Neuro-Ophthalmology & Neuro-Otology
Aug. 17, 2022
At vero eos et accusamus et iusto odio dignissimos ducimus qui blanditiis praesentium voluptatum deleniti atque corrupti quos dolores et quas.
The author reviews paroxysmal hemicrania, one of the trigeminal autonomic cephalalgias that is indomethacin-responsive (123). Strictly unilateral pain of short attack length (typically 20 minutes), high frequency of attacks (often 10 per day), and associated lateralized cranial autonomic features (such as lacrimation, conjunctival injection, and nasal symptoms) should trigger the diagnosis to be considered. When paroxysmal hemicrania is considered, investigation for pituitary gland pathology with dedicated MRI and blood tests is recommended, based on large case series. Paroxysmal hemicrania is absolutely responsive to indomethacin, when it is tolerated, at doses from 25 to 275 mg daily.
• Paroxysmal hemicrania is a unilateral, severe, short-lasting headache, typically of 20 minutes duration and occurring approximately 10 times a day.
• Paroxysmal hemicrania attacks are associated with cranial autonomic symptoms, such as lacrimation, conjunctival infection, and nasal symptoms.
• Paroxysmal hemicrania responds absolutely to indomethacin given orally.
• It is recommended that any patient with a diagnosis of paroxysmal hemicrania undergo investigation by MRI and blood tests for pituitary gland dysfunction.
In 1974, Sjaastad and Dale reported what they described rather aptly as a new treatable headache entity (105). They subsequently coined the term “chronic paroxysmal hemicrania” to describe these patients (106). Later, a remitting form of the disease was recognized and termed “episodic hemicrania continua” (54; 10; 77; 40; 120). The release of the second edition of the International Headache Society Classification resulted in the introduction of the umbrella terminology “paroxysmal hemicrania,” which is recognized to have both an episodic and a chronic form (46).
Paroxysmal hemicrania is characterized by multiple, brief, intense, daily focal attacks of head pain. The pain is unilateral and always affects the same side. The pain is usually most severe in the auriculotemporal area, the forehead, and above or behind the ear, although it can occur anywhere in the head. It may spread to involve the ipsilateral shoulder, arm, and neck. The pain is described as excruciating, throbbing, boring, or pulsating. Between attacks, the patient may have tenderness in the symptomatic area (21). During attacks, the patient usually sits quietly or may "curl up in bed," which contrasts with the usually increased restlessness and activity during attacks of cluster headache (38). Some patients have interparoxysmal pain, so the differential diagnosis with hemicrania continua can be problematic as both are indomethacin-sensitive (20). This may be the basis for a reported transition from hemicrania continua to paroxysmal hemicrania in a single case report (75).
Paroxysmal hemicrania usually begins in adulthood. The mean age is 34 years, with a range of 2 to 81 years. Paroxysmal hemicrania has been reported in children (52; 15; 36; 71; 114; 121; 116; 89; 56; 49; 33), and their symptoms are very much the same as those in adults (09). As in adults, the differential diagnosis with cluster headache can be troublesome (109). The author has seen a number of children between 4 and 12 years of age with typical indomethacin responses. In the author’s experience, the disorder settles in this age group after 2 to 3 years, and one patient has been reported with that behavior (115). The attack profile of paroxysmal hemicrania is highly characteristic. The frequency of attacks ranges from 2 to 30 per 24-hour period (102). Attacks usually last between 2 and 25 minutes and occasionally as long as 60 minutes. In a prospective study of 105 attacks, the mean duration was found to be 13 minutes, with a range of 3 to 46 minutes and a mean attack frequency of 14 in a 24-hour period (range 4 to 38) and even circadian distribution (91). Nocturnal attacks associated with the REM phase of sleep have also been described (51).
Ipsilateral lacrimation is reported to occur in about 80% of attacks. When the lacrimation is bilateral, it is predominantly on the side of the pain (04). Conjunctival injection on the symptomatic side occurs during two thirds of attacks (21). Some patients have ipsilateral eyelid edema and slight miosis. Photophobia may accompany some attacks, and increased forehead sweating, especially on the symptomatic side, is observed in a few patients. Photophobia, when present, is often unilateral to the side of the pain (48). Although not widely recognized, a sense of aural fullness is reported in paroxysmal hemicrania (21) and may account for some patients reported not to have cranial autonomic features (86).
Three cases are documented of nonlateralized, short-lasting headache that responded to indomethacin, but these patients did not have cranial autonomic symptoms (85; 74; 08); one might consider this a different syndrome--perhaps paroxysmal cephalalgia (67). Cases of short-lasting presentation with bilateral cranial autonomic features responding to indomethacin have also been reported (31; 27), as has a case with unilateral pain without apparent cranial autonomic features (58). In the latter case, the description of the relevant negatives is incomplete, and the nosology of these entities is unclear.
Although the majority of attacks are sudden and spontaneous, approximately 10% may be precipitated mechanically, either by bending or by rotating the head (04). Attacks may also be provoked by external pressure against the transverse processes of C2, C3, C4, or the greater occipital nerve (107). The relationship between menstruation and paroxysmal hemicrania attacks is undetermined (60; 59). Birth control pills do not seem to influence the attack frequency, and menopause has no reported effect. Unusual presentation with otalgia and a feeling of fullness in the ear has been reported (14), as has extratrigeminal pain (29). A report from the Mayo Clinic suggests that the largest part of the phenotype can be seen without a response to indomethacin (12). Cases of mixed trigeminal autonomic cephalalgias have been reported and underline the premise of ICHD-3 that there is shared pathophysiology (117).
The features of the International Classification of Headache Disorders (ICHD) third edition beta criteria are listed below (46).
Paroxysmal hemicrania has two key forms. The episodic form occurs during a period that lasts 7 days to 1 year and is separated by pain-free periods that last 1 month or more. The chronic form has attacks that occur for more than 1 year without remission or with remissions lasting less than 1 month.
To be diagnosed with paroxysmal hemicrania, patients must experience each of the following:
A. At least 20 attacks fulfilling criteria B through E
B. Severe, unilateral orbital, supraorbital, or temporal pain that lasts 2 to 30 minutes
C. Headache associated with at least one of the following symptoms or signs, ipsilateral to the pain:
- Conjunctival injection and/or lacrimation
D. Attack frequency of more than five headaches daily for more than one half of the time
E. Attacks are prevented absolutely by therapeutic doses of indomethacin
F. Not better accounted for by another ICHD-3 diagnosis
The new criteria are very much like ICHD-2 (45), with the notable standardization of the cranial autonomic features amongst the Trigeminal Autonomic Cephalalgias. The sense of aural fullness, which is common in trigeminal autonomic cephalalgias (38), has been removed in ICHD-3, which this author sees as a mistake to be corrected in ICHD-4.
There are no systematic data on the prognosis of paroxysmal hemicrania. Sjaastad followed a patient for 45 years, during which time the patient was on indomethacin for 33 years. The drug never stopped working, and the problem seemed to become less troublesome with time (100).
No secondary complications exist except those that arise as a result of long-term indomethacin therapy, such as gastritis, gastric ulcers, and gastrointestinal bleeding.
The etiology of paroxysmal hemicrania is unknown; however, given its similarities with cluster headache (69), a disorder of the CNS seems the most likely explanation. Head or neck trauma is reported by approximately 20% of patients (04), although these findings do not differ significantly from cluster headache or migraine (53; 99).
The pathophysiological mechanisms responsible for pain in paroxysmal hemicrania are unknown. As with other primary headaches, it seems likely that there is a hereditary element, and a family with paroxysmal hemicrania has been reported (24). The pain is strictly unilateral and appears to be in the distribution of the ipsilateral trigeminal nerve. The pain and the cranial autonomic symptoms, such as lacrimation and conjunctival injection, have linked the disorder to cluster headache under the classification of a trigeminal autonomic cephalalgia (40). It is likely to be a brain disorder, as is cluster headache (37).
Dynamic tonometric studies have shown that in chronic paroxysmal hemicrania significant attack-related increases in corneal indentation pulse amplitudes, ocular blood flow, and intraocular pressure occur in paroxysmal hemicrania (106; 92). These findings probably result from acute vasodilatation resulting in increased intraocular volume due to a neurogenic impulse and are mediated by the release of vasoactive neuropeptides (39). Similar to cluster headache, calcitonin gene-related peptide and vasoactive intestinal polypeptide are released in the cranial circulation during acute attacks of chronic paroxysmal hemicrania (39; 30). POEMS can have a paroxysmal hemicrania-like presentation (82), again implicating peptide mechanisms.
Increased tearing, conjunctival injection, and nasal secretion are probably due to cranial parasympathetic activation. Attack-related heart rhythm disturbances have been observed in a few patients. These cardiac abnormalities include bradycardia and sinoatrial block, bundle-branch block with episodes of atrial fibrillation, and multiple extrasystoles (105; 84; 91). This may indicate dysfunction in the central control of the autonomic nervous system during chronic paroxysmal hemicrania attacks that affects both sympathetic and parasympathetic systems. A lesion in the pericarotid area inside the cavernous sinus where trigeminal, sympathetic, and parasympathetic fibers come together has been advanced to explain the pain and cranial autonomic changes (73); however, this fails to explain the paroxysmal nature of the pain attacks (40). There is no cognitive deficit in paroxysmal hemicrania (32).
An important finding for paroxysmal hemicrania is activation on PET in the posterior hypothalamic region (64), as seen in other trigeminal autonomic cephalalgias (22), and in the ventral midbrain, as seen in hemicrania continua (65). A case of a patient in the arctic region with less troublesome attacks in the light season reinforces the possibility of hypothalamic mechanisms in this disorder (07). The latter may be a link to indomethacin sensitivity, and the whole picture reinforces the unique nature of this condition.
The mechanism behind the absolute responsiveness to indomethacin is now starting to be unraveled. There is a clear effect on nitric oxide synthase of indomethacin that other NSAIDS, such as ibuprofen and naproxen, do not have (111; 112). It appears to be independent of indomethacin’s effect on prostaglandin synthesis, as other nonsteroidal anti-inflammatory agents have little effect (101) or no (80; 02) effect on chronic paroxysmal hemicrania. Sjaastad’s first case (personal communication) and cases this author has seen respond well to aspirin.
Paroxysmal hemicrania is a rare syndrome, although with increasing awareness of it, it is perhaps being recognized more frequently. The prevalence of paroxysmal hemicrania is not known, although an important contribution from the Vaga study of headache epidemiology was to identify one possible patient with the syndrome among 1838 adults aged 18 to 65 who were quizzed (104). The data are important because the authors also reported data for cluster headache, 0.3% of the cohort (103), and saw two patients with SUNCT and one with hemicrania continua in the same group (104). Remarkably, this suggests that SUNCT is more common than paroxysmal hemicrania and that paroxysmal hemicrania is much less common than cluster headache. This is certainly consistent with this author’s referral practice.
Chronic paroxysmal hemicrania was originally reported in females, although clearly males can be affected (88; 90). In a 1989 survey of 84 reported cases, 59 were females and 25 males, a female-to-male ratio of 2.36 to 1 (04). In the largest single case series, the male-to-female ratio was equal (21).
No method of preventing paroxysmal hemicranias has been established. Once the diagnosis is made, indomethacin will prevent the attacks; but when indomethacin is discontinued, symptoms usually reappear within 12 hours to a few days (02).
The differential diagnosis for paroxysmal hemicrania is between other trigeminal autonomic cephalalgias (40; 97) and secondary forms of chronic paroxysmal hemicrania. Mistaking chronic paroxysmal hemicrania for cluster headache is problematic because, generally speaking, treatments for cluster headache are not effective in chronic paroxysmal hemicrania. There is a case report of the two conditions existing together (18), and there have been claims that cluster headache can respond to indomethacin (87). These cases need to be objectively determined with blinded indomethacin tests. Chronic paroxysmal hemicrania has been said to be more often seen in females, although the largest case series does not bear this out (21). Given the effect of indomethacin is relatively rapid in onset (02), it could be advocated that all patients diagnosed with trigeminal autonomic cephalgias who do not have a contraindication could have a short course of indomethacin at the start of treatment to detect the indomethacin-sensitive group, at least until a reliable biological marker becomes available. Whether indomethacin-insensitive paroxysmal hemicrania exists is moot (12; 34), and more research is needed. A comparison found no particular clinical difference between paroxysmal hemicrania and cluster headache (125), and the editorial commentary (11) is reminiscent of Sjaastad’s original remark that indomethacin “hits the nail on the head” in paroxysmal hemicrania. The only trial of indomethacin in cluster headache suggests it does not work (01), but published placebo-controlled experience is lacking. Episodic paroxysmal hemicrania is differentiated by periods in which there are no attacks, and the considerations for indomethacin are similar.
The coexistence of chronic paroxysmal hemicrania with trigeminal neuralgia, or chronic paroxysmal hemicrania-tic syndrome, is recognized (42). It is important to recognize because both the chronic paroxysmal hemicrania and the trigeminal neuralgia require treatment (17; 43; 62; 126). Acetazolamide has been suggested as an addition to this therapeutic armamentarium (94). In line with the proposed changes to the International Headache Society classification and respecting the fact that both episodic and chronic paroxysmal hemicrania are reported (13), the term “paroxysmal hemicrania-tic” is preferable (41). Chronic paroxysmal hemicrania may coexist with benign cough headache, and both syndromes respond to indomethacin (63).
Secondary paroxysmal hemicrania is well described, and investigations are required to identify or exclude treatable underlying causes (118). Secondary paroxysmal hemicrania should be considered if the clinical picture is typical but treatment response is poor or if there are neurologic signs. Paroxysmal hemicrania may be seen after head trauma and even with typical migraine aura where comorbidity with migraine is a reasonable explanation (66; 96; 83). A reasonably complete screen of a patient with chronic paroxysmal hemicrania, considering the associated clinical problems reported, would include a blood count looking for thrombocythemia (57); erythrocyte sedimentation rate and vasculitic investigations (70); a brain-imaging procedure looking for an intracranial tumor, such as a lesion in the region of the sella turcica (35; 122; 95) or orbit (19); cerebral metastases (61) or elsewhere (70); or a brain hemorrhage (06). It is important to exclude pituitary pathology by imaging and blood tests (55; 124; 05). Other structural mimics of chronic paroxysmal hemicrania include an arteriovenous malformation (78), anterior clinoid meningioma (25), cavernous sinus meningioma, or the condition has been reported after carotid aneurysm embolization (47). A secondary form of the disorder has been reported with the use of phosphodiesterase inhibitors (113). Secondary chronic paroxysmal hemicrania is more likely if the patient requires high doses (greater than 200 mg/day) of indomethacin (108), although this author has seen a case in which a patient took 400 mg daily, and extensive investigations over the years have yielded nothing. Should the pain become bilateral, a lumbar puncture should be carried out to look for intracranial hypertension, even in the face of a response to indomethacin (42). When appropriate, an electrocardiogram and Holter monitor should be considered to look for bundle branch block or atrial fibrillation (91), and a chest x-ray should be considered to look for a Pancoast tumor (28).
A good clinical history, a detailed neurologic examination, and a therapeutic trial of indomethacin are necessary to make a diagnosis of paroxysmal hemicrania. It is generally agreed that an MRI brain with pituitary views is appropriate at baseline.
The treatment of chronic paroxysmal hemicrania is prophylactic (38). Indomethacin is the drug of choice, and its effect on symptoms is absolute. When a patient with a unilateral headache has a relatively high attack frequency (more than four attacks in 24 hours), a drug trial with indomethacin should be considered. Trial doses should be increased to at least 150 mg per 24 hours for 3 or 4 days, with a maximum dose of 275 mg daily. The beneficial effect is seen within a few hours to 5 days. The maintenance dosage is usually 25 to 100 mg/day but varies among individuals, and the response persists for years (79). When indomethacin is discontinued, symptoms usually reappear within 12 hours to a few days (02). To prevent gastric complications, long-term indomethacin therapy may be combined with antacids and histamine blockers or other gastric mucosa protective agents. Nausea, vomiting, vertigo, and purpura have been reported during indomethacin use. There has been some controversy as to whether sumatriptan is effective (81) or ineffective (26; 03) in chronic paroxysmal hemicrania. The issue is unresolved, but it seems possible that it is a matter of attack length. The COX-2 inhibitors celecoxib (68), valdecoxib (withdrawn from the United States and other markets due to adverse effects) (98), and rofecoxib (withdrawn from United States and other markets due to adverse effects) (Goadsby unpublished observations) may be useful in paroxysmal hemicrania. This author and others have had useful responses from topiramate (23; 16). As with cluster headache, pterygopalatine ganglion blockade has been reported to be useful (72). Neuromodulation with noninvasive vagus nerve stimulation has been reported to be useful (119; 50).
Patients who respond to indomethacin will continue to do so indefinitely.
In several patients with chronic paroxysmal hemicrania, attacks have disappeared or improved during pregnancy and returned after delivery (102; 110), although this author has managed a female with otherwise typical disease who had attacks throughout pregnancy. There is no risk to the mother or child if chronic paroxysmal hemicrania occurs during pregnancy. It is not advisable to continue indomethacin. No specific treatment is available during pregnancy for this condition, and its management is extremely difficult.
There is no contraindication to the use of general or local anesthetics in patients undergoing surgical procedures.
Peter J Goadsby MD PhD
Dr. Goadsby of King’s College London and the University of California, San Francisco, received consulting fees from Abbvie, Aeon Biopharma, Amgen, Biohaven Pharmaceuticals Inc, Electrocore LLC, Eli Lilly, eNeura, Epalex, GlaxoSmithKline, Impel Neuropharma, Lundbeck, Novartis, Pfizer, Prxis, Sanofi, Santara Therapeutics, Satsuma, and Teva Neurosciences and grants from Amgen, Celegene, and Eli Lilly.See Profile
Shuu-Jiun Wang MD
Dr. Wang of the Brain Research Center, National Yang-Ming University, and the Neurological Institute, Taipei Veterans General Hospital, has no relevant financial relationships to disclose.See Profile
Nearly 3,000 illustrations, including video clips of neurologic disorders.
Every article is reviewed by our esteemed Editorial Board for accuracy and currency.
Full spectrum of neurology in 1,200 comprehensive articles.
Neuro-Ophthalmology & Neuro-Otology
Aug. 17, 2022
Headache & Pain
Jul. 13, 2022
Headache & Pain
Jul. 11, 2022
Headache & Pain
Jul. 01, 2022
Jun. 11, 2022
Jun. 06, 2022
Jun. 04, 2022
Jun. 04, 2022