Abstract
Background
Nummular headache, or coin-shaped cephalagia, is defined as a mild to moderate, pressure-like pain that is felt exclusively in a circumscribed area. More than 200 cases of nummular headache have been reported since it was defined in 2002, but the pathogenesis remains unclear.
Methods
A patient with nummular headache who had the symptomatic area of his scalp removed but suffered headache reappearance was reported. All published cases of nummular headache in the English literature were reviewed and analyzed for demographic and clinical features, image and laboratory findings, and response to treatment.
Results
The patient with nummular headache had the symptomatic area of the scalp removed but suffered reappearance of headache in another area that overlapped with the former one. The literature review showed that nummular headache was a chronic, mild to severe, pressure-like pain with a circular or elliptical shape of 1–10 cm in diameter. The parietal region was the most affected region. Exacerbations and sensory disturbances in the affected area were reported in 43% and 56% of cases, respectively. Observational data suggested botulinum toxin type A (BoNTA) and gabapentin may be beneficial.
Discussion
Our case and evidence from the literature review support the peripheral mechanism of nummular headache. Nummular headache might be a local pain disorder stemming from terminal branches of a sensory nerve and could induce peripheral sensitization in one or several primary sensory neurons.
Introduction
Nummular headache (NH), or coin-shaped cephalagia, was first described as a chronic, mild to moderate, pressure-like pain that is felt exclusively in a circumscribed area with a diameter of 2–6 cm, in the absence of any underlying lesions of the head. Since defined by Pareja et al. in 2002 (1), more than 200 NH cases have been reported worldwide (2). Usually, NH consists of continuous pain with exacerbations or remissions, and the symptomatic area does not change in shape or size with time (3,4). A variable combination of sensory symptoms or trophic changes may be present in the affected areas (1,5–7). Accompanying symptoms of photophobia or phonophobia were reported in a few NH patients with or without coexistent headaches (8–10). Magnetic resonance imaging (MRI) and computed tomography (CT) scans have discovered some underlying lesions, such as meningioma and arachnoid cysts that posited secondary cause of NH (11,12). Neurological tests and blood examinations are normal.
Here we report a patient who had the symptomatic area of the scalp removed but suffered reappearance of NH in another area that overlapped with the former one. We reviewed all published NH cases since 2002 and analyzed their demographic and clinical features, image and laboratory findings, as well as response to treatment. Through reviewing our case and all published cases, we aimed to clearly define NH and discuss its possible pathogenesis.
Methods
This study was conducted in two parts. The first part provided a case report from our clinic, and the second part entailed a literature review.
Diagnostic criteria for nummular headache.
ICHD-II: The International Classification of Headache Disorders, second edition.
Information abstracted for each case included: (1) gender; (2) age at onset and reported; (3) duration; (4) headache location; (5) side of headache; (6) number of foci; (7) shape of affected area; (8) foci size of the affected area; (9) headache quality; (10) intensity (by verbal rating scale (VRS) and visual analogue scale (VAS)); (11) temporal pattern; (12) exacerbation; (13) sensory disturbances; (14) trophic changes; (15) accompanying symptoms; (16) concurrent headaches; (17) trauma history; (18) abnormal imaging; and (19) abnormal laboratory tests.
Cases from the literature review might have been recorded more than once since some cases were published multiple times without indication.
Results
Case report
A 63-year-old male started suffering from focal head pain after a cold shower in the summer of 1994. The painful area was perfectly circular, with the diameter of about 10 cm in the vertex. The pain was pulsating, continuous and severe, with the intensity fluctuating around 9 on a VAS of 10 (7). The patient had exacerbations lasting for minutes to hours after poor sleep or cold stimulation. There were no accompanying symptoms and no associated neurological focal symptoms. He had no known family history of migraine, stroke, dementia, or psychiatric disorders. He had had hypertension for 10 years.
