Cluster headaches are the most common of the primary headache type known as trigeminal autonomic cephalgia. They are relatively rare, plaguing 0.1% of the population, making a study of the disease process difficult at best. They have earned consideration as one of if not the most severe type of headache, and so, despite their rarity, recognition and treatment are vital. There is some degree of genetic correspondence, as first-degree family members are 18 times more likely to be diagnosed with cluster headaches. However, the mode of inheritance is unclear. Some families seem to pass the symptoms along with an autosomal recessive pattern, while others seem to be autosomal dominant.
The definition of cluster headaches is a unilateral headache with at least one autonomic symptom ipsilateral to the headache. These headaches occur between every other day up to eight times a day. They usually occur at approximately the same time of day, most often at night. Most patients are episodic, with daily attacks for weeks to months, followed by a remission for months to years.
The exact etiology of cluster headaches is not clear. There have been several theories suggested and tested. Unfortunately, due to the relative rarity of patients suffering from this condition, sample sizes are limited. There is a known association between the trigeminovascular system, the parasympathetic nerve fibers involved in the trigeminal autonomic reflex, and the hypothalamus. However, it is not certain how these structures interact to cause these headaches. Moreover, as there is a definite familial connection, considerable effort has gone into studying the genetics of the condition, which has also led to a biochemical assessment.
There is a definitive link between vasodilation and a pain attack. Activation of the trigeminovascular system causes perivascular afferent nerves to cause vasodilation. The belief is that the activation of the trigeminal nerve is unilateral; that said, it has not been detected and confirmed with functional imaging. Also, complete trigeminal nerve root section does not affect the number or frequency of attacks. This lack of change does not rule out the trigeminovascular system as a component of the cluster headache; it merely shows that there are other components.
The hypothalamus has a definite association with cluster headaches. The attacks have a circadian periodicity, happening most often at night. PET scans have shown activation of the inferior hypothalamic grey matter while the patient is having an attack. Morphology studies have also demonstrated anatomic abnormalities within this same region of the hypothalamus. Of note, though, stimulation of the hypothalamus does not trigger attacks. There is even some research suggesting that stimulation of the hypothalamus may abort an attack.
The parasympathetic nerve fibers are part of the trigeminal autonomic reflex, which is what causes the autonomic symptoms, including conjunctival injection or lacrimation, rhinorrhea, and facial vasodilation. Like the hypothalamus, it is a known component of the cluster headache, but the exact trigger of how the trigeminal reflex is activated is uncertain.
Cluster headaches occur in 0.1% of the general population. Any age is possible for onset, but the typical age is approximately 30 years old. Men are three times more likely to suffer from this condition than women, though current research has shown that the ratio has been decreasing over time. This incidence may be due to an improvement in diagnostic accuracy; often, women receive an erroneous diagnosis of migraines rather than cluster headaches. Of interest, 85% of all patients diagnosed with cluster headaches are cigarette smokers. Patients are 18 times more likely to be diagnosed if they have a 1st-degree relative who also had a diagnosis of cluster headaches. There is a broad spectrum in research suggesting anywhere between 30% to 80% of patients with cluster headaches will also have sleep apnea.
As stated above, there is no single clear source of cluster headaches. There is a circadian periodicity, leading to investigation of the hypothalamus, which ultimately shows abnormalities in morphology studies. There is also the involvement of the trigeminovascular system and the parasympathetic nerve fibers. Some research studies have suggested a defect in the central pathway of pain control and autonomic nervous system dysregulation leading to dysfunction in supraspinal control of pain and cognitive processing. Researchers have also noted dysfunction in inter and intracellular signaling pathways of GABA, ion channels, and inflammation-related molecules, including IL-2, adhesion molecules, and histamine.
Genetic analysis has many possible sources, including PER3 (associated with the circadian rhythm), orexin-B (associated with the sleep-wake cycle, food intake, and modulation of nociceptive neurotransmission), and PACAP receptor gene, being that PCAP (pituitary adenylate cyclase-activating polypeptide) increases in the blood during attacks. More research is necessary for all suggested genetic pathways.
A thorough history is essential to the diagnosis of cluster headaches. Patients will reveal that they experience ten out of ten unilateral pain, most often located intra or supraorbitally. Patients must have at least one autonomic symptom ipsilateral to the side with pain.
The autonomic symptoms are as follows:
Up to 33% of patients will also experience general allodynia; this is more common in women, patients with recent attacks, those who had a low age at the time of onset, and those with comorbid depression or migraines. There is often suicidal ideation associated with the attacks, but the risk of actual suicide is low.
