Headache Disorders
Read Time: 13 mins

Cluster Headache – Diagnosis and Treatment

Copy Link
Published Online: Jun 4th 2011
Authors: Todd D Rozen
Quick Links:
Article Information

Cluster headache is a primary headache syndrome that is under-diagnosed and in many instances undertreated. The pain produced during a cluster headache is more severe than that generated by any other primary headache. Cluster headache is very stereotyped in its presentation and is fairly easy to diagnose with an in-depth headache history. Cluster headache is easy to treat in most individuals if the correct medications are used and the correct dosages are prescribed. This article presents information on the clinical presentation of cluster headache and both medicinal and surgical interventions.


There is no more severe pain than that sustained by a cluster headache sufferer and if not for the rather short duration of attacks most cluster sufferers would choose death rather than continue suffering. Cluster has been nicknamed the ‘suicide headache’ because cluster sufferers typically have thought about taking or have taken their lives during a cluster headache.

Diagnosis of Cluster Headache
Recently, Klapper et al.1 determined that the average time it takes for a cluster sufferer to be diagnosed correctly by the medical profession is 6.6 years. The average number of physicians seen prior to a correct diagnosis is four and the average number of incorrect diagnoses before a correct diagnosis of cluster is four. This statistic is unacceptable based on the pain and suffering cluster patients must endure when they are not treated correctly or when not being treated at all.

Cluster is a stereotypic episodic headache disorder marked by frequent attacks of short-lasting, severe, unilateral head pain with associated autonomic symptoms. A cluster headache is defined as an individual attack of head pain, while a cluster period or cycle is the time that a patient is having daily cluster headaches. Episodic cluster headache (the most common form) is defined by a cluster period lasting seven days to one year separated by a pain-free period lasting one month or longer. Chronic cluster headache is defined by attacks that occur for greater than one year without remission or with remissions lasting less than one month.

Typical cluster headache location is retro-orbital, periorbital and occipitonuchal. Maximum pain is normally retro-orbital in greater than 70% of patients. Pain quality is described as boring, stabbing, burning, or squeezing. Cluster headache intensity is always severe, never mild, although headache pain intensity may be less at the beginning and end of cluster periods. Cluster headaches that awaken a patient from sleep will be more severe than those occurring during the day.

The one-sided nature of cluster headaches is a trademark. Cluster sufferers will normally experience cluster headaches on the same side of the head their entire lives. The headaches will only shift to the other side of the headache in 15% at the next cluster period and sideshifting during the same cluster cycle will only occur in 5% of patients. The duration of individual cluster headaches is between 15 minutes and 180 minutes with greater than 75% attacks being less than 60 minutes. Attack frequency is between one to three attacks per day with most patients experiencing two or less headaches in a day. Peak time periods for daily cluster headache onset is 1am to 2am, 1pm to 3pm, and after 9pm so that most cluster patients can complete their occupation requirements without experiencing headaches during the working day. The headaches have a predilection for the first rapid eye movement (REM) sleep phase so the cluster patient will awaken with a severe headache 60 to 90 minutes after falling asleep. Cluster period duration normally lasts between two to 12 weeks and patients generally experience one or two cluster periods per year. Remission periods (headache-free time in-between cluster cycles) average six months to two years. Cluster headache is marked by its associated autonomic symptoms that typically occur on the same side as the head pain but can be bilateral. Lacrimation is the most commonly associated symptom occurring in 73% of patients followed by conjunctival injection in 60%, nasal congestion 42%, nasal rhinorrhea 22% and a partial Horner’s syndrome in 16% to 84%. Symptoms generally attributed to migraine can also occur during a cluster headache including nausea, vomiting, photophobia, and phonophobia. Photophobia and phonophobia probably occur as frequently in cluster as in migraine.Vingen et al.2 found a self-reported frequency of photophobia in 91% and phonophobia in 89% of 50 cluster patients. These symptoms may not be syndrome-specific but may just be markers of trigeminal-autonomic pathway activation.The occurrence of so called ‘migrainous symptoms’ in cluster has probably led to the high rate of misdiagnosis of cluster patients. Cluster headache is really a state of agitation as remaining motionless appears to make the pain worse. Some cluster patients state that they will lie down with a cluster headache but when questioned it has been discovered they do not lie still but roll around on the bed in agony. Many patients will develop their own routine during a cluster attack including banging their heads against a wall, crawling on the floor, taking hot showers or just screaming out in pain. Only approximately 3% can lie still during an attack.3The face of cluster patients has been described as having a ‘leonine appearance’ with thick, coarse facial skin, peau d’orange appearance, marked wrinkling of the forehead and face with deep furrowed brows. In addition, Kudrow 4 reported that two-thirds of the patients in his large series had hazel colored eyes.These features may actually reflect a history of smoking and alcohol overuse, which is common in cluster sufferers.

