Abstract
It has recently been suggested that the trigeminocervical complex plays a crucial role in the pathophysiology of neck discomfort that accompanies migraine attacks. Clinical and neurophysiological data have shown that pain within the occipital area may be transmitted by the first trigeminal branch, which supports an anatomical and functional link between cervical and trigeminal modulation of peripheral afferents. We describe a patient with an acute symptomatic migraine attack and chronic occipital neuralgia, both due to bleeding of a bulbocervical cavernoma. The clinical presentation is also discussed and related to recent scientific data on the role of the trigeminocervical complex in both the clinical picture and underlying pathophysiological mechanisms of cervical and head pain.
Introduction
In clinical practice, frontotemporal headache and neck discomfort are often associated during a migraine attack. Neck discomfort is often characterized by both muscle tenderness and cervical pain and may, at least for a subgroup of migraineurs, be more disabling than headache itself. In some migraine attacks, the occipital location of pain may represent the sole manifestation, thereby making the diagnosis of primary headache more difficult. Patients with headaches of cervical origin [e.g. cervicogenic headache, occipital neuralgia (ON)] may even have pain that spreads to trigeminal territories.
These clinical observations point to the existence of a close anatomical and functional relationship between the trigeminal nuclei and upper spinal cord. Indeed, earlier neurosurgical findings have shown that stimulation of structures in the neck, which are innervated by the upper cervical roots, not only elicits occipital pain sensations, but may even be perceived in trigeminally innervated dermatomes (1). Similarly, direct mechanical stimulation of the supratentorial dura mater and large cranial vessels leads especially to pain in the first (ophthalmic) division of the trigeminal nerve (2), but which may even be perceived in dermatomes supplied by the upper cervical roots (3).
One mechanism that may explain these clinical observations is that the afferent inflow from the meninges and from the upper cervical roots converges onto the same central neuron. Indeed, functional coupling between nociceptive meningeal afferents and cervical afferents at the second-order neuron level has been shown to exist (4). In this regard, studies in the rat (5), cat (6) and monkey (7) have demonstrated that the trigeminal nucleus extends beyond the traditional nucleus caudalis to the dorsal horn of the high cervical region (C2–C3 segments) in an anatomofunctional continuum named the trigeminocervical complex (TCC).
We would like to add to the above-mentioned body of evidence by describing a case of a migraine-like headache associated with central sensitization phenomena followed by chronic ON in a patient with a bleeding bulbocervical cavernoma.
Case report
A 32-year-old White, otherwise healthy, man came to our Emergency Department (ED) complaining of a daily, refractory headache that had started 2 weeks earlier. He had been treating the headache with daily non-steroidal anti-inflammatory drugs (NSAIDs), which provided good, although temporary, relief; he had decided to come to the ED because of the persistent nature of the headache.
Upon admission, he underwent neurological examination and cerebral computed tomography (CT), both of which were normal. After parenteral administration of cyclooxygenase (COX)-1/COX-2 inhibitors led to the remission of pain, he was discharged and referred to our headache centre with the ED physician's diagnosis of primary headache. His admission to our headache centre was 4 days after being discharged, headache free, from the ED.
A more comprehensive clinical visit revealed that the patient had had a similar headache attack 2 years earlier. On that occasion, the symptoms had resolved spontaneously after 1 week without medical assistance. There was no family history of headache. The headache that had led to admission to the ED resembled clinically a migraine attack, insofar as it was throbbing, unilateral on the right side, frontotemporal and of moderate/severe intensity. Moreover, besides reporting that the headache was seriously exacerbated by physical activity, the patient also complained of a stiff neck upon awaking in the morning. There were neither facial autonomic features nor any other symptoms. At the time of the first visit to the headache centre, the patient reported that he had had right facial numbness and dysaesthesia for the previous 12 days and hiccup for the previous 7 days. He also referred to the persistence of neck stiffness when he awoke, although no evidence of this sign was found on neurological examination. His cranial nerves were intact and his limb examination, co-ordination and gait normal. However, neurological examination revealed hyperalgesia to brush, touch and pinprick stimuli and cutaneous tactile allodynia along the right great occipital nerve, in the right auricular and retroauricular regions, as well as along the three branches of the right trigeminal nerve.
Given the peculiarity of the history and clinical presentation, magnetic resonance imaging (MRI) was performed, with particular attention being paid to the midbrain and upper cervical medulla regions.
Standard MRI 1.5 T (Philips Intera) was used to obtain 3 mm thick slices. A comprehensive MRI study of the brain and cervical region was performed by including a T2-weighted sequence, a T2 fast field echo (FFE) sequence and precontrast and postcontrast T1-weighted sequences(16 ml of Magnevist, 0.2 mmol/kg). MRI showed a clearly hypointense lesion on the T2 FFE images, with an inhomogeneous, hyperintense signal on T2, located in the right posterior portion of the bulbocervical tract with a caudal extension up to C2. No pathological enhancement was observed after intravenous contrast administration.