In 1994, the patient was given more than 30 local nerve blocks (LNBs) with lidocaine, aiming at the greater occipital nerve, lesser occipital nerve, and the supraorbital nerve as well as the stellate ganglion, with no relief. He also received acupuncture and traditional Chinese medicine, as well as a variety of agents, either singly or in combination, including carbamazepine, somedon, diazepam, and flunarizine, with hardly any relief. At the patient’s insistence, in 2004, the scalp at the tendinous part of the epicranius muscle of the symptomatic area was removed at a local hospital and a piece of skin from his thigh was transplanted there instead (green circle, Figure 1). After surgery, the headache disappeared, but several days later, the pain reappeared in a circular area in the left temporal region with a diameter of about 5 cm (yellow circle, Figure 1). The new symptomatic area overlapped with the former one. All the other characteristics were the same as before.
Symptomatic areas before and after scalp surgery.
The patient’s neurological examination was normal, as were the blood tests, electrocardiography, and an MRI scan of the brain. Psychiatric tests were normal.
Amitriptyline at 25 mg and gabapentin at a final dose of 900 mg per day were prescribed the first time he presented to our clinic (July 21, 2011). Two weeks later, he returned with no obvious relief, and the therapy was changed to amitriptyline 25 mg with valproate sodium 600 mg per day. Two weeks later, the painful area was reduced to two-thirds of what he had before. Four months after the first visit, the symptomatic area was said to be around 1 cm in diameter, with the intensity fluctuating around 8–9 on a VAS, and the patient continued on medication. Six months since the first visit, the patient reported that the condition was slowly improving, and he was kept on amitriptyline and valproate sodium. At 17 months after the first visit (December 2012), intensity was 7–8 on a VAS and the diameter of the painful area was about 0.5 cm.
The patient is a Buddhist and never regretted the surgery. Before surgery, he made an agreement with the surgeon to be responsible for all consequences.
Literature review
Demographic and clinical characteristics of nummular headache patients in the literature.
Includes crossover locations: four temporo-parieto-occipital, two parieto-frontal, two parieto-occipital.
Includes seven middle occipital and five vertex.
VRS: verbal rating scale; 59 cases contain 28 cases transferred from visual analogue scale (VAS).
Sensory disturbances including allodynia, tender to touch, hypoesthesia, hyperalgesia, paresthesia, and dysesthesia.
Two photophobia and one photophobia and phonophobia.
Therapeutic responses of nummular headache patients in the literature.
Some antiepileptics, nonsteroidal anti-inflammatory drugs (NSAIDS), antidepressants, and analgesics are named by category (e.g. ibuprofen, NSAIDS).
BoNTA: botulinum toxin type A; TENS: transcutaneous electrical nerve stimulation.
Epidemiology and demographic information
NH was considered a rare headache, but its exact prevalence and incidence were uncertain (10). The incidence was 6.4/100,000/year in one hospital series (31). NH represented 1.25% of all headache patients in a study in a general neurology outpatient office (32,33); in one headache outpatient office in a territory hospital, the figure reached 4.6% (32). NH had a female predominance, with a gender ratio of 1.5:1. The age at onset was usually middle-aged, but could be from childhood to old age (3 to 73 years). The mean age at reporting is 47.5 years. The longest reported history of NH is 50 years. Of all NH cases, 15 reported complete relief, four without treatment.
Clinical features
NH localized in the parietal region in nearly half of cases (44%), with a few cases reporting in two contiguous regions (8%). The pain presented more often on the right side (56%). NH was strictly one focal in the majority of cases (93%). Bifocal (6%) and multifocal (0.5%) pain seemed to be rare. The affected area was typically circular (81%), which did not change with time. The diameter of the symptomatic area ranged from 1 to 10 cm, with the mean value of 3.6 cm. The quality of pain was pressing in nearly half of the cases (45%). Intensity could be mild, moderate, severe or very severe; severe (25%) and very severe (2%) pain were not rare. Pain intensity was evaluated by VAS and VRS (7). The mean VAS of 102 areas was 5.53. Among the 59 areas evaluated by VRS, 19 had mild pain (32%), 24 had moderate (41%), 15 had severe (25%), and one had very severe (2%). The temporal pattern of NH was usually continuous and chronic since onset (65%). About half of patients experienced superimposed exacerbations of pain, lasting from seconds to days (43%). Sensory disturbances including allodynia, tenderness to touch, hypoesthesia, hyperalgesia, paresthesia, and dysesthesia were reported in nearly half of the cases (56%). About 5% of NH cases had trophic changes in the symptomatic areas. Accompanying symptoms such as photophobia and phonophobia were rare (2%). About 4% of cases claimed a head trauma history. The poor correlation of NH with head trauma removed a likely possibility of posttraumatic origin of NH. Concurrent headaches existed in 13% of NH cases, 11% of which were migraine.
Image and laboratory findings
Neurological images such as MRI or CT scans are necessary since meningioma, arachnoid cyst, and cranial lesions were detected, which were thought to be the secondary cause of NH. One case with meningioma reported complete relief after surgery. Two cases with arachnoid cysts refused surgery. These were recognized as secondary forms of NH by the study authors. Since NH is a primary disorder, secondary causes might be ruled out. Laboratory findings were normal in all reported patients.
Response to treatment
No specific treatment for NH has been discovered. Antiepileptics were tried in most patients, with gabapentin proving effective in nearly half of the cases. Botulinum toxin type A (BoNTA) was prescribed in a few cases and proved to be effective in most of them (nine of 11). This treatment was supported with a grade C recommendation in 2011 (34). LNBs showed effectiveness in nearly one-fourth of patients (five of 20). It seemed that various treatments had resulted in inconsistent benefit. Since NH can be severe, more controlled studies with a larger sample of cases are needed to find a proper therapy. One NH patient who was diagnosed with meningioma got relief after surgery. It is worth noting that in all NH cases who received treatment, four were considered secondary forms (one meningioma, two arachnoid cysts, and one cranial lesion). This could result in the unsatisfactory response to treatment. Pregabalin was effective in one NH case with arachnoid cyst, while paracetamol, metamizole, and ibuprophen were ineffective. One patient with a cranial lesion was given LNBs with effect; he was also given metamizole, gabapentin, oxcarbazepine, indomethacin, and amitriptyline with no effect. One NH patient with meningioma and one with an arachnoid cyst were not given any medicine. In some case reports, the names of the medicines were not precisely interpreted (e.g. ibuprofen as a nonsteroidal anti-inflammatory drug (NSAID)), so we present only the available data.
Discussion
Our literature review showed that NH was a chronic, mild to severe, pressure-like pain with a circular or elliptical shape of 1–10 cm in diameter. The parietal region was most often affected. Exacerbations and sensory disturbances in the affected area were reported in half of cases. Concurrent headaches existed in 13% of NH cases, 11% of which were migraine. The prevalence of migraine was reported as about 9.6%–14.6% (35–37), which implies that NH is an entity that might not be caused by other types of headache such as migraine. Observational data suggested that BoNTA and gabapentin might be beneficial treatments. Since severe NH cases were not rare, a proper therapy was of primary importance.
Pathogenesis
Evidence that supports the peripheral or central mechanism.
Hypoesthesia, dysesthesia, allodynia, and paresthesia; PPT: pressure pain threshold (17); BoNTA: botulinum toxin type A.
Some evidence supports the central origin of NH. Subcutaneous injections of anesthetics were unsatisfactory in the majority of patients (1,4,5,11,16,17,28). For some NH patients, the painful area extends across the midline of the scalp; this cannot be easily explained by peripheral nerve innervation (4,29,30,39). Some patients are reported to have bifocal (9,20) or multifocal (5) headache, which also supports a central mechanism, since a peripheral disorder would be likely to generate symptoms in a more restricted area.
