Cluster Headache

National Organization for Rare Disorders, Inc.

Skip to the navigation


It is possible that the main title of the report Cluster Headache is not the name you expected. Please check the synonyms listing to find the alternate name(s) and disorder subdivision(s) covered by this report.


  • familial cluster headaches
  • histamine cephalalgia
  • vasogenic facial pain

Disorder Subdivisions

  • None

General Discussion


Cluster headaches (CH) are an uncommon, severe form of primary neurovascular headaches. CH are the most painful form of headaches, with the pain occurring on one side of the head and behind or above the eye or at the temple most commonly. The pain has been described as searing, burning and stabbing. The age of onset of CH is most often between 20 and 40, and they are more common in men than women with a ratio of 2:1. CH attacks also include one or more cranial autonomic symptoms (CAS) on the same side of the head as the pain (ipsilaterally) such as red eye (conjunctival injection), eyelid swelling (edema), forehead and facial sweating, tearing (lacrimation), abnormally small size of the pupil (miosis), nasal congestion, runny nose (rhinorrhea), and drooping eyelid (ptosis). CH is divided into episodic and chronic. Episodic cluster headache patients usually suffer from 1 to 4 short headaches a day that can individually last between 15-120 minutes when they are having a series of attacks. These attacks (cluster periods) last for weeks or months and are separated by months or years of remission periods where the patients are pain-free. Chronic cluster headache patients suffer without remissions for 1 year or more or with remissions so brief they do not even span a month. Less than 20% of cluster headache patients have the chronic form. It is not yet clear what causes CH, but scientists have discovered a lot of recent evidence that links it to the part of the brain called the hypothalamus. There is no cure, and treatment is determined on an individual basis. However, the two most effective types of acute or symptomatic treatment, high-flow inhaled oxygen and injections of subcutaneous sumatriptan, have been proven to be effective in reducing the pain from CH.


Cluster headaches were first completely described by the London neurologist Wilfred Harris in 1926. At that time, it was believed that nearly 90% of CH patients were men, while women instead suffered with migraine headaches. Although it has been confirmed that men are more at risk for developing CH, that ratio has decreased significantly. Researchers theorize that this is because women have long been misdiagnosed with migraines instead. Treatment for migraine headaches differ in several ways from CH, so an accurate diagnosis is important.


In most cases, cluster headaches occur more often at night than during the day, and more often in the spring and fall. Headaches also often occur at the same time every day. The headaches usually last approximately 30 minutes, but can last as long as two hours and uncommonly longer. The pain of a cluster headache is deep, agonizing, usually non-throbbing and affects only one side of the head (unilateral). The area of the head or face involved is always the same from day to day, but occasionally switches to the other side before a cluster ends entirely. The cluster can switch sides with the next onset of attacks. The pain usually occurs in the eye socket area (orbit) and spreads to affect the face, temple and/or forehead. Often the headache is tolerated better by moving, pacing, rocking or even banging their head and usually not by remaining still. They may be irritable and aggressive. When the headache passes, affected individuals may fall into a deep sleep only to be awakened again by another headache. This may continue several times a day, sometimes for weeks or months. Some affected individuals may be unable to continue a normal work schedule. It is important to note that both drinking alcohol and smoking can act as triggers and can bring on attacks in persons who suffer from CH.

Other symptoms include excessive watering of the eye and nose on the same side of the head as the pain occurs. Facial sweating, nasal congestion, drooping eyelids (ptosis), and eyelid swelling (edema) are also common. Rarely before an attack patients will report seeing an aura or having other visual disturbances.

Chronic and episodic CH patients both experience the same type of severe, unilateral pain, which generally occurs 1 to 4 times a day for about 30 minutes. However, chronic CH patients do not go through periods of remission or these periods are shorter than 1 month, while cyclic CH patients may experience episodic attacks for weeks or months but are then relieved by symptom-free periods that last months or years.


While the underlying cause of CH is not known, the direct cause of the pain is due to the dilation of blood vessels which creates pressure on the trigeminal nerve. Recently, position emission tomography (PET) scans have shown activity in the hypothalamus, the part of the brain responsible for circadian rhythms, or the biological clock, during CH attacks. This reinforces a previous hypothesis that the hypothalamus may be the cluster headache generator.

Although originally not thought to be a genetic disease, researchers now know that genetic factors are involved in approximately 10% of CH cases. It has been suggested that the disorder can be inherited as an autosomal dominant trait that appears to skip generations (incomplete penetrance) or as an autosomal recessive trait.

Dominant genetic disorders occur when only a single copy of an abnormal gene is necessary to cause a particular disease. The abnormal gene can be inherited from either parent or can be the result of a new mutation (gene change) in the affected individual. The risk of passing the abnormal gene from affected parent to offspring is 50% for each pregnancy. The risk is the same for males and females.

Recessive genetic disorders occur when an individual inherits two copies of an abnormal gene for the same trait, one from each parent. If an individual receives one normal gene and one gene for the disease, the person will be a carrier for the disease but usually will not show symptoms. The risk for two carrier parents to both pass the defective gene and have an affected child is 25% with each pregnancy. The risk to have a child who is a carrier like the parents is 50% with each pregnancy. The chance for a child to receive normal genes from both parents and be genetically normal for that particular trait is 25%. The risk is the same for males and females.

