Cluster Headache: When to See a Doctor

At a glance
- Prevalence / approximately 1 in 1,000 people, with a 3:1 male-to-female ratio
- Attack duration / 15 to 180 minutes of strictly unilateral pain
- Attack frequency / one to eight attacks per day during a cluster bout
- Key autonomic signs / ipsilateral lacrimation, conjunctival injection, ptosis, rhinorrhea, or miosis
- First-line acute treatment / high-flow oxygen at 12 L/min via non-rebreather mask
- First-line preventive / verapamil 240 to 960 mg daily during cluster periods
- Episodic vs. chronic / episodic form accounts for roughly 80 to 85% of cases
- Diagnostic standard / ICHD-3 criteria require at least five attacks meeting defined features
- Average delay to diagnosis / 5.3 years from symptom onset in European cohort studies
- Emergency red flag / a first attack mimics subarachnoid hemorrhage and requires urgent imaging
What Cluster Headache Actually Feels Like
Cluster headache produces what the International Headache Society classifies as the most painful primary headache disorder [1]. The pain is strictly one-sided, centered behind or around the eye, and peaks within 5 to 10 minutes. Attacks last between 15 and 180 minutes, then resolve completely until the next episode.
Unlike migraine, patients with cluster headache rarely lie still. Restlessness and agitation during an attack are so characteristic that the ICHD-3 diagnostic criteria list "a sense of restlessness or agitation" as a supportive feature [1]. Pacing, rocking, and pressing on the affected orbit are common behaviors. Patients frequently describe the pain as a hot poker driven through the eye socket. The term "suicide headache," though informal, reflects the intensity that patients endure.
Autonomic symptoms accompany the pain on the same side. These include tearing, redness of the eye, a drooping or swollen eyelid, nasal congestion or clear runny nose, forehead sweating, and pupil constriction. At least one autonomic sign must be present for an ICHD-3 diagnosis [1]. A 2019 population study published in Cephalalgia found that 97% of cluster headache patients reported at least two ipsilateral autonomic features during attacks [2].
Attacks cluster in bouts. During a cluster period (typically lasting 6 to 12 weeks), patients may experience one to eight attacks per day, often at the same time each night, usually 1 to 2 hours after falling asleep [3]. Between bouts, patients are completely headache-free, sometimes for months or years.
Why Cluster Headache Happens
The exact cause remains incompletely understood, but converging evidence points to the hypothalamus as the central generator. Functional MRI studies have consistently demonstrated posterior hypothalamic activation during cluster attacks [4]. This region governs circadian rhythm, which explains the clocklike regularity of attacks and their seasonal recurrence.
The trigeminovascular system mediates the pain itself. Activation of trigeminal nerve fibers around meningeal blood vessels releases calcitonin gene-related peptide (CGRP) and vasoactive intestinal peptide (VIP), which drive both pain signaling and the parasympathetic autonomic features [5]. Blood levels of CGRP rise during cluster attacks and normalize after successful treatment with sumatriptan [5].
Genetic factors play a role. First-degree relatives of cluster headache patients carry a 14- to 39-fold increased risk compared to the general population [6]. Twin studies support a heritable component, though no single gene is responsible.
Common triggers during a cluster period include alcohol (even small amounts), histamine, nitroglycerin, strong odors, and sudden changes in sleep schedule. Outside of a cluster bout, these same triggers typically do not provoke an attack, reinforcing the concept of a "vulnerable window" governed by hypothalamic cycling [3].
When to Go to the Emergency Department
A first-ever attack of severe unilateral headache with autonomic symptoms requires emergency evaluation. The presentation can mimic subarachnoid hemorrhage, carotid artery dissection, or acute angle-closure glaucoma. A CT head without contrast (followed by CT angiography or lumbar puncture if indicated) is standard practice to exclude secondary causes [7].
Go immediately if you experience any of the following:
- The worst headache of your life with sudden onset ("thunderclap" quality reaching peak intensity within 60 seconds)
- Neurological deficits: limb weakness, speech difficulty, visual field loss, or altered consciousness
- Fever and neck stiffness accompanying the headache
- Headache following head trauma within the past 4 weeks
- New headache in a person over age 50 who has never had similar episodes
The American Headache Society's 2016 position paper on emergency headache evaluation states: "Any first or worst headache warrants neuroimaging to exclude life-threatening secondary etiologies before assigning a primary headache diagnosis" [7].