Unlike patients with migraines, these patients are unlikely to hold still. Many patients pace or rock in place; there are descriptions of patients even banging their heads to distract themselves from the headache. Attacks typically last anywhere from 15 minutes to 3 hours. They can occur up to eight times a day, although most commonly, patients experience attacks twice a day, usually at night. Most patients will have attacks for weeks to months and then have remission for months up to years. This pattern usually lasts up to fifteen years.
The diagnosis is clinical. While there are some lab abnormalities, such as increased histamine in the blood during attacks, and anatomical changes found on a morphology study, these have not been found to be useful in a clinical setting. A questionnaire asking about headache duration (less than 180 minutes) and autonomic symptoms (conjunctival injection or lacrimation) has a sensitivity of 81% and a specificity of 100%. Clinicians must have a degree of suspicion, as cluster headaches are usually misdiagnosed, and patients do not receive a clinical diagnosis for up to five years.
There are two sides to the treatment of cluster headaches: treatment of the acute attack and preventative treatment.
Perhaps the most well-known treatment for cluster headaches is 100% oxygen therapy. This approach is a level A recommendation and is unique to cluster headaches vs. all other types of headaches. At least 66% of patients respond to oxygen therapy. It is effective in less than 10 minutes. Oxygen use carries no risks or side effects, making it an excellent choice for treatment. Unfortunately, insurance often will not pay for oxygen therapy for the cluster headache patient, and it can be difficult to obtain.
Triptans are the only other level A recommended treatment. Delivery of these drugs can be subcutaneous sumatriptan or zolmitriptan delivered via nasal spray. An oral medication of any form is not recommended, as the time of onset is often longer than the headache is.
Other treatment options include intranasal lidocaine (with reported 33% response), octreotide, and ergotamine. Unfortunately, 10% to 20% of patients with chronic cluster headaches develop drug resistance. Patients should be instructed to avoid triggers, especially alcohol. While there is no proof that smoking cessation will decrease the risk of headaches, patients should also be encouraged to stop smoking.
The only level A recommended treatment for the prevention of cluster headaches is a suboccipital blockade. Adverse events are nonserious, including transient injection site pain and low-level headache.
Verapamil is the most widely prescribed preventative drug. It is recommended to do regular ECGs to monitor cardiac function when a patient is using this drug. Curiously, despite verapamil's popularity among providers, it only has a level C recommendation.
Other pharmacological options include lithium, oral steroids, valproic acid, melatonin, and intranasal capsaicin.
There has been much research into electrical stimulation. Sites of stimulation include the sphenopalatine ganglion, occipital nerve, and vagus nerve. Deep brain stimulation of the hypothalamus has been very successful and has been proven quite useful in the treatment of drug-resistant patients. Vagus nerve stimulation can be an option with a nonimplanted device.
Cluster headache often goes undiagnosed for years. Other differential diagnoses are as follows:
Episodic: 85% to 90% of patients. Patients experience daily attacks for weeks to months and then have a remission lasting months or years. Patients typically have 1 to 2 episodes a year, most often in spring or autumn.
Chronic: 15% to 20% of patients. Patients experience attacks lasting more than a year without remission or remission that lasts less than one month;10% to 20% of these patients will develop drug resistance.
Probable: patients whose attacks fulfill all but one criterion.
Around 25% of patients who experience a cluster headache will never have another one. Another 15% to 20% of patients will have chronic cluster headaches, 10% to 20% of which will develop drug resistance. This condition often resolves in approximately 15 years.
Patients can experience detriments to their mental health. There will be physical burdens and loss of quality of life. Many patients describe suicidal ideation, but actual follow-through action is rare.
There is a frequent association with psychiatric conditions, including depression, anxiety, and aggressive behavior. Suicidal attempts are rare, but these conditions do increase the disease burden. There is evidence supporting systemic autonomic function being affected, which is usually subclinical and includes bradycardia, tachycardia, hypertension, and arrhythmias (such as AV block and SA block). The most concerning of the autonomic dysregulation is an increased risk of dysregulation of systemic blood pressure. As in sleep apnea, patients have an increased risk of not undergoing physiological blood pressure dipping during sleep, which can lead to end-organ damage and increases the risk of cardiovascular disease. This situation is accompanied by an increased risk of an abnormal ankle-brachial index (ABI), indicating peripheral arterial disease, which confirms that there is also an increased risk for cardiovascular disease.
Alcohol is the most common trigger for a pain attack. Most patients recognize this trigger and stop alcohol use on their own, but it may be necessary to provide patients with cessation counseling. While no evidence exists that quitting smoking affects the number of pain attacks, patients should be encouraged to stop. Patients should also be educated on pharmacological triggers, specifically vasodilators. Patients may require aid in lifestyle management, and patients should receive an offer for psychiatric counseling.
The average cluster headache patient can take approximately five years to diagnose and only after visiting several different clinicians. Cluster headaches are often confused for migraines. The only way to improve diagnosis and management is with an interprofessional team dedicated to the treatment of headaches.
The healthcare provider (i.e., physician, nurse practitioner) must strive to recognize the episodic nature of cluster headaches, autonomic dysregulation, and even the activity of the patient during an attack.
Unfortunately, though oxygen is a level A recommended treatment for cluster headaches, it may not be covered by insurance - this may be due to a lack of understanding about cluster headaches by the insurance companies or a misunderstanding about the price differential between prescribing a triptan versus prescribing oxygen. The out of pocket cost for a patient buying oxygen depends on the state and can be anywhere from $500 to $10000 per year. Some patients ask about welder's oxygen, which is non-medical grade oxygen; however, this is usually not any cheaper than medical oxygen and may have impurities. Practitioners are recommended to refer insurers to the American Headache Society Guidelines during peer to peer conversations. Medicare and Medicaid have ruled that they will not cover oxygen at all for cluster headaches. Because oxygen is an excellent abortive treatment, the social worker may be involved in the care to ensure that the patient does not lack this therapy because of finances.
The neuroscience or pain nurse should educate the patient on avoidance of alcohol and the importance of getting good sleep. In addition, the patient should be told that smoking cessation is important. The nurse should also emphasize the importance of a stress-free lifestyle and avoiding hot temperatures. Finally, close follow up is essential in preventing future attacks. The members of the interprofessional team should communicate if anyone makes a change in the medication, as one of the common causes of recurrent attacks is a low dose of the prophylactic medication.armacist should review medication dosage, check for drug-drug interactions, and educate patients about side effects.
Verapamil is generally regarded as prophylactic maintenance therapy by healthcare providers despite only a level C recommendation. There is discordance between evidence-based medicine and provider activity.
Cluster headaches require an interprofessional team approach, including physicians, specialists, specialty-trained nurses, and pharmacists, all collaborating across disciplines to achieve optimal patient results. [Level V]
This research was supported (in part or whole) by HCA Healthcare and/or an HCA healthcare affiliated entity. The views expressed in this publication represent those of the author and do not necessarily represent the official views of HCA Healthcare or any of its affiliated entities.
|||Hoffmann J,May A, Diagnosis, pathophysiology, and management of cluster headache. The Lancet. Neurology. 2018 Jan; [PubMed PMID: 29174963]|
|||Weaver-Agostoni J, Cluster headache. American family physician. 2013 Jul 15; [PubMed PMID: 23939643]|
|||Láinez MJ,Guillamón E, Cluster headache and other TACs: Pathophysiology and neurostimulation options. Headache. 2017 Feb; [PubMed PMID: 28128461]|
|||Tepper DE, Cluster headache. Headache. 2015 May; [PubMed PMID: 25989208]|
|||Barloese MC, A Review of Cardiovascular Autonomic Control in Cluster Headache. Headache. 2016 Feb; [PubMed PMID: 26616143]|
|||Eising E,Pelzer N,Vijfhuizen LS,Vries B,Ferrari MD,'t Hoen PA,Terwindt GM,van den Maagdenberg AM, Identifying a gene expression signature of cluster headache in blood. Scientific reports. 2017 Jan 11; [PubMed PMID: 28074859]|
|||Wilbrink LA,Louter MA,Teernstra OPM,van Zwet EW,Huygen FJPM,Haan J,Ferrari MD,Terwindt GM, Allodynia in cluster headache. Pain. 2017 Jun; [PubMed PMID: 28267062]|
|||Robbins MS,Starling AJ,Pringsheim TM,Becker WJ,Schwedt TJ, Treatment of Cluster Headache: The American Headache Society Evidence-Based Guidelines. Headache. 2016 Jul; [PubMed PMID: 27432623]|
|||Tepper D, Oxygen for Cluster Headache. Headache. 2017 Oct; [PubMed PMID: 28967120]|
|||Lasaosa SS,Diago EB,Calzada JN,Benito AV, Cardiovascular Risk Factors in Cluster Headache. Pain medicine (Malden, Mass.). 2017 Jun 1; [PubMed PMID: 28034970]|
|||Zeeberg P,Olesen J,Jensen R, Efficacy of multidisciplinary treatment in a tertiary referral headache centre. Cephalalgia : an international journal of headache. 2005 Dec [PubMed PMID: 16305604]|