Headache • TAC • Reference material

Trigeminal Autonomic Cephalalgias: Cluster Headache, SUNCT and SUNA

Trigeminal autonomic cephalalgias (TACs) are a group of primary headache disorders characterised by strictly or predominantly unilateral pain together with ipsilateral cranial autonomic symptoms. The group includes cluster headache, paroxysmal hemicrania, hemicrania continua, and short-lasting unilateral neuralgiform headache attacks classified as SUNCT and SUNA.

These syndromes differ primarily in attack duration and frequency, clinical course and characteristic treatment response. Accurate recognition of the phenotype is therefore essential.

Classification

What belongs to the trigeminal autonomic cephalalgias?

TACs are a small but clinically important group of primary headache disorders. They are unified by activation of trigeminal-autonomic mechanisms, manifesting as unilateral pain together with cranial autonomic features on the same side.

  • Cluster headache — the best-known TAC, with attacks lasting 15–180 minutes.
  • Paroxysmal hemicrania — shorter, more frequent attacks with an absolute response to indomethacin.
  • Hemicrania continua — continuous strictly unilateral headache with exacerbations, also showing an absolute response to indomethacin.
  • SUNCT and SUNA — very short, usually frequent unilateral attacks with cranial autonomic symptoms.
Cranial autonomic symptoms may include lacrimation, conjunctival injection, nasal congestion or rhinorrhoea, eyelid oedema, facial sweating, miosis and ptosis. The presence of these features alone does not establish a TAC diagnosis: the entire temporal and clinical pattern matters.
Cluster headache

Clinical characteristics

Cluster headache is among the most intense primary pain disorders. A typical attack consists of severe or very severe strictly unilateral pain in the orbital, supraorbital and/or temporal region.

15–180 min Typical duration of an untreated attack.
1 every other day – 8/day Formal attack-frequency range in ICHD-3.
Unilateral Orbital, supraorbital and/or temporal pain.
Restlessness Motor agitation and inability to remain still are characteristic.

One or more cranial autonomic symptoms usually occur on the painful side: lacrimation, conjunctival injection, nasal congestion, rhinorrhoea, eyelid oedema, facial sweating, miosis or ptosis.

Unlike a typical migraine attack, during which movement commonly worsens pain and the patient often prefers to lie still, cluster headache usually produces marked motor restlessness: patients may pace, rock, or find it difficult to remain seated or recumbent.

Circadian pattern and cluster periods

Cluster headache shows a striking temporal organisation. Attacks may occur at approximately the same time each day, frequently at night, and active periods may recur seasonally.

During an active cluster period, alcohol can rapidly trigger an attack in many patients, whereas this association typically disappears during remission.

Side of pain

Within a single cluster period, attacks almost always occur on the same side. In most patients the side remains stable, although in some individuals it may alternate between different cluster periods.

Epidemiology

Cluster headache is rare compared with migraine. Onset most commonly occurs between 20 and 40 years of age. Historically, a marked male predominance was reported, although sex differences appear less pronounced in more recent clinical series.

Psychological impact

The intensity of pain, recurrent nocturnal attacks and anticipation of the next attack may lead to marked anxiety, depressive symptoms and substantial impairment in quality of life.

Suicidal ideation in cluster headache is a clinically important problem. Reported prevalence varies substantially between studies and recruitment settings, but particularly high rates have been described among severely affected patients seen in specialist centres. Any immediate risk of self-harm or suicide requires urgent intervention.
ICHD-3

Diagnostic criteria and forms of cluster headache

The diagnosis is clinical. ICHD-3 requires at least five attacks meeting the characteristic temporal and clinical features.

Core attack criteria

  • severe or very severe unilateral orbital, supraorbital and/or temporal pain;
  • duration 15–180 minutes when untreated;
  • ipsilateral cranial autonomic features and/or marked restlessness or agitation;
  • frequency from one attack every other day to 8 attacks per day;
  • the disorder is not better accounted for by another diagnosis.

Variation is possible

During a smaller part of the active disease period, attacks may be shorter or longer, less severe, or occur somewhat more or less frequently than the formal limits.