These MRI findings are compatible with a diagnosis of cavernoma, accompanied by signs of recent bleeding associated with blood degradation products in various catabolic phases (Figs 1–4).
Weighted fast field echo magnetic resonance imaging axial scan. A clearly hypointense area in the bulbocervical junction due to haemosiderin deposits.
Axial scan. Presence of an inhomogeneous area at the level of the bulbocervical junction ascribable to haemorrhagic events and the presence of methaemoglobin.
T1-weighted magnetic resonance imaging saggital scan after administration of gadolinium with diethylenetriamine penta-acetic acid (Gd-DTPA). The inhomogeneous area of altered signal is not enhanced by Gd-DTPA.
T2-weighted magnetic resonance imaging coronal scan. Presence of inhomogeneous area of altered signal located in the posterior right zone of the bulbocervical junction characterized by a central hyperintense component surrounded by a hypointense rim (haemosiderin), which indicate blood products in different catabolic phases.
During the second visit to the headache centre, performed 2 days after the first (i.e. 20 days after the onset of headache), we reassessed the patient's pathology in light of the MRI findings. In view of these findings, a neurosurgical consultation was requested during the same visit to discuss a possible neurosurgical operation. However, the high risk related to the cavernoma site combined with the generally good condition of the patient ruled out a neurosurgical approach to the pathology.
As regards the clinical symptoms, the patient reported a spontaneous, progressive reduction in the hyperalgesia and hiccup; he also reported the persistence of tactile allodynia and the appearance of daily, severe attacks of a paroxysmal stabbing, electric-like pain, which lasted milliseconds and was accompanied by persistent aching in between the attacks in the region of the right greater occipital nerve (GON), often radiating to the frontal region. Following the onset of these attacks, which induced anxiety and insomnia in the patient, we started pharmacological therapy with amitriptyline (30 mg/day) and asked the patient to return for a follow-up visit 1 month later. At this follow-up visit, we observed a good response to amitriptyline without adverse events. In particular, the patient reported a significant reduction in the frequency of the attacks (from four to five per day to one to two per week) and in the intensity of pain (using a 10-point numeric rating scale, from 8/10 to 3/10), as well as a marked reduction in the tactile allodynia.
No further episodes of migraine-like headache were registered during the follow-up period (6 months after referral to the headache centre), which supports the hypothesis that the patient did not have migraine and that his headache attacks were secondary in nature.
Discussion
We have divided the discussion of this case report into two parts. In the first, we discuss the clinical and physiopathological correlates of acute–subacute headache in our patient. In the second, we focus on the ON observed in the follow-up phase and its clinical, physiopathological and therapeutic aspects.
We describe a male patient who came to the ED because of headache with the following clinical features: (i) unilateral on the right side; (ii) throbbing; (iii) frontotemporal; (iv) lasting >1 week; (v) exacerbated by physical activity; (vi) refractory to symptomatic therapy. These clinical features are typical of a complication of migraine named ‘migraineous status’ (8). After both the neurological examination and the brain CT proved negative and intravenous NSAID administration effectively relieved symptoms, the patient was discharged and referred to our headache centre with a diagnosis of primary headache.
In spite of the ED diagnosis of primary headache, the negative brain CT and the headache-free status of the patient, when he was referred to our headache centre it was decided to continue the diagnostic process with an additional radiological study. The presence of right facial numbness and dysaesthesia as well as hiccup, combined with hyperalgesia and cutaneous tactile allodynia in the region of the right GON and along the three branches of the right trigeminal nerve, led us to study the bulbocervical region by means of an MRI examination.
As it is difficult to evaluate the above-mentioned symptoms objectively in an ED setting, the case was not deemed to require either emergency treatment or radiological studies other than a brain CT scan. Moreover, neurological examination did not reveal clear signs of myelopathy, which is in accordance with another case of upper cervical cavernous angioma in which the only symptom was ON (9).
The patient subsequently underwent an MRI study, which revealed a bleeding vascular lesion in the bulbar and upper cervical spinal regions. Collaboration between the ED and the headache centre thus led to a correct diagnosis of secondary headache in a patient with a migraineous status-like syndrome. In this regard, we support calls for specific guidelines not only for the diagnosis and treatment of primary headache in the ED (10), but also to help detect suspicious clinical symptoms, in the ED setting, in patients with secondary headache who are normal on neurological examination. Moreover, since headache is a common complaint in ED visits (11), the clinical collaboration between ED physicians and the headache team in evaluating all ED headache patients is, we believe, of utmost importance: it may enhance knowledge of migraine epidemiology, shed light on the more obscure aspects of this widespread pathology and lead to clinical guidelines for a comprehensive evaluation of ED headache patients being drawn up.