Table 1: Abortive Treatment Options

All cluster headache patients require treatment. Other primary headache syndromes can sometimes be managed non-medicinally but in regard to cluster headache medication, sometimes even polypharmacy is indicated. Cluster headache treatment can be divided into three classes. Abortive therapy is a treatment given at the time of an attack for that individual attack alone. Transitional therapy can be considered an intermittent or short-term preventive treatment. An agent is started at the same time as the patient’s true maintenance preventive. The transitional therapy will provide the cluster patient attack relief while the maintenance preventive is being built up to a therapeutic dosage. Preventive therapy consists of daily medication that is supposed to reduce the frequency of headache attacks, lower attack intensity, and lessen attack duration. The main goal of cluster headache preventive therapy should be to make a patient cluster-free on preventives even though they are still in a cluster cycle. As most cluster headache patients have episodic cluster headaches, medications are only utilized while a patient is in cycle and is stopped during remission periods.

Abortive Therapy
The goal of abortive therapy for cluster headache is fast, effective, and consistent relief. A sumatriptan injectable can normally alleviate a cluster headache attack within 15 minutes.There is no role for over-the-counter (OTC) agents or butalbital-containing compounds in cluster headache and little if any need for opiates (see Table 1).

Table 2:Transitional Treatment Options
DHE = dihydroergotamine; IM = intramuscular.

Subcutaneous sumatriptan is the most effective medication for the symptomatic relief of cluster headache. In a placebo-controlled study, 6mg of injectable sumatriptan was significantly more effective than placebo, with 74% of patients having complete relief by 15 minutes compared with 26% of placebotreated patients.5 In long-term, open-label studies, sumatriptan is effective in 76% to 100% of all attacks within 15 minutes even after repetitive daily use for several months.6 Interestingly, sumatriptan appears to be 8% less effective in chronic cluster headache than episodic cluster headache. Sumatriptan is contraindicated in patients with uncontrolled hypertension, past history of myocardial infarction or stroke.As almost all cluster patients have a strong history of cigarette smoking, the physician must closely monitor cardiovascular (CV) risk factors in these patients.

Sumatriptan nasal spray (20mg) has been shown to be more effective than placebo in the acute treatment of cluster attacks. In over 80 patients tested, intranasal sumatriptan reduced cluster headache pain from very severe, severe, or moderate to mild or no pain at 30 minutes in 58% of sumatriptan users, compared with 30% of patients given placebo on the first attack treated, while the rates were 50% (sumatriptan) compared with 33% (placebo) after the second treated attack. 7 Sumatriptan nasal spray appears to be efficacious for cluster headache but less effective than subcutaneous injection. Sumatriptan nasal spray should be considered as a cluster headache abortive in patients who cannot tolerate injections or when, situationally (e.g. an office setting), injections would be considered socially unacceptable.

In many instances cluster headache patients may need to use sumatriptan more than once a day for days to weeks at a time. Hering 8  noted that the use of daily injectable sumatriptan in four cluster patients led to a marked increase in the frequency of cluster attacks three to four weeks after initiating treatment. In three patients the character of the cluster headache changed while two patients experienced prolongation of their cluster headache period. Withdrawal of sumatriptan reduced the frequency of headaches. Even though daily sumatriptan may be benefiting a cluster headache patient the goal should be to have them cluster free on preventive medication not using abortives to achieve cluster-free status.

