Cluster Headache: Labs, Diagnosis, and Next Steps

Why Cluster Headache Is a Clinical Diagnosis

Cluster headache is identified at the bedside, not in the lab. The ICHD-3 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 sign (lacrimation, conjunctival injection, nasal congestion, ptosis, miosis, eyelid edema, or forehead sweating) or a sense of restlessness 1. A European headache-center survey (N=1,163) reported that patients waited a mean of 5.3 years and consulted an average of 3.9 physicians before receiving the correct diagnosis 2.

The reason for this delay is simple. Cluster headache is rare. Its lifetime prevalence sits near 124 per 100,000, roughly 0.1% of the general population 3. Many clinicians see only a handful of cases across an entire career, and early attacks are frequently misattributed to migraine, sinusitis, or dental pathology. Autonomic features like tearing and nasal congestion reinforce the sinus-headache misdiagnosis in primary care settings.

Pattern recognition matters most. Attacks cluster in bouts lasting weeks to months, typically with circadian regularity, often striking 1 to 2 hours after sleep onset. That clockwork periodicity, paired with the ipsilateral autonomic signs and the patient's inability to stay still, separates cluster headache from migraine with high specificity.

Laboratory Tests: What to Order and Why

No blood marker diagnoses cluster headache. Labs serve one purpose: ruling out secondary headaches that mimic the cluster phenotype. A reasonable baseline panel includes a complete blood count (CBC), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), thyroid-stimulating hormone (TSH), and a basic metabolic panel.

ESR and CRP matter most in patients over 50. Giant cell arteritis (GCA) can present with strictly unilateral periorbital pain and autonomic features. The American College of Rheumatology notes that an ESR above 50 mm/h combined with new-onset headache in a patient older than 50 warrants urgent temporal artery biopsy or ultrasound 4. Missing GCA carries the risk of irreversible vision loss within days.

TSH screening catches thyroid dysfunction. Both hyperthyroidism and hypothyroidism alter headache frequency and can complicate verapamil dosing later.

Pituitary hormone panels (prolactin, IGF-1, morning cortisol, free T4) become necessary when MRI reveals a sellar or parasellar lesion. Pituitary adenomas are the single most common structural mimic of cluster headache. A 2020 review in Cephalalgia identified 84 published cases of "secondary cluster headache" attributable to pituitary pathology 5. If imaging is clean, pituitary labs are not needed.

For patients already on verapamil or being started on it, an ECG before initiation and after each dose increase is mandatory. Verapamil prolongs the PR interval and can cause heart block at doses above 480 mg/day. The European Headache Federation guideline recommends ECG monitoring at baseline, at 10 days after each uptitration, and whenever the dose exceeds 480 mg 6.

Brain Imaging: When and What to Request

Every patient with a new diagnosis of cluster headache should receive brain MRI with gadolinium contrast and dedicated pituitary cuts. This is not optional. The American Headache Society and European Headache Federation both recommend imaging at first presentation to exclude structural lesions in the cavernous sinus, hypothalamus, and pituitary 6.

The hypothalamus is central to cluster headache pathophysiology. PET imaging studies by May et al. demonstrated ipsilateral posterior hypothalamic activation during cluster attacks 7. Standard clinical MRI will not show this activation, but it will catch lesions that disrupt hypothalamic circuitry, including tumors, vascular malformations, and demyelinating plaques.

CT head is insufficient. It misses small pituitary adenomas, cavernous sinus lesions, and posterior fossa pathology. Reserve CT for emergency department settings where MRI is not immediately available and the clinical picture raises concern for subarachnoid hemorrhage or acute stroke.

MR angiography (MRA) of the circle of Willis adds value when the pain profile is atypical. A posterior communicating artery aneurysm or carotid dissection can produce periorbital pain with a partial Horner syndrome, closely mimicking cluster headache. If ptosis or miosis persists between attacks, order MRA or CT angiography of the head and neck.

Acute Treatment: Stopping an Attack in Progress

Two therapies carry grade A evidence for aborting cluster attacks. Both work fast.

High-flow oxygen at 12 to 15 L/min via a non-rebreather mask for 15 to 20 minutes relieves pain in approximately 78% of attacks within 15 minutes, per the randomized controlled trial by Cohen et al. (N=76) published in JAMA 8. Oxygen has no systemic side effects and no dose ceiling. Patients can use it multiple times per day. The practical barrier is access: home oxygen requires a prescription, a supplier, and a portable setup for use outside the home.

Subcutaneous sumatriptan 6 mg provides complete relief in 74% of attacks within 15 minutes 9. It is the fastest pharmacologic option. Sumatriptan nasal spray (20 mg) is an alternative but slower, with a 57% response rate at 30 minutes. Oral triptans are too slow for cluster attacks, which peak in minutes. The maximum daily dose of subcutaneous sumatriptan is 12 mg (two injections), and it is contraindicated in patients with uncontrolled hypertension, coronary artery disease, or prior stroke.

Dr. Peter Goadsby, a leading headache researcher at King's College London, described the acute treatment hierarchy: "Oxygen and injectable sumatriptan are the two pillars. If you don't have both available for your cluster headache patient, the management is incomplete" 10.

Intranasal zolmitriptan (5 mg) holds level B evidence as an alternative when sumatriptan is contraindicated. Lidocaine 4% to 10% applied intranasally to the ipsilateral sphenopalatine fossa provides modest adjunctive relief but should not replace oxygen or triptans as primary therapy.

