Abstract
Background
Epicrania fugax is included in the appendix of the International Classification of Headache Disorders and is characterized as recurrent brief attacks of linear or zigzag pain moving across the cranial surface, commencing and terminating in the distribution of different nerves. We present a new case of epicrania fugax in which the headache was the presenting symptom of a cerebellar abscess.
Case report
We present a 58-year-old woman with prior history of Chiari I malformation who underwent suboccipital craniectomy. Two weeks after surgery, she experienced paroxysmal pain episodes of 1–3 seconds, with constant linear trajectory from the right occipital surface to the right orbital region, remaining pain free between episodes. Cranial tomography showed a hypodense intraaxial lesion in the right cerebellar hemisphere. Magnetic Resonance Imaging exhibited intralesional bleeding and peripheral enhancement after gadolinium administration. Post-surgical cerebellar abscess was diagnosed and antibiotic therapy was started; the patient underwent urgent surgical drainage. Pain disappeared after the surgery and the patient remains pain free with 12 months of follow-up.
Conclusion
Posterior fossa abnormalities have been described as a possible cause of secondary epicrania fugax. The presence of red flags should encourage conducting of paraclinical tests to rule out a symptomatic form.
Introduction
Epicrania fugax (EF) is an entity originally described by Pareja et al. in 2008 (1). It was initially included in the appendix of the beta version of the third edition of International Classification of Headache Disorders (ICHD) (2) and it is characterized by the presence of recurrent pain attacks lasting 1–10 seconds, in which pain is described as moving with a linear or zigzag trajectory across the cranial surface, beginning and ending in the distributions of different nerves (3).
Since the first description, phenotypic variants have been described, such as patients with antero-posterior pain radiation (4), facial onset (5), multidirectional distribution (6), or interictal sensory disturbance (7).
Given the superficial topography of pain, usually it is considered an epicranial headache. However, two symptomatic cases have been recently described in relation with posterior fossa abnormalities (8,9). We present here a new case of EF in which the headache was the presenting symptom of a cerebellar abscess.
Case report
We present a 58-year-old woman with a history of episodic headache occurring 6–12 days per month, of 1–4 hours duration, described as holocranial, with a tightening quality, usually 6 out of 10 intensity in Verbal Analogic Scale, without hypersensitivity to external stimuli, or nausea or vomiting. The headache appeared spontaneously but coughing and straining aggravated it. She experienced an increase in frequency up to daily episodes and, because of the aggravation by coughing, a magnetic resonance study was requested. This study revealed 5 mm descent of cerebellar tonsils and she received a diagnosis of symptomatic type I Chiari malformation, being referred directly to neurosurgery without any specific evaluation by neurology. Neurosurgery considered the radiological findings as symptomatic and the patient underwent suboccipital craniectomy.
Six weeks after the surgery, she was referred to our headache office for the first time because of a new headache, which started two weeks after the surgery. It was described as paroxysmal pain lasting 1–3 seconds, occurring two to five times per day, with electric quality and 10 out of 10 intensity in Verbal Analogical Scale (0: No pain, 10: The worst imaginable pain). All the episodes started in the same precise point, located at the right occipital scalp about 3 cm above the surgical scar and then moved forward with a linear trajectory towards the right supraorbital region. Between episodes, the patient remained pain free. She did not identify triggers and pain was not associated to nausea, vomiting, hypersensitivity to stimuli or trigeminal-autonomic symptoms. The patient had not presented fever, malaise, or nuchal rigidity since surgery.
The physical examination did not show hypersensitivity to palpation of pericranial nerves, hypoesthesia, hyperalgesia or allodynia in the pain territory. Inspection and palpation of the territory did not reveal anything abnormal. Temporal artery pulses were normal, and we did not find active myofascial trigger points. The neurological examination was strictly normal.