An isolated deviation from the “ideal” temporal profile does not necessarily exclude the diagnosis.

Episodic cluster headache

In episodic cluster headache, active periods last from 7 days to 1 year and are separated by remissions lasting at least 3 months. In routine practice, cluster periods more commonly last several weeks to several months.

Chronic cluster headache

In chronic cluster headache, attacks continue for at least 1 year without remission or with remission periods lasting less than 3 months.

Differential diagnosis

Cluster headache, other TACs and trigeminal neuralgia

Attack duration, attack frequency, presence of continuous background pain, mechanical triggers, and a characteristic response to indomethacin are particularly useful in practical differential diagnosis.

Syndrome Typical duration Frequency / course Autonomic features Triggers / behaviour Key feature
Cluster headache 15–180 min 1 every other day to 8/day; often in cluster periods Prominent, ipsilateral Marked restlessness; alcohol may trigger attacks during an active bout Response to oxygen and subcutaneous sumatriptan; verapamil for prevention
Paroxysmal hemicrania 2–30 min Usually more than 5/day Often prominent May be associated with restlessness Absolute response to indomethacin
Hemicrania continua Continuous pain with exacerbations Continuous unilateral pain for more than 3 months May occur during exacerbations Restlessness or worsening with movement may occur during exacerbations Absolute response to indomethacin
SUNCT / SUNA 1–600 sec At least 1/day by criteria, often dozens per day At least one cranial autonomic symptom is required Often triggered by touch, chewing, washing the face Very short neuralgiform attacks; lamotrigine is the principal long-term preventive option
Trigeminal neuralgia Fraction of a second – 2 min Repeated short paroxysms Usually not dominant Typical trigger zones: touch, shaving, tooth brushing, chewing Electric shock-like or shooting pain in one or more trigeminal divisions
SUNCT/SUNA and trigeminal neuralgia may be particularly difficult to distinguish because both can produce brief neuralgiform attacks with mechanical triggers. Prominent cranial autonomic symptoms favour SUNCT/SUNA.
Safety

When should a secondary cause be considered?

Even a very typical clinical phenotype does not remove the need to consider secondary causes, particularly at first presentation or when the clinical course is atypical.

  • a new sudden headache of maximal intensity;
  • new persistent focal neurological deficit;
  • persistent diplopia or visual loss;
  • fever or systemic illness;
  • a new persistent Horner syndrome outside typical attacks;
  • new neck pain associated with unilateral headache;
  • a substantial change in the usual phenotype;
  • onset at an unusual age or other atypical features;
  • altered consciousness;
  • suicidal ideation or immediate risk of self-harm.

The choice of urgent investigation depends on the clinical situation and may include CT, MRI and vascular imaging.

See also: new headache and red flags .

Diagnosis

MRI in trigeminal autonomic cephalalgias

TACs are primary headache disorders, but a similar phenotype may rarely occur as a manifestation of structural disease. Neuroimaging therefore remains important at first diagnosis, in atypical cases, or when the neurological examination is abnormal.

Standard brain MRI

In most situations, high-quality brain MRI, with contrast when appropriate, forms the basis of structural assessment.

Particular attention is paid to regions capable of mimicking a TAC phenotype, including the sellar and parasellar regions, cavernous sinus, posterior fossa and trigeminal nerve pathways.

Does everyone need dedicated pituitary MRI?

Routine dedicated pituitary imaging was previously proposed for all patients with cluster headache. Later data do not support a separate pituitary MRI protocol for every patient.

A dedicated pituitary protocol is more appropriate when standard imaging shows an abnormality in this region or when symptoms, history or laboratory findings suggest pituitary disease.

When vascular imaging may be required

MRA or CTA may be appropriate when a vascular cause is suspected, for example with an atypical Horner syndrome, prominent neck pain, suspected arterial dissection or other red flags.

Additional investigations

Hormonal testing, sleep studies or other investigations are not required routinely. They are selected when the history, examination or imaging raises a specific clinical question.

Current therapy

Treatment of cluster headache

Management has three major goals: to stop an individual attack rapidly, to suppress attacks during the beginning of an active cluster period, and to reduce subsequent attack frequency.