Preventive Treatment: Shortening the Bout

Preventive therapy starts at bout onset for episodic cluster headache and runs continuously for the chronic form. The goal is reducing attack frequency and intensity across the entire cluster period.

Verapamil is the established first-line preventive. Starting dose is 240 mg/day in three divided doses, titrated upward by 80 mg every 10 to 14 days as needed. Many patients require 480 to 720 mg/day. Some need 960 mg/day. A retrospective analysis by Leone et al. found that 69% of patients achieved greater than 50% reduction in attack frequency on verapamil 11. ECG monitoring is required at each dose step due to the risk of cardiac conduction delays.

Galcanezumab (Emgality) earned FDA approval in 2019 specifically for episodic cluster headache, making it the only FDA-approved preventive for the condition. The phase 3 trial (N=106) demonstrated a reduction of 3.5 attacks per week compared to 0.2 for placebo over the first three weeks of treatment (P<0.001) 12. The dose is 300 mg subcutaneous (three consecutive 100 mg injections) at cluster bout onset. It did not meet its primary endpoint in chronic cluster headache.

Short-course corticosteroids bridge the gap while verapamil is being titrated. Prednisone 60 to 80 mg/day for five days, tapered over two to three weeks, suppresses attacks rapidly. A Cochrane review noted that steroids are widely used as transitional therapy despite limited randomized trial data 13. Repeated steroid courses risk avascular necrosis, hyperglycemia, and adrenal suppression.

Greater occipital nerve (GON) block with local anesthetic and corticosteroid provides 2 to 4 weeks of reduced attack frequency. A randomized trial by Ambrosini et al. (N=23) found that suboccipital steroid injection reduced attack frequency by over 50% in 85% of patients during the first month 14. It is safe, repeatable, and useful as a bridge.

Lithium, topiramate, and melatonin represent second-line options when verapamil fails or is contraindicated. Lithium requires serum level and renal function monitoring.

Identifying Triggers and Modifiable Risk Factors

Alcohol is the single most reliable cluster headache trigger during an active bout. Even small amounts of beer, wine, or spirits can provoke an attack within 30 to 60 minutes during a cluster period. Outside the bout, alcohol typically has no effect. Patients should abstain completely during active cycles.

Smoking prevalence among cluster headache patients is strikingly high. A meta-analysis by Rozen and Fishman found that 65% of cluster headache patients were current or former smokers compared to roughly 20% in the general population 15. Whether smoking is causal or a shared-risk association remains debated, but cessation is advised regardless.

Nitroglycerin and other nitric oxide donors reliably trigger attacks during active cluster periods. This sensitivity is so consistent that sublingual nitroglycerin has been used as a provocation test in research settings 16. Patients taking cardiac nitrates should discuss alternatives with their cardiologist during cluster bouts.

Sleep-wake disruption worsens cluster frequency. Attacks cluster around the first REM sleep period. Shift work, jet lag, and obstructive sleep apnea all increase bout severity. Polysomnography is reasonable in patients who snore, have witnessed apneas, or have refractory cluster headache. Treating comorbid sleep apnea with CPAP may reduce attack frequency.

When to Seek Emergency Care

Most cluster attacks, though excruciating, resolve on their own within 180 minutes. But certain presentations demand immediate evaluation.

The European Headache Federation guideline states: "Any patient presenting with a first-ever cluster-like headache requires neuroimaging before the diagnosis of primary cluster headache can be made" 6. The same urgency applies when an established patient develops a change in attack character: longer duration, new neurologic symptoms, or persistent deficits between attacks.

Red flags that require emergency imaging or referral:

• New-onset cluster-pattern headache after age 50
• Persistent Horner syndrome (ptosis and miosis) between attacks
• Papilledema or visual field loss
• Fever, weight loss, or elevated ESR with new headache
• Attacks lasting beyond 3 hours
• Failure to respond to previously effective oxygen or sumatriptan
• Focal weakness, speech difficulty, or seizure associated with headache

Suicidal ideation is a genuine risk. Cluster headache is sometimes called "suicide headache" for a reason. A survey of 1,134 cluster headache patients found that 55% had experienced suicidal thoughts during attacks 17. Ask about suicidal ideation directly. Connect patients with mental health support and ensure they have reliable acute abortive therapy available at home.

Building a Long-Term Management Plan

Cluster headache requires a coordinated approach between the patient, a headache specialist or neurologist, and sometimes a cardiologist (for verapamil monitoring) and a pulmonologist (for home oxygen).

The first visit should include: a detailed headache diary review, MRI brain with pituitary protocol if not already done, baseline ECG, and baseline labs (CBC, ESR, CRP, TSH, basic metabolic panel). Prescribe both acute therapies (oxygen prescription at 12 to 15 L/min and subcutaneous sumatriptan auto-injector) and initiate preventive therapy if the patient is in an active bout.

Follow-up at 2 weeks allows ECG reassessment after verapamil titration and attack-frequency review. Patients with episodic cluster headache who have predictable seasonal bouts can begin verapamil 2 weeks before the expected bout onset and taper it 2 weeks after attacks cease.

For chronic cluster headache, defined as remission periods lasting fewer than 3 months over one year, referral to a headache center is appropriate. Refractory chronic cases may be candidates for occipital nerve stimulation or sphenopalatine ganglion stimulation, both of which have shown efficacy in randomized controlled trials 18.

Patients should keep a headache diary tracking attack time, duration, severity, autonomic symptoms, and abortive response. Diary data guides preventive titration and identifies trigger patterns. The recommended minimum follow-up is every 3 months during active periods and annually during remission.