Oxygen inhalation is an excellent abortive therapy for cluster headache.Typical dosing is 100% oxygen given via a non-rebreather face mask at seven liters to 10 liters per minute for 20 minutes. Past studies indicate that about 70% of cluster patients respond to oxygen therapy.9 In some patients oxygen is completely effective at aborting an attack if taken when the pain is at maximal intensity, while in others the attack is only delayed for minutes to hours rather than completely alleviated. It is not uncommon for a cluster patient to be headache-free while on oxygen but immediately redevelop pain when the oxygen is removed. Oxygen is overall a very attractive therapy as it is completely safe and can be used multiple times during the day, unlike sumatriptan or ergots, for example, which if used too frequently could cause cardiac ischemia. Large oxygen tanks are prescribed for cluster patients’ homes while portable tanks can be taken to the workplace. There may be a gender discrepancy in response to oxygen. Rozen et al.10 reported that only 59% of female cluster patients at their academic center responded to oxygen while 87% of men responded to oxygen. A recent study showed that individuals who do not respond to typical oxygen dosing may respond at higher flow rates up to 15 liters per minute.11 A small, open-label study of hyperbaric oxygen (2atm) delivered over 30 minutes demonstrated efficacy in six of seven cluster patients within five to 13 minutes, with these patients reporting complete or partial interruption of the cluster period.12

Transitional Therapy
Transitional cluster therapy is a short-term preventive treatment that bridges the time between cluster diagnosis and the time when the true traditional maintenance preventive agent becomes efficacious. Transitional preventives are started at the same time the traditional preventive is begun. The transitional preventive should provide the cluster patient with almost immediate pain relief and allow the patient to be headache-free or near headache-free while the traditional preventive medication dose is being tapered up to an effective level.When the transitional agent is tapered off the maintenance preventive will have kicked in, thus the patient will have no gap in headache preventive coverage (see Table 2).

A short course of corticosteroids is the best known transitional therapy for cluster headache. Typically, within 24 to 48 hours of administration, patients become cluster-free and by the time the steroid taper has ended the patients’ main preventive agent has started to become effective. Prednisone or dexamethasone are the most typically used corticosteroids in cluster. A typical taper would be 80mg of prednisone for the first two days followed by 60mg for two days, 40mg for two days, 20mg for two days, 10mg for two days then ceasing to use the agent. There is no set manner in which to dose corticosteroids in cluster headache. Preventive Therapy
Preventive agents are absolutely necessary in cluster headaches unless the cluster periods last less than two weeks. Preventive medications are only used while the patient is in cycle and they are tapered off once a cluster period has ended. If a patient decides to remain on a preventive agent even after they have gone out of cycle this does not appear to prevent a subsequent cluster period from starting.The maintenance preventive should be started at the time a transitional agent is given. Most physicians treating cluster headache will increase the dosages of the preventive agents very quickly to obtain a desired response.Very large dosages, much higher than that suggested in the Physician’s Desk Reference (PDR), are sometimes necessary when treating cluster headache. A well-recognized trait of cluster patients is that they can tolerate medications much better than non-cluster patients. Most of the recognized cluster preventives can be used in both episodic and chronic cluster headache. Polypharmacy is not discouraged in cluster headache prevention. Not unlike the multiple preventive regime utilized in trigeminal neuralgia, cluster attacks are so extreme that severe add-on therapy is encouraged rather than ceasing treatment with one agent having the attacks worsen again and trying another single agent (see Table 3).

Verapamil appears to be the best first-line therapy for both episodic and chronic cluster headache.3 13 It can be used safely in conjunction with sumatriptan, ergotamine, and corticosteroids, as well as other preventive agents. Leone et al.14 compared the efficacy of verapamil with placebo in the prophylaxis of episodic cluster headache. After five days of run-in, 15 patients received verapamil (120mg tid) and 15 received placebo (tid) for 14 days. The authors found a significant reduction in attack frequency and abortive agent consumption in the verapamil group. The initial starting daily dosage of verapamil is 80mg three times a day or building up to this dosage within three to five days. The non-sustained release formulation appears to function better than the sustained release preparation but there is no literature proving this. Dosages are typically increased by 80mg every three to seven days. If a patient needs greater than 480mg per day then an electrocardiogram (ECG) is necessary before each dose change thereafter to guard against heart block. It is not uncommon for cluster patients to need dosages as high as 800mg to gain cluster remission. Most headache specialists will push the dose as high as 1g if tolerated. Constipation is the most common side effect, but dizziness, edema, nausea, fatigue, hypotension, and bradycardia may also occur.