Dr. Peter Goadsby, a neurologist at King's College London and leading cluster headache researcher, has noted: "Cluster headache is a medical emergency when it first presents, because the differential diagnosis includes conditions that kill within hours" [8]. After secondary causes are excluded, the pattern of recurrent attacks over weeks or months typically confirms the diagnosis.
When to Schedule a Neurology Appointment
Not every cluster headache visit needs the emergency department. Once you have an established diagnosis, see your neurologist promptly if:
- Your cluster bout pattern changes (attacks become more frequent, longer, or switch sides)
- Your current preventive medication stops working
- You develop chronic cluster headache, defined as remission periods lasting fewer than 3 months over 1 year [1]
- You need a transitional bridge treatment such as a short course of oral corticosteroids or a greater occipital nerve block
- You are experiencing more than two attacks per day despite preventive therapy
A referral is overdue if you have been treating recurrent severe one-sided headaches with over-the-counter analgesics for months without a formal diagnosis. The average diagnostic delay for cluster headache in a European multicenter study was 5.3 years, with 77% of patients initially receiving incorrect diagnoses including migraine, sinusitis, or dental pathology [9]. Shorter time-to-diagnosis correlates with better quality of life and earlier access to effective abortive therapy.
How Cluster Headache Is Diagnosed
Diagnosis is clinical, based on the ICHD-3 criteria published by the International Headache Society [1]. No blood test or imaging finding confirms cluster headache. Instead, the diagnosis requires at least five attacks of severe unilateral orbital, supraorbital, or temporal pain lasting 15 to 180 minutes, accompanied by at least one ipsilateral autonomic feature or restlessness, with attacks occurring at a frequency between one every other day and eight per day.
Brain MRI with and without gadolinium is recommended at least once to rule out structural lesions, particularly pituitary adenomas and posterior fossa pathology that can mimic cluster headache [7]. A 2006 retrospective series found that 3.8% of patients initially diagnosed with cluster headache had a secondary cause identified on MRI [10].
A headache diary is one of the most useful diagnostic tools. Recording the time of day, duration, side affected, autonomic features, and response to treatment helps distinguish cluster headache from other trigeminal autonomic cephalalgias such as paroxysmal hemicrania (which responds to indomethacin) and SUNCT/SUNA syndromes (which feature shorter, more frequent attacks).
Acute Treatment That Works
High-flow oxygen is first-line. Inhaling 100% oxygen at 12 to 15 L/min through a non-rebreather mask for 15 minutes aborts an attack in approximately 78% of patients within 15 minutes, according to a randomized, double-blind, placebo-controlled crossover trial published in JAMA [11]. The mechanism likely involves cerebral vasoconstriction and inhibition of trigeminal nociceptive neurons. Oxygen has no significant side effects and can be used for multiple daily attacks without concern for medication overuse.
Subcutaneous sumatriptan 6 mg is the fastest pharmacological option. A Cochrane review found that sumatriptan injection produced pain-free status at 15 minutes in 48% of attacks versus 17% for placebo (NNT = 3.3) [12]. Sumatriptan nasal spray 20 mg is an alternative when injection is impractical, though onset is slower (approximately 30 minutes).
Intranasal zolmitriptan 5 mg or 10 mg is also effective. A randomized trial demonstrated a 40% response rate at 30 minutes for the 10 mg dose [13]. Oral triptans are generally too slow for cluster attacks, which often peak and resolve before oral absorption occurs.
Lidocaine 4% intranasal solution applied to the ipsilateral sphenopalatine fossa provides moderate benefit as an adjunctive measure but is not sufficient as monotherapy [3].
Over-the-counter analgesics such as ibuprofen, acetaminophen, and aspirin do not work for cluster headache. Their onset of action is far too slow, and reliance on them delays access to effective treatment.
Preventive Medications to Discuss with Your Doctor
Verapamil is the standard preventive agent. The European Headache Federation recommends verapamil as first-line prophylaxis for both episodic and chronic cluster headache, at doses of 240 to 960 mg daily in divided doses [14]. An open-label study by Leone et al. showed that 69% of episodic cluster headache patients achieved complete or near-complete suppression of attacks at a mean dose of 360 mg/day [14]. ECG monitoring is required before initiation and at each dose increase because verapamil can cause heart block.
Short-course corticosteroids serve as a bridge therapy while verapamil is titrated. Prednisone 60 to 80 mg daily tapered over 2 to 3 weeks provides rapid suppression but cannot be used long-term due to bone loss, hyperglycemia, and adrenal suppression [3].