All individuals carry 4-5 abnormal genes. Parents who are close relatives (consanguineous) have a higher chance than unrelated parents to both carry the same abnormal gene, which increases the risk to have children with a recessive genetic disorder.

CH has also been associated with tobacco smoking as well as alcohol consumption.

Affected Populations

Cluster headaches affect males more often than females with a ratio that is 2:1. Symptoms usually start to appear between the ages of 20-40 with a mean age of onset of 30.

Standard Therapies


Currently there are no diagnostic tests to confirm CH.

According to the criteria of the International Classification of Headache Disorders, at least 5 attacks meeting the following are required for diagnosis:

1 - Severe unilateral, orbital (around the eye) and/or temporal (around the temple) pain lasting 15-180 minutes if untreated.

2 - Headache must be accompanied by at least one of the following:

·Red eye or tearing on the side of the headache

·Nasal congestion or runny nose on the side of the headache

·Eyelid swelling on the side of the headache

·Forehead and facial sweating on the side of the headache

·Small pupil or eyelid droop on the side of the headache

·A sense of restlessness or agitation

3 - Attacks have a frequency from one every other to 8 per day

Clinical Testing and Work-Up

Brain imaging with MRI with and without contrast is used at presentation of the very first cycle or when CH patients do not respond well to any of the common treatments, in order to exclude an alternative cause. Imaging studies in CH patients are normal most of the time, but lesions responsible for the headache can be seen leading to diagnosis of a secondary cluster headaches.


Treatment for CH varies greatly from patient to patient, depending on what works best to relieve pain for each individual. A common form of acute treatment at the first sign of pain symptoms is pure oxygen inhalation at 7-15 liters per minute by facial mask. Often this is successful at aborting an attack. Another effective, acute treatment is subcutaneous sumatriptan. When injected or inhaled, this drug acts on several different processes involving blood vessels and pain nerves. Both of these treatments narrow or constrict blood vessels, but the exact mechanism by which headache is stopped is not known. . Drugs that are used to end the cluster episode are grouped as transitional and chronic prevention. Occipital nerve block may terminate an episode. Corticosteroids are commonly used transitionally for up to three weeks. Verapamil is the customary chronic preventative, while others include antiepileptic drugs and melatonin. Transitional rugs are slowly tapered off even with continued headaches, while preventatives discontinued about two weeks after the last headache for episodic cluster. They are continued long term in chronic cluster. In cases of chronic CH only, lithium may be used.

Investigational Therapies

In extreme cases, electrical stimulation of the occipital nerve, deep brain stimulation or surgery to destroy certain facial nerves (radiofrequency retrogasserian rhizotomy) may provide relief.

Information on current clinical trials is posted on the Internet at All studies receiving U.S. government funding, and some supported by private industry, are posted on this government web site.

For information about clinical trials being conducted at the NIH Clinical Center in Bethesda, MD, contact the NIH Patient Recruitment Office:

Tollfree: (800) 411-1222

TTY: (866) 411-1010


For information about clinical trials sponsored by private sources, contact:

Contact for additional information about cluster headache:

Frederick R. Taylor, MD FAAN, FAHS

Park Nicollet Headache Center

Adjunct Professor of Neurology

University of Minnesota

Minneapolis, Minnesota 55426



Sargeant LK, Blanda M. Headache, Cluster. Emedicine. Updated May 20, 2012. Accessed May 25, 2012.

Robert T. Cluster Headache. The American Headache Society. Edited August 14, 2010. Accessed May 25, 2012.

Leroux E, Ducros A. Cluster headache. Orphanet Journal of Rare Diseases. Published July 23, 2008. Accessed May 25, 2012.

Cluster Headaches. National Headache Foundation. Accessed May 25, 2012.

Online Mendelian Inheritance in Man (OMIM). The Johns Hopkins University.

Cluster Headache, Familial. Entry No: 119915. Last Edited October 4, 2007. Available

at: Accessed May 25, 2012.


National Headache Foundation

820 N. Orleans

Suite 411

Chicago, IL 60610-


Tel: (312)274-2650

Fax: (312)460-9049

Tel: (888)643-5552



NIH/National Institute of Neurological Disorders and Stroke

P.O. Box 5801

Bethesda, MD 20824

Tel: (301)496-5751

Fax: (301)402-2186

Tel: (800)352-9424

TDD: (301)468-5981


American Council for Headache Education

19 Mantua Road

Mount Royal, NJ 08061

Tel: (856)423-0043

Fax: (856)423-0082

Tel: (800)255-2243



Organization for Understanding Cluster Headaches (O.U.C.H.)

c/o Sandy Cullen P.O. Box 902

Jamestown, NC 27282


Tel: (214)783-1899



Genetic and Rare Diseases (GARD) Information Center

PO Box 8126

Gaithersburg, MD 20898-8126

Tel: (301)251-4925

Fax: (301)251-4911

Tel: (888)205-2311

TDD: (888)205-3223


The Cluster Headache Support Group, Inc.

PO Box 21

Fountainville, PA 18923

Tel: (267)293-9689

Fax: (215)249-1993

Email: or


For a Complete Report

This is an abstract of a report from the National Organization for Rare Disorders, Inc.® (NORD). Cigna members can access the complete report by logging into For non-Cigna members, a copy of the complete report can be obtained for a small fee by visiting the NORD website. The complete report contains additional information including symptoms, causes, affected population, related disorders, standard and investigational treatments (if available), and references from medical literature. For a full-text version of this topic, see