Lithium Carbonate
Lithium carbonate therapy is still considered a mainstay of cluster prevention but its narrow therapeutic window and high side effect profile makes it less desirable than other, newer, preventives. Since 2001, there have been 28 clinical trials looking at the efficacy of lithium in cluster therapy. For chronic cluster 78% of patients treated (in 25 trials) have improved on lithium while 63% of episodic patients have gained cluster remission on lithium.When lithium was compared with verapamil in a single trial, both agents were found to be effective but verapamil caused fewer side effects and had a more rapid onset of action.15 A single double-blind, placebo-controlled trial failed to show the superiority of lithium (800mg sustained release) over placebo. However, this study was halted one week after treatment began, and there was an unexpectedly high placebo response rate of 31%.16 The treatment period was therefore too short to be conclusive.

The initial starting dosage of lithium is 300mg at bedtime with dose adjustments usually no higher than 900mg per day. Lithium is often effective at serum concentrations (0.3–0.8mM) lower than those usually required for the treatment of bipolar disorder. Most cluster patients benefit from dosages between 600mg and 900mg a day. During the initial treatment stages, lithium serum concentrations should be checked repeatedly to guard against toxicity. Serum lithium concentrations should be measured in the morning 12 hours after the last dose. In addition, prior to starting lithium, renal and thyroid functions need to be checked. Adverse events related to lithium include tremor, diarrhea, and polyuria. Valproic Acid
In a open label investigation 26 patients (21 chronic cluster, five episodic cluster) were treated with divalproex sodium.17 The mean decrease in headache frequency was 53.9% for the chronic cluster patients and 58.6% for the episodic cluster patients. The mean dose of divalproex sodium used was 838mg, which could be considered a low dose by cluster standards. Recently, a double-blind placebo controlled study of sodium valproate (1,000–2,000mg/day) in cluster was completed. Ninety-six patients were included, 50 in the sodium valproate group and 46 in the placebo group. After a seven-day run-in period, patients were treated for two weeks. Primary efficacy was the percentage of patients having an at least 50% reduction in the average number of attacks per week between the run-in period and the last week of treatment. Fifty per cent of subjects in the sodium valproate group and 62% in the placebo group had significant improvement (P=0.23). Due to the high success rate seen with the placebo, the authors felt they could make no conclusion about the efficacy of sodium valproate in cluster.18 The extended release preparation of valproic acid appears to work well and dosing up to 3,000mg qhs can be effective.

Topiramate is a more recent antiepileptic that may be efficacious in both migraine and cluster headache prevention. Lainez et al.19 treated 26 patients (12 episodic, 14 chronic) with topiramate to a maximum dose of 200mg. Topiramate rapidly induced cluster remission in 15 patients, reduced the number of attacks by more than 50% in six patients, and reduced the cluster period duration in 12. The mean time to remission was 14 days, but in seven patients remission was obtained within the first days of treatment with very low dosages (25–75mg a day). Six patients discontinued treatment due to side effects (all with daily dosages over 100mg) or lack of efficacy.

Topiramate should be initiated at a dose of 25mg per day and increased in 25mg increments every five days up to 75mg.The patient should be monitored at this dose for several weeks before deciding if the dose needs to be increased. Dosages up to 400mg have been needed in some cluster patients. Anecdotally, there appears to be a therapeutic window for topiramate in cluster. Some patients have experienced worsening of attacks when the dose is raised above a certain limit and improvement again when the dose is lowered back down.