1. Acute treatment

The best-supported options are:

  • 100% oxygen at a flow rate of at least 12 L/min for approximately 15 minutes, usually through a non-rebreather mask;
  • subcutaneous sumatriptan 6 mg — one of the fastest pharmacological methods for terminating an attack.

An intranasal triptan may be an alternative in selected situations, although onset is generally slower.

2. Transitional treatment

Transitional therapy is used to suppress attacks quickly while preventive therapy is beginning to take effect.

  • a short course of systemic corticosteroids;
  • greater occipital nerve block, typically using local anaesthetic and/or corticosteroid according to the protocol used.

Transitional treatment does not replace longer-term prevention.

3. Preventive treatment

Verapamil remains the principal preventive medication. Doses may exceed those commonly used in routine cardiovascular practice, so ECG monitoring is required before and during dose escalation.

Alternatives include lithium and topiramate, depending on the clinical context.

Galcanezumab

Galcanezumab, a monoclonal antibody targeting CGRP, demonstrated efficacy in a clinical trial of episodic cluster headache and is included in EAN guidance as an alternative option for the episodic form.

Its regulatory status for cluster headache differs between jurisdictions. For example, the EMA did not extend the European indication for Emgality to episodic cluster headache. Local registration therefore needs to be considered.

Non-invasive vagus nerve stimulation

According to EAN guidance, non-invasive vagus nerve stimulation has evidence of efficacy in episodic cluster headache, but comparable efficacy has not been demonstrated for chronic cluster headache.

Invasive neuromodulation

Invasive approaches are not routine therapy and are considered only for carefully selected patients with very severe, refractory disease in specialised headache centres.

Electrical stimulation of the greater occipital nerve is not recommended by EAN for routine use because of the balance between benefit and complications.

Conventional oral analgesics are generally not useful for aborting a cluster attack, because pain reaches peak intensity rapidly and the attack may resolve before an oral analgesic has had time to produce a meaningful effect.
Special situations

Children, pregnancy and refractory disease

Children and adolescents

Cluster headache can begin in childhood or adolescence, although it is much less common than in adults. Because of its rarity, diagnostic delay may be particularly pronounced in younger patients.

Pregnancy and breastfeeding

During pregnancy, treatment choices must be individualised according to gestational age, attack frequency and severity, and potential treatment risks. Oxygen has an obvious advantage as a non-pharmacological acute treatment.

Use of triptans and preventive medication requires an individual decision taking obstetric context and current safety data into account.

Refractory cluster headache

When severe disease does not respond to several adequately performed treatment strategies, the diagnosis should be reviewed, a secondary TAC phenotype reconsidered, and management discussed in a specialised headache centre.

SUNCT and SUNA

Short-lasting unilateral neuralgiform headache attacks

SUNCT and SUNA belong to the group of short-lasting unilateral neuralgiform headache attacks. They are characterised by very short attacks of moderate or severe unilateral pain in the distribution of the trigeminal nerve together with cranial autonomic symptoms.

1–600 sec Formal duration of an individual attack in ICHD-3.
Often many per day The formal minimum is at least one attack per day during active disease.
Neuralgiform pain Stabbing, shooting, electric shock-like; a saw-tooth pattern may occur.
Triggers Touch, chewing, washing the face and other mechanical stimuli.

SUNCT

SUNCT is defined by the presence of both conjunctival injection and lacrimation on the side of the pain.

SUNA

In SUNA, one or both of these two specific signs may be absent — that is, conjunctival injection and lacrimation are not both required.

This does not mean that cranial autonomic symptoms can be absent altogether: at least one ipsilateral cranial autonomic feature is still required for the diagnosis.

SUNCT/SUNA and trigeminal neuralgia

The clinical overlap with trigeminal neuralgia can be considerable. Both may produce brief electric shock-like pains triggered by touch or chewing.

Prominent cranial autonomic features and the characteristic attack pattern support SUNCT/SUNA, whereas classical trigeminal neuralgia more typically presents with triggerable electric shock-like pain without such a prominent autonomic component.

Neuroimaging

In SUNCT/SUNA, structural causes should be excluded, particularly lesions involving the posterior fossa, sellar and parasellar regions, and possible neurovascular compression of the trigeminal root.