Case reports have often been used to lend support to the pathophysiological mechanisms hypothesized to underlie the migraine attack. In this regard, bleeding cerebrovascular malformations have recently attracted attention as a peculiar cause of symptomatic migraine attacks (12, 13). This is understandable if we consider that 10–30% of cavernous angiomas involve the brainstem and cerebellum (14) and that 5–10% of central nervous system cavernous angiomas arise within the spinal cord, particularly at the cervical level (15). Indeed, these regions have been implicated in the mechanisms that lead to a migraine attack (16) and chronic migraine (17).
This is, to our knowledge, the first case report to show a direct role of the upper cervical spine in the pathophysiology of migraine attacks in human. In our patient, bleeding occurred in the TCC, an anatomical and functional structure that plays a major role in allodynic phenomena during the migraine attack (18). Interestingly, our patient came to the ED with a unilateral migraine-like attack and, when admitted to the headache centre, presented homolateral cutaneous tactile allodynia. In this regard, this case supports the pathophysiological link between an involvement of the TCC and the central sensitization of trigeminal nociception in the development of migraine attacks in humans (19).
It is noteworthy that the site of the lesion corresponds to that believed to induce hiccup (20). The patient's symptomatology combined with the site of the cavernoma support the hypothesis that the anatomical correlates of hiccup lie in the dorsolateral region of the middle medulla.
At one of the follow-up visits, 20 days after the onset of headache, our patient presented right ON associated with the persistence of tactile allodynia in the region of the right GON, right lesser occipital nerve and right greater auricular nerve, a clinical picture very closely resembling a recent report of ON secondary to an upper cervical spine cavernous angioma (9). In both cases, neurological examination was negative for myelopathy, while ON was the main symptom of cervical cavernous angioma. Moreover, not only did both cases display frontal radiation of pain (beyond the territory of the GON) and respond well to amytriptiline, but the cervical tract location of the lesion was also similar, with the cervical medulla being involved at the C1–C2 level. There are, however, marked differences in the clinical history of the two cases: Cerrato et al. described the case of a woman with a 5-year history of pain without any MR evidence of recent bleeding (9), whereas in our patient the clinical course was acute–subacute and MRI revealed the presence of a lesion that had recently bled. These differences in the clinical picture were probably due to the acute bleeding of the vascular lesion in our patient, an hypothesis supported by the migraine syndrome observed when the patient came to the ED. Indeed, the bulbocervical nature of the lesion together with the meningeal irritation (i.e. the neck stiffness upon awakening described by the patient during the first visit to the headache centre) may explain this more complex syndrome comprising secondary neurovascular headache followed by secondary ON.
Possible aetiologies for ON are irritation of the GON at its origin because of arthritis in the C1–C2 segment, entrapment at the points where the GON pierces the semispinalis capitis muscle, trapezius muscle aponeurosis, or an entrapment in the suboccipital region in the GON's course through the neck muscles, especially the obliquus capitis inferior muscle (21).
We believe that that the development of ON in our patient may have been related to molecular mechanisms (caused, in this case, by an upper cervical haemorrhage) that underlie central neuropathic pain. Indeed, it should be borne in mind that his neuralgic symptoms started 20 days after the onset of the migraine-like pain. Animal model studies have shown that spinal cord sensitization and gene expression changes may begin several days after the injury that is responsible for neuropathic pain (22). This may explain the latency between the onset of migraine-like symptoms (and presumably the onset of bleeding) and the onset of ON.
From a clinical point of view, the frontal radiation of the ON described by our patient, combined with the timing of the onset of pain and allodynia, strongly suggests an involvement of TCC mediated by facilitation or sensitization of central nociceptive neurons.
In accordance with Cerrato et al. (9), we suggest that an upper cervical cavernous angioma may be a potential cause of secondary ON. An MRI study of the cervical spine is, consequently, to be recommended even in patients with ON with no signs of myelopathy on neurological examination.
We conclude that close collaboration between ED physicians and headache specialists may help to make a diagnosis in patients referred to the ED for both primary and secondary headaches. A more comprehensive clinical evaluation of ED headache patients may also encourage development of clinical guidelines for the diagnosis and treatment of primary and secondary headaches in ED. Moreover, our case points to a role of the TCC not only in the pathophysiology of primary headaches such as migraine, but even in other kinds of chronic pain in the craniofacial region.
In conclusion, we strongly believe that a more comprehensive and extensive radiological study including the bulbar and cervical regions should be performed in patients who present a clinical syndrome suggestive of impairment of the TCC. In particular, we propose that migraine-like headache attacks or ON associated with homolateral persistent, widespread cervico-facial allodynia and long-lasting hiccup should be investigated even in patients with a negative neurological examination.