Serum melatonin levels are reduced in patients with cluster headache, particularly during a cluster period. This loss of melatonin may be the inciting event necessary to at least produce nocturnal cluster attacks. Providing back melatonin via an oral supplementation route theoretically could act as a cluster preventive. The efficacy of 10mg of oral melatonin was evaluated in a double-blind, placebocontrolled trial.20 Cluster headache remission within three to five days occurred in five of 10 patients who received melatonin compared with zero of 10 patients who received placebo. Melatonin only appeared to work in episodic cluster patients. Recently,melatonin has also been shown to be an effective preventive in chronic cluster headache.21 A negative study was published utilizing melatonin for cluster prevention but the dosing was lower than the other studies and a sustained preparation was given.22 The author believes that melatonin should be initiated in all cluster patients as a first-line preventive sometimes even before verapamil. It has minimal side effects and in a number of patients it can turn off nocturnal clusters within 24 hours. Melatonin also appears to prevent daytime attacks. In addition, even when melatonin does not completely resolve all of the attacks it appears to lower the dose necessary of the other addon preventives. The typical dose of melatonin used is 9mg at bedtime (three 3mg tablets) but higher dosages may be necessary. If one brand of commercial melatonin does not work another should be tried because the true amount of melatonin in various OTC brands varies widely.

Surgical Treatment of Cluster Headache
The surgical treatment of cluster headache should only be considered after a patient has exhausted all medicinal options or when a patient’s medical history precludes the use of typical cluster abortive and preventive medications. Episodic cluster patients should rarely be referred for surgery because of the presence of remission periods. Once a cluster patient is deemed a medical failure only those who have strictly side-fixed headaches should be considered for surgery. Other criteria for cluster surgery include pain mainly localizing to the ophthalmic division of the trigeminal nerve, a psychologically stable individual and one without an addicting personality. Cluster patients must understand that, in most instances, to alleviate their cluster pain, the trigeminal nerve will have to be injured, leaving them not only with facial analgesia but a risk of developing severe adverse events including corneal anesthesia and anesthesia dolorosa. Surgical Techniques for Cluster Headache

Surgeryon the Cranial Parasympathetic System
The parasympathetic autonomic pathway can be interrupted by sectioning the greater superficial petrosal nerve, the nervus intermedius, or the sphenopalatine ganglion. Based on the trigeminal autonomic (TAC) reflex pathway hypothesis for cluster pathogenesis, this technique should obliterate the autonomic symptoms associated with a cluster headache but would not appear likely to affect the cluster associated pain because this is a trigeminal nerve-driven response, although the nervus intermedius may have nociceptive fibers. From reports in literature, techniques targeting the autonomic system in cluster have provided very inconsistent pain relief in patients and when deemed initially effective have had high recurrence rates.

Surgery on the Sensory Trigeminal Nerve

Procedures directed toward the sensory trigeminal nerve include:

  • alcohol injection into supra-orbital and infraorbital nerves;
  • alcohol injection into the Gasserian (trigeminal) ganglion;
  • avulsion of infraorbital/supraorbital/supratrochlear nerves;
  • retrogasserian glycerol injection;
  • radiofrequency trigeminal gangliorhyzolysis; and
  • trigeminal root section.

Based on the TAC reflex hypothesis this would mechanistically make the most sense for aborting both the pain and possibly the autonomic symptoms related to the cluster attack.Overall, these techniques have been the most successful at alleviating cluster pain, especially radiofrequency trigeminal gangliorhyzolysis.5 With some of the procedures there is the possibility of very severe adverse events including anesthesia dolorosa.

A New Direction

Hypothalamic Stimulation
A recent series of patients reported by Leone et al.23 may completely change the way that chronic intractable cluster headache is treated. Based upon the positron emission tomography (PET) studies by May et al.24 suggesting a hypothalamic generator for cluster, Leone et al. have treated several chronic cluster patients by electrode implantation into the posterior inferior hypothalamus.When the stimulator is activated in these patients the cluster pain vanishes. When the stimulator is turned off the headaches reappear. This technique is novel and more investigation is necessary before it can be considered a rational treatment of cluster. What is exciting about this is that knowledge of pathogenesis will help to discover new and better therapies for cluster headache.

Cluster headache is a primary headache syndrome that is under-diagnosed and in many instances undertreated. Cluster headache is very stereotyped in its presentation and fairly easy to diagnose with an in depth headache history. Treatment of cluster headache can be very successful if the correct medications are used and the correct dosages are prescribed. New understanding of cluster pathogenesis has led to better medicinal and surgical treatment strategies.