The MRI protocol depends on the clinical context, and dedicated sequences for the trigeminal nerve and neurovascular relationships may be used when appropriate.

SUNCT / SUNA

Treatment

Because individual attacks are extremely short, conventional treatment of a single attack is generally impractical. The principal strategy is prevention.

Long-term prevention

The strongest contemporary observational evidence supports lamotrigine as the principal preventive treatment.

Oxcarbazepine is an important alternative. Topiramate, carbamazepine and several other antiepileptic medications may also be considered, although evidence and tolerability vary.

Severe exacerbation

During very frequent, severe and refractory attacks, intravenous lidocaine may be used as transitional treatment in an inpatient setting with appropriate monitoring.

The largest prospective series showed a substantial response in many patients, but this approach requires specialist use.

The response of SUNCT/SUNA to sodium-channel blockers highlights the clinical overlap between these syndromes and trigeminal neuralgia, although they remain distinct diagnostic entities.
Key points

What is important to remember

  • Trigeminal autonomic cephalalgias are a group of syndromes, not a single disease, and differ in attack duration, frequency and treatment response.
  • Cluster headache is characterised by very severe unilateral attacks lasting 15–180 minutes with cranial autonomic symptoms and/or marked motor restlessness.
  • Paroxysmal hemicrania and hemicrania continua are distinguished by an absolute response to indomethacin.
  • SUNCT and SUNA produce attacks lasting 1–600 seconds, often have mechanical triggers and require cranial autonomic symptoms.
  • A typical TAC phenotype is diagnosed clinically, but neuroimaging is important for excluding structural mimics in the appropriate clinical context.
  • Routine dedicated pituitary MRI is not required for every patient with typical cluster headache when standard brain MRI is unremarkable and there are no features suggesting pituitary disease.
  • The principal acute treatments for cluster headache are 100% oxygen and subcutaneous sumatriptan; verapamil remains the main preventive treatment with appropriate ECG monitoring.
  • For SUNCT/SUNA, lamotrigine remains the best-supported long-term preventive medication, while intravenous lidocaine can be used for severe exacerbations in specialist settings.
References

Selected sources

  1. Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition (ICHD-3). Cephalalgia. 2018;38(1):1–211. doi:10.1177/0333102417738202
  2. May A, Evers S, Goadsby PJ, et al. European Academy of Neurology guidelines on the treatment of cluster headache. Eur J Neurol. 2023. PubMed
  3. Schindler EAD, Burish MJ. Recent advances in the diagnosis and management of cluster headache. BMJ. 2022;376:e059577. doi:10.1136/bmj-2020-059577
  4. Robbins MS, Starling AJ, Pringsheim TM, Becker WJ, Schwedt TJ. Treatment of cluster headache: the American Headache Society evidence-based guidelines. Headache. 2016;56(7):1093–1106. doi:10.1111/head.12866
  5. Grangeon L, O'Connor E, Danno D, et al. Is pituitary MRI screening necessary in cluster headache? Cephalalgia. 2021;41(7):779–788. doi:10.1177/0333102420983303
  6. Lambru G, Stubberud A, Rantell K, et al. Medical treatment of SUNCT and SUNA: a prospective open-label study including single-arm meta-analysis. J Neurol Neurosurg Psychiatry. 2021;92(3):233–241. doi:10.1136/jnnp-2020-323999
  7. Lee MJ, Cho SJ, Park JW, et al. Increased suicidality in patients with cluster headache. Cephalalgia. 2019;39(10):1249–1256.
  8. Benkli B, Kim SY, Koike N, et al. Circadian features of cluster headache and migraine: a systematic review, meta-analysis, and genetic analysis. Neurology. 2023;100(22):e2224–e2236.
  9. Winsvold BS, Harder AVE, Ran C, et al. Cluster headache genomewide association study and meta-analysis identifies eight loci and implicates smoking as causal risk factor. Ann Neurol. 2023;94(4):713–726.

This page is intended as medical reference material and does not replace individual diagnosis. Regulatory status of medications, availability of specific treatments and local clinical protocols may differ between countries.

Made on
Tilda