In light of the presentation of these symptoms in close temporal relationship with the surgical intervention, an emergent neuroimaging study was requested. A computed tomography showed an intraaxial hypodense lesion without well-defined borders located in the right cerebellar hemisphere. Brain magnetic resonance imaging (MRI) was performed urgently, which showed the same intraaxial hemispheric cerebellar lesion with intralesional bleeding foci and peripheral enhancement after gadolinium administration (Figures 1 and 2). Because of the suspicion of a postsurgical cerebellar abscess, antibiotic therapy with meropenem 1 gr/8 h, vancomycin 2 gr/12 h and ceftriaxone 2 gr/24 h was started. The patient underwent urgent surgical drainage and lesion evacuation. None of the five samples of exudate obtained during the surgery showed any bacterial growth in cultures. After the surgery, she still complained about weekly episodes of holocranial headache lasting from 1 to 2 hours, pressing quality, without vegetative symptoms, photophobia, phonophobia or osmophobia and not modified by coughing or any physical activity. Intensity is mild enough that she does not need any symptomatic medication. EF episodes disappeared after surgery and, after 12 months of follow up, the patient never presented them again. Imaging during follow up after 1 month and 6 months showed resolution of the intracranial lesion and postsurgical changes.
Pre- and post-gadolinium axial T1 FSE weighted images showing hypointense intra axial lesion located in right cerebellar hemisphere, showing ring-enhancement after gadolinium administration. Sagittal CUBE FLAIR pre-gadolinium and sagittal T1 FSE weighted post-gadolinium showing the ring-enhancing lesion located in the right cerebellar hemisphere.

Discussion
Some authors subdivide headaches into epicranias and splachnocranias. The first group include entities in which the origin of pain is considered to be in the superficial extracranial structures, such as scalp, subcutaneous tissue and small fiber nerves. A pain described in the cranial surface in a well-delimitated area characterizes them. Nummular headache, terminal branch neuralgias or primary stabbing headache are considered part of this group by some authors (1,10).
EF was considered to be part of this group since its first description because of the superficial topography of pain and the precise delimitation of it. Additionally, this idea was supported after the description of cases in close relation with other epicranial headaches, such as terminal branch neuralgias and nummular headache (11,12).
Recently, the two first cases of secondary EF have been reported. Curiously, both cases occurred in patients with posterior fossa disorders. The first one, described by Fernandez-Matarrubia et al. (8) had undergone stereotactic radiotherapy to treat a meningioma of the skull base in contact with the trigeminal nerve. A paroxysmal pain like that of EF appeared one month after the surgery, with frontal onset ipsilateral to the lesion. Jaimes et al. (9) reported the second case, as a secondary form in the context of an infarction in the right dorsolateral medulla secondary to a right vertebral artery occlusion.
In the previously reported cases the latency was 1 and 9 months, respectively. In our case, episodes started two weeks after surgery.
In the characterization of a secondary headache, ICHD requests either, or both, close temporal relation between the onset of the headache and the event and/or significant improvement in parallel with the stabilization or improvement of the noxious condition (3).
All three reported cases appeared to be related to ipsilateral posterior fossa conditions. To date, pathogenesis of EF remains unknown. Nociceptors located in the terminal fibers of trigeminal nerve and the first occipital nerves can be activated by many different intra- or extracranial stimuli, leading to trigeminal cervical complex activation. The activation of the trigeminal nucleus caudalis has been described not only in secondary entities but also in many primary headache disorders (1,13). The convergence of trigeminal and first cervical sensory afferents in the brainstem could be the explanation of the referred pain at the cranial or facial territory when the lesion is located in the occipital nerves territory (14). For instance, in strokes affecting the brainstem or cerebellum, pain is commonly described in the periocular and frontal territory (16–18). Description of superficial pain caused by intracranial disorders can be also explained by the intra-extracranial nerve connections of cranial nerves (19).
We hypothesize that secondary forms of EF (8,9) could be a possible form of referred pain due to convergence of sensory branches. It is yet to be determined if “primary” EF forms originate from intracranial structures or peripherally. In our opinion, given that EF could be related with structural lesions, imaging should be routinely considered in all patients (3) with particular attention to posterior fossa structures. Another lesson from this case is that before a surgical treatment of a headache it is always desirable to have an expert evaluation, given that in this case the initial headache did not fulfil IHCD criteria for headache attributed to Chiari malformation type I and other possible causes were not considered.
Abbreviations
EF: epicrania fugax; ICHD: International Classification of Headache Disorders; MRI: magnetic resonance imaging; TNC: trigeminal nucleus caudalis.
Clinical implications
This is the first case in which epicrania fugax was the presenting symptom of an intracranial condition. Epicrania fugax is associated with posterior fossa conditions. A secondary form of epicrania fugax should be ruled out.
Footnotes
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The patient read and signed an informed consent form and agreed with the publication.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