Greater occipital nerve (GON) blocks with a local anesthetic and corticosteroid offer another transitional option. A randomized controlled trial showed that a single suboccipital injection of cortivazol reduced attack frequency by 57% over 4 weeks compared to saline [15].
For patients with chronic cluster headache refractory to verapamil, options include lithium carbonate (300 to 900 mg daily, monitored with serum levels and thyroid function tests), topiramate (100 to 200 mg daily), and galcanezumab. In the CGAL study, galcanezumab 300 mg subcutaneously monthly reduced weekly attack frequency by 8.7 attacks compared to 5.2 for placebo (P = 0.04) in episodic cluster headache, though it did not show superiority in chronic cluster headache [16].
Lifestyle and Trigger Management During a Cluster Period
Alcohol is the most reliable trigger during an active cluster bout. Even a single glass of wine or beer can provoke an attack within 30 to 60 minutes during a cluster period. Complete alcohol avoidance during bouts is strongly recommended [3].
Maintain a rigid sleep schedule. Because the hypothalamus drives both circadian rhythm and cluster cycling, disruptions to sleep timing (shift work, jet lag, weekend sleep-ins) can increase attack frequency. Going to bed and waking at the same time every day, including weekends, is a practical measure with biological rationale.
Avoid daytime napping during cluster periods. Brief naps are a common attack trigger, likely because they engage the same hypothalamic sleep-wake transitions that occur at night [3].
Smoking cessation does not stop cluster periods (a common misconception), but it does reduce overall cardiovascular risk in a population already at higher risk for coronary events [6]. Patients should not expect that quitting smoking will eliminate their clusters.
Strong scents, volatile solvents, and high altitudes have all been reported as triggers, though evidence is anecdotal. Keeping a headache diary during a cluster bout helps identify personal triggers.
Emerging Treatments and Research
Non-invasive vagus nerve stimulation (nVNS) with the gammaCore device received FDA clearance for adjunctive treatment of episodic cluster headache in 2017. The ACT1 randomized controlled trial demonstrated that nVNS increased the proportion of pain-free attacks at 15 minutes compared to sham (34.2% vs. 10.6%, P < 0.001) [17].
Sphenopalatine ganglion (SPG) stimulation via an implanted microstimulator (the Pulsante device) showed promise in the Pathway CH-1 study. Full stimulation achieved pain relief in 67.1% of attacks compared to 7.4% for sham, and 43% of patients experienced a >50% reduction in attack frequency [18].
Psilocybin and other 5-HT2A agonists are under active investigation. A retrospective survey published in Neurology reported that 52% of cluster headache patients who used psilocybin experienced termination of a cluster period, and 88% reported reduced severity or frequency of attacks [19]. Controlled trials are ongoing at Yale and other centers, but psilocybin remains a Schedule I substance in most jurisdictions, and self-treatment carries legal and medical risks.
Anti-CGRP monoclonal antibodies beyond galcanezumab (fremanezumab, erenumab) are being studied for cluster headache, though results so far have been mixed. The biology of cluster headache is distinct from migraine, and response to CGRP-targeted therapies may differ between the two conditions.
The Cost of Delayed Diagnosis
The 5.3-year average delay between symptom onset and correct diagnosis has measurable consequences [9]. Patients with undiagnosed cluster headache report higher rates of depression, job loss, relationship breakdown, and suicidal ideation. A cross-sectional survey of 1,134 cluster headache patients found that 55% had experienced suicidal thoughts during a cluster bout, compared to a general population rate of approximately 4% [20].
Early diagnosis shortens this suffering. Access to oxygen therapy alone can reduce attack duration from 60 to 90 minutes to under 15 minutes. Verapamil prophylaxis can eliminate attacks entirely for the duration of a cluster period. Neither treatment is available to patients who have not yet received a correct diagnosis.
If your headaches match the pattern described in this article (severe one-sided orbital pain, 15 to 180 minutes, autonomic features, clocklike recurrence), bring this information to your doctor and request a referral to a headache specialist or neurologist. A headache diary covering at least two weeks of attacks is the single most useful thing you can bring to that appointment.
Frequently asked questions
›What causes cluster headache?
›How is cluster headache diagnosed?
›When should I worry about cluster headache?
›Is cluster headache the same as migraine?
›Can cluster headache be cured?
›Does oxygen therapy really work for cluster headache?
›Why do cluster headaches happen at the same time every day?
›What medications prevent cluster headache?
›Can alcohol trigger cluster headache?