Some NH cases are described as “symptomatic NH.” This implies symptoms that are secondary to underlying structural lesions or other causes, such as meningioma (11), arachnoid cyst (12), fusiform aneurysms (40), and after local neurosurgery and a benign cranial lesion (21). These observations posed the secondary form of NH. Since NH is a primary disorder, the secondary causes are worth attention but might be excluded when making a diagnosis.
Except for the hypotheses of pathogenesis above, NH patients are reported with psychiatric disturbances and NH is suggested to have a psychiatric etiology (30). Fernández-de-las-Peñas et al. tested some patients and indicated that depression and anxiety were not related to the presence of NH (19). Some patients are wakened by the pain from sleep (31). Now, increasingly subjective symptom reports and objective measurements make this indication seem very unlikely (17).
Our case report fulfilled the proposed diagnostic criteria for NH (7). The case supported the peripheral origin of NH: When the scalp of the symptomatic area was removed, the headache ceased. The surgery removed the tissue from the scarfskin to the tendinous part of the epicranius muscle, including the algesireceptor, and resulted in relief from localized headache in the very place. Several days later, NH reappeared in another place that partly overlapped with the former pain area. The two symptomatic areas were both located in the innervation area of the terminal braches of the supraorbital nerve. Supposing that NH originated completely from the central mechanism, surgery would be of no use and the headache would remain.
Based on our case and evidence from our literature review, we propose the peripheral mechanism, with NH as a local pain disorder stemming from terminal branches of the sensory nerve. And it might induce peripheral sensitization in one or several primary sensory neurons. This hypothesis may explain the special topography of “nummular” as well as the reappearance of NH in our case. After removal of the scalp, sensory nerves in the former area were destroyed and NH disappeared, but sensitization in primary sensory neurons persisted and caused reappearance of NH in the innervation area of other branches of the nerves. Since sensory nerves from primary sensory neurons give out terminal branches that innervate overlapping but slightly different fields, the reappearance of NH overlapped with the former area (41) (Figure 2). As the effect of LNBs aimed at peripheral nerves was unsatisfactory for our case as well as for the majority of cases in the literature, we hypothesize that sensitization might occur at the level of primary sensory neurons. Also, there exists the possibility that the surgery might directly sensitize nearby sensory nerves because of injury, inducing reappearance of NH; this also supports the peripheral mechanism of NH (42).
Peripheral sensitization of sensory neurons. Peripheral mechanism of nummular headache. Former nummular headache area is in light blue, supplied by the peripheral nerve. Nummular headache disappears when the focal scalp is removed, but sensitization of primary sensory neurons persists, causing reappearance of nummular headache in the innervation area (dark blue) of other branches.
The literature review showed that the parietal region was the most commonly affected. This area is innervated by the distal branches of the supraorbital and occipital nerves, which is in accordance with the peripheral mechanism. BoNTA turned out to be the most effective treatment and it is believed that BoNTA is able to reduce peripheral sensitization induced by local neurotransmitter release (24,43). LNBs aimed at peripheral nerves might be unsatisfactory because the headache was related to the persistence of sensitization of sensory neurons. Therefore, LNB aimed at sensory neurons might be more useful for NH. Even around the midline of the head, the innervation of sensory nerves partly overlaps; this might be why NH cases have a painful area extended across the midline.
Meanwhile, we cannot definitely exclude the central hypothesis, since peripheral nociception might also initiate or maintain a process of central sensitization (44). NH might also be caused by a complex pathogenesis involving both peripheral and central mechanisms. This question still remains open. Additional clinical observations and basic research may help us unveil the mystery.
Clinical implications
We presented a fairly special case of nummular headache (NH) in a patient who had the symptomatic area of the scalp cut off and suffered reappearance of NH in another place that overlapped with the former one. We reviewed all reported NH cases since it was defined in 2002, and analyzed its demographic and clinical features, image and laboratory findings, and response to treatment. Since there is no literature review for the reported 238 cases, we consider our work helpful for the medical practitioners and researchers. We posited a hypothesis of peripheral sensitization of the pathogenesis of NH. The pathogenesis of NH has thus far been unclear.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
None declared.