Further Resources

Share this Article
Related Content In Headache Disorders
  • Copied to clipboard!
    accredited arrow-down-editablearrow-downarrow_leftarrow-right-bluearrow-right-dark-bluearrow-right-greenarrow-right-greyarrow-right-orangearrow-right-whitearrow-right-bluearrow-up-orangeavatarcalendarchevron-down consultant-pathologist-nurseconsultant-pathologistcrosscrossdownloademailexclaimationfeedbackfiltergraph-arrowinterviewslinkmdt_iconmenumore_dots nurse-consultantpadlock patient-advocate-pathologistpatient-consultantpatientperson pharmacist-nurseplay_buttonplay-colour-tmcplay-colourAsset 1podcastprinter scenerysearch share single-doctor social_facebooksocial_googleplussocial_instagramsocial_linkedin_altsocial_linkedin_altsocial_pinterestlogo-twitter-glyph-32social_youtubeshape-star (1)tick-bluetick-orangetick-red tick-whiteticktimetranscriptup-arrowwebinar Sponsored Department Location NEW TMM Corporate Services Icons-07NEW TMM Corporate Services Icons-08NEW TMM Corporate Services Icons-09NEW TMM Corporate Services Icons-10NEW TMM Corporate Services Icons-11NEW TMM Corporate Services Icons-12Salary £ TMM-Corp-Site-Icons-01TMM-Corp-Site-Icons-02TMM-Corp-Site-Icons-03TMM-Corp-Site-Icons-04TMM-Corp-Site-Icons-05TMM-Corp-Site-Icons-06TMM-Corp-Site-Icons-07TMM-Corp-Site-Icons-08TMM-Corp-Site-Icons-09TMM-Corp-Site-Icons-10TMM-Corp-Site-Icons-11TMM-Corp-Site-Icons-12TMM-Corp-Site-Icons-13TMM-Corp-Site-Icons-14TMM-Corp-Site-Icons-15TMM-Corp-Site-Icons-16TMM-Corp-Site-Icons-17TMM-Corp-Site-Icons-18TMM-Corp-Site-Icons-19TMM-Corp-Site-Icons-20TMM-Corp-Site-Icons-21TMM-Corp-Site-Icons-22TMM-Corp-Site-Icons-23TMM-Corp-Site-Icons-24TMM-Corp-Site-Icons-25TMM-Corp-Site-Icons-26TMM-Corp-Site-Icons-27TMM-Corp-Site-Icons-28TMM-Corp-Site-Icons-29TMM-Corp-Site-Icons-30TMM-Corp-Site-Icons-31TMM-Corp-Site-Icons-32TMM-Corp-Site-Icons-33TMM-Corp-Site-Icons-34TMM-Corp-Site-Icons-35TMM-Corp-Site-Icons-36TMM-Corp-Site-Icons-37TMM-Corp-Site-Icons-38TMM-Corp-Site-Icons-39TMM-Corp-Site-Icons-40TMM-Corp-Site-Icons-41TMM-Corp-Site-Icons-42TMM-Corp-Site-Icons-43TMM-Corp-Site-Icons-44TMM-Corp-Site-Icons-45TMM-Corp-Site-Icons-46TMM-Corp-Site-Icons-47TMM-Corp-Site-Icons-48TMM-Corp-Site-Icons-49TMM-Corp-Site-Icons-50TMM-Corp-Site-Icons-51TMM-Corp-Site-Icons-52TMM-Corp-Site-Icons-53TMM-Corp-Site-Icons-54TMM-Corp-Site-Icons-55TMM-Corp-Site-Icons-56TMM-Corp-Site-Icons-57TMM-Corp-Site-Icons-58TMM-Corp-Site-Icons-59TMM-Corp-Site-Icons-60TMM-Corp-Site-Icons-61TMM-Corp-Site-Icons-62TMM-Corp-Site-Icons-63TMM-Corp-Site-Icons-64TMM-Corp-Site-Icons-65TMM-Corp-Site-Icons-66TMM-Corp-Site-Icons-67TMM-Corp-Site-Icons-68TMM-Corp-Site-Icons-69TMM-Corp-Site-Icons-70TMM-Corp-Site-Icons-71TMM-Corp-Site-Icons-72