›How long does a cluster headache last?
›Is cluster headache hereditary?
›Are there any new treatments for cluster headache?
References
- Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition. Cephalalgia. 2018;38(1):1-211. https://pubmed.ncbi.nlm.nih.gov/29368949
- Louter MA, Wilbrink LA, Haan J, et al. Cluster headache and depression. Neurology. 2016;87(18):1899-1906. https://pubmed.ncbi.nlm.nih.gov/27694266
- 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(10):1066-1077. https://pubmed.ncbi.nlm.nih.gov/16987158
- May A, Bahra A, Büchel C, Frackowiak RSJ, Goadsby PJ. Hypothalamic activation in cluster headache attacks. Lancet. 1998;352(9124):275-278. https://pubmed.ncbi.nlm.nih.gov/9690407
- Goadsby PJ, Edvinsson L. Human in vivo evidence for trigeminovascular activation in cluster headache. Brain. 1994;117(3):427-434. https://pubmed.ncbi.nlm.nih.gov/7518321
- Russell MB. Epidemiology and genetics of cluster headache. Lancet Neurol. 2004;3(5):279-283. https://pubmed.ncbi.nlm.nih.gov/15099542
- Dodick DW. Pearls: headache. Semin Neurol. 2010;30(1):74-81. https://pubmed.ncbi.nlm.nih.gov/20127586
- Goadsby PJ. Pathophysiology of cluster headache: a trigeminal autonomic cephalgia. Lancet Neurol. 2002;1(4):251-257. https://pubmed.ncbi.nlm.nih.gov/12849458
- Voiticovschi-Iosob C, Allena M, De Cillis I, et al. Diagnostic and therapeutic errors in cluster headache: a hospital-based study. J Headache Pain. 2014;15(1):56. https://pubmed.ncbi.nlm.nih.gov/25178541
- Mainardi F, Trucco M, Maggioni F, Palestini C, Dainese F, Zanchin G. Cluster-like headache: a comprehensive reappraisal. Cephalalgia. 2006;26(10):1165-1175. https://pubmed.ncbi.nlm.nih.gov/16961782
- Cohen AS, Burns B, Goadsby PJ. High-flow oxygen for treatment of cluster headache: a randomized trial. JAMA. 2009;302(22):2451-2457. https://pubmed.ncbi.nlm.nih.gov/19996400
- Law S, Derry S, Moore RA. Sumatriptan plus naproxen for the treatment of acute migraine attacks in adults. Cochrane Database Syst Rev. 2016;4:CD008541. https://pubmed.ncbi.nlm.nih.gov/27096438
- Cittadini E, May A, Straube A, Evers S, Bussone G, Goadsby PJ. Effectiveness of intranasal zolmitriptan in acute cluster headache. Neurology. 2006;67(6):1016-1017. https://pubmed.ncbi.nlm.nih.gov/17000972
- Leone M, D'Amico D, Frediani F, et al. Verapamil in the prophylaxis of episodic cluster headache. Neurology. 2000;54(6):1382-1385. https://pubmed.ncbi.nlm.nih.gov/10746617
- 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(1-2):92-96. https://pubmed.ncbi.nlm.nih.gov/16202532
- Goadsby PJ, Dodick DW, Leone M, et al. Trial of galcanezumab in prevention of episodic cluster headache. N Engl J Med. 2019;381(2):132-141. https://pubmed.ncbi.nlm.nih.gov/31291515
- Silberstein SD, Mechtler LL, Kudrow DB, et al. Non-invasive vagus nerve stimulation for the acute treatment of cluster headache: findings from the randomized, double-blind, sham-controlled ACT1 study. Headache. 2016;56(8):1317-1332. https://pubmed.ncbi.nlm.nih.gov/27593728
- Schoenen J, Jensen RH, Lantéri-Minet M, et al. Stimulation of the sphenopalatine ganglion (SPG) for cluster headache treatment: Pathway CH-1. Cephalalgia. 2013;33(10):816-830. https://pubmed.ncbi.nlm.nih.gov/23314784
- Sewell RA, Halpern JH, Pope HG Jr. Response of cluster headache to psilocybin and LSD. Neurology. 2006;66(12):1920-1922. https://pubmed.ncbi.nlm.nih.gov/16801660
- Rozen TD, Fishman RS. Cluster headache in the United States of America: demographics, clinical characteristics, triggers, suicidality, and personal burden. Headache. 2012;52(1):99-113. https://pubmed.ncbi.nlm.nih.gov/22077141