blocks and trigger point injections among headache practitioners in the USA: Results of the American Headache Society Interventional Procedure Survey (AHS- IPS). Headache. 2010;50: Blumenfeld A, Ashkenazi A, Napchan U, et al. Expert consensus recommendations for the performance of peripheral nerve blocks for headaches A narrative review. Headache. 2013;53: Ambrosini A, Vandenheede M, Rossi P, et al. Suboccipital injection with a mixture of rapid- and longacting steroids in cluster headache: A double-blind placebo-controlled study. Pain. 2005;118: Schwedt TJ, Vargas B. Neurostimulation for treatment of migraine and cluster headache. Pain Med. 2015;16: Mueller O, Diener HC, Dammann P, et al. Occipital nerve stimulation for intractable chronic cluster headache or migraine: A critical analysis of direct treatment costs and complications. Cephalalgia. 2013;33: Strand NH, Trentman TL, Vargas BB, Dodick DW. Occipital nerve stimulation with the Bion(R) microstimulator for the treatment of medically refractory chronic cluster headache. Pain Physician. 2011;14: Magis D, Gerardy PY, Remacle JM, Schoenen J. Sustained ness of occipital nerve stimulation in drug-resistant chronic cluster headache. Headache. 2011;51: Fontaine D, Christophe Sol J, Raoul S, et al. Treatment of refractory chronic cluster headache by chronic occipital nerve stimulation. Cephalalgia. 2011;31: Mueller OM, Gaul C, Katsarava Z, Diener HC, Sure U, Gasser T. Occipital nerve stimulation for the treatment of chronic cluster headache Lessons learned from 18 months experience. Cent Eur Neurosurg. 2011;72:84-89.

14 1106 July/August Burns B, Watkins L, Goadsby PJ. Treatment of intractable chronic cluster headachebyoccipitalnervestimulationin14patients.neurology. 2009;72: Burns B, Watkins L, Goadsby PJ. Treatment of medically intractable cluster headache by occipital nerve stimulation: Long-term follow-up of eight patients. Lancet. 2007;369: Magis D, Allena M, Bolla M, De Pasqua V, Remacle JM, Schoenen J. Occipital nerve stimulation for drugresistant chronic cluster headache: A prospective pilot study. Lancet Neurol. 2007;6: Wilbrink LA, Teernstra OP, Haan J, et al. Occipital nerve stimulation in medically intractable, chronic cluster headache. The ICON study: Rationale and protocol of a randomised trial. Cephalalgia. 2013;33: Martelletti P, Jensen RH, Antal A, et al. Neuromodulation of chronic headaches: Position statement from the European Headache Federation. J Headache Pain. 2013;14: Gaul C, Diener HC, Silver N, et al. Non-invasive vagus nerve stimulation for PREVention and Acute treatment of chronic cluster headache (PREVA): A randomised controlled study. Cephalalgia. 2016;36: Bhola R, Kinsella E, Giffin N, et al. Single-pulse transcranial magnetic stimulation (stms) for the acute treatment of migraine: Evaluation of outcome data for the UK post market pilot program. J Headache Pain. 2015;16: Jammes JL. The treatment of cluster headaches with prednisone. Dis Nerv Syst. 1975;36: Monstad I, Krabbe A, Micieli G, et al. Preemptive oral treatment with sumatriptan during a cluster period. Headache. 1995;35: Nilsson Remahl AI, Ansjon R, Lind F, Waldenlind E. Hyperbaric oxygen treatment of active cluster headache: A double-blind placebocontrolled cross-over study. Cephalalgia. 2002;22: Donnet A, Demarquay G, Ducros A, et al. French guidelines for diagnosis and treatment of cluster headache (French Headache Society). Rev Neurol (Paris). 2014;170: Saper JR, Evans RW. Oral methylergonovine maleate for refractory migraine and cluster headache prevention. Headache. 2013;53: Ashkenazi A, Schwedt T. Cluster headache Acute and prophylactic therapy. Headache. 2011; 51: Tepper SJ, Stillman MJ. Cluster headache: Potential options for medically refractory patients (when all else fails). Headache. 2013;53: Stillman MJ. Testosterone replacement therapy for treatment refractory cluster headache. Headache. 2006;46: Rozen TD. Clomiphene citrate as a preventive treatment for intractable chronic cluster headache: A second reported case with long-term follow-up. Headache. 2015;55: Bratbak DF, Nordgard S, Stovner LJ, et al. Pilot study of sphenopalatine injection of onabotulinumtoxina for the treatment of intractable chronic cluster headache. Cephalalgia. 2016;36: