Abstract
A 54 year-old female has a clinical diagnosis of chronic paroxysmal hemicrania (CPH) according to IHS-criteria (2004) three years ago. Six months after that, she developed symptoms compatible with trigeminal neuralgia (TN) involving the second and third trigeminal territories also at the left side. Cranial and cervical spine resonance magnetic images showed extensive cervical syringomyelia that included nucleus caudalis (C2 level) region and a posterior fossa Chiari I malformation without hydrocephalus. The association between CPH and TN is called chronic paroxysmal hemicrania-tic syndrome (CPH-tic). This is the first clinical description of a secondary cause of CPH-tic syndrome in the literature.
Keywords
Introduction
The association of paroxysmal hemicrania (PH) with trigeminal neuralgia (TN) was recently described and called paroxysmal hemicrania-tic syndrome. There are a number of reports in the literature describing the coexistence of PH and intracranial lesions, which might play a role in the causation of PH. Secondary causes of PH–tic have not been previously described in the literature. We report an unusual association of PH–tic and Chiari I malformation.
Case report
A 54-year-old female patient had a 3-year history of severe recurrent left temporal-parietal and orbital pain. Each painful episode lasted for 2–20 min. The frequency of the attacks ranged from five to 12 a day. Painful episodes were accompanied by moderate ipsilateral autonomic activation (tearing and conjunctival injection). During this period, she had no pain-free episodes.
Six months after the beginning of these painful episodes, the patient developed another type of pain described as stabbing, with a duration of 1–2 s, affecting only the second and third left trigeminal areas. It occurred many times a day, daily, spontaneously or induced by speaking, eating, washing her face or brushing her teeth. Due to the high frequency of both types of pain, the patient stated that occasionally both types of pain started at the same time. She could not recall precisely how often that happened.
She was first prescribed carbamazepine (CBZ) 1000 mg daily, with a poor clinical response (she presented only a decrease in pain intensity but no change in the frequency of the TN symptoms). Next, she was started on 400 mg a day of gabapentine (GP) and both symptoms worsened. Her first appointment in our out-patient clinic occurred almost 4 years after both symptoms had started. At this time she was again taking only CBZ 600 mg, but still had frequent episodes of severe pain.
Clinical and neurological examinations were normal. Cranial and cervical spine magnetic resonance imaging revealed an extensive cervical syringomyelia with its upper limit at the level of C2 that included the spinal trigeminal nucleus (nucleus caudalis). There was no hydrocephalus, but cerebellar tonsils had descended into the vertebral spinal canal (posterior fossa), characterizing a Chiari I malformation (Figs 1 and 2).
Sagittal T1 weighted image after gadolinium injection shows ectopic tonsils (arrow). Note the effacement of the convexity without significant hydrocephalus.
Sagittal T1 weighted image of the cervical spine shows an extensive hydrosyringomyelia suggestive of Chiari I malformation (arrow). The involvement extends to the upper cervical levels.
The association of indomethacin (ID) (150 mg/day) and CBZ 600 mg/day substantially decreased the number of both pain episodes. When the ID dose was raised to 250 mg/day with CBZ maintained at 1000 mg/day, the patient was pain free. Her last appointment occurred in July 2006, when she complained of short-lasting PH episodes twice a week with autonomic activation and no more TN symptoms. The patient has had no new neurological symptoms since the time of the first diagnosis.
Discussion
The trigeminal-autonomic cephalgias (TACs), which include cluster headache (CH), paroxysmal hemicranias and short-lasting unilateral neuralgiform headache with conjunctival injection and tearing, are a group of recurrent primary headaches of brief duration that probably share common pathophysiological mechanisms (1).
Differential diagnosis between CH and PH is limited by the considerable overlap of their clinical characteristics. Maximal pain localization, mean attack duration, mean attack frequency, age and gender can help in the differential diagnosis. There are no clinical characteristics that exclusively belong to one of these headache entities (2). Boes wrote in an editorial that, given the phenotypic overlap between CH and PH, requiring an indomethacin response is the only way to identify a group that can be rigorously studied to determine the aetiology of the disorder (3).
The pain episodes described in this report met the International Headache Society (IHS) criteria for PH and TN (1). The association between these two entities (PH–tic) is rare and, to the best of our knowledge, only eight cases had been described in the literature (4, 5). The relationship between the two entities is not completely understood. It is possible that abnormalities include impaired inhibitory mechanisms that normally control afferent activity in trigeminal nucleus, as well as a hypothalamic dysfunction (6).
Secondary causes of PH described in the literature include: gangliocytoma of the sella turcica, collagen vascular disease, cerebrovascular disease, Pancoast's tumour, cavernous sinus meningioma, frontal lobe tumour, intracranial hypertension, internal carotid artery aneurysm, tuber cinereum hamartoma, occipital infarction and severe orbital imflamation (7–9).
In 2001, Penarrocha described a patient with clinical symptoms of orofacial pain (neurogenic trigeminal pain) and episodes similar to CH on the same side as the sole manifestation of syringobulbia-syringomyelia with Chiari malformation. The symptoms resolved after amplification of foramen magnum (10). There is only one case report of a possible secondary PH–tic which describes a patient with vertebrobasilar ectasia that had a close anatomical relationship to the left trigeminal nerve root entry zone (4).
Syringomyelia is a disorder attributed to different causes, characterized by obstruction to the passage of cerebral spinal fluid at the level of the posterior fossa, usually associated with an anatomical malformation at this level (11).
Chiari I malformation is characterized by the presence of the cerebellar tonsils 3–5 mm below the magnum foramen (11). Patients with Chiari I can be asymptomatic or have symptoms such as headache, paraesthesias, motor deficits and gate disturbances, but TN is rarely the only manifestation of the malformation (12–14). They often present syringomyelia.
There are fewer than 10 reports with the association of Chiari I and TN (15). The trigeminal pathway of the pain is not yet completely understood and the relationship between these two entities is still uncertain. Compression or deformity of the trigeminal nucleus in the brainstem due to a posterior fossa malformation has been implicated as one possible explanation (15, 16).
Seijo-Martinez et al. published a case report of cluster-like headache in association with cervical syringomyelia and Chiari malformation. Several features were atypical, such as the fact that some pain attacks were provoked by positioning the head in a stooped posture (17). The authors attempted to explain this association of disorders by the fact that the spinal trigeminal nucleus and its pathway extend from the junction of the pons and medulla to the uppermost segments of the spinal cord, probably including C2 and C3 levels. They believed that the trigeminal pathway of pain was also potentially affected by syringomyelia (17). Our patient described no similar or other pain trigger such as performing a Valsalva manoeuvre. In patients with TN and Chiari I malformation described in the literature and surgically treated (posterior fossa decompression, ventricular peritoneal shunt or third ventriculostomy), complete resolution of TN symptoms has been observed (12, 15, 16, 18).
In 2004, Trucco published a review article about TACs, in which only one patient with PH and Chiari I malformation was found, who did not meet the IHS criteria for PH (9).
Conclusion
The association of PH–tic and Chiari I with syringomyelia has not been previously described in the literature. Since our patient responded in the usual way to standard therapy for both conditions, the question remains, whether there is any underlying mechanism between these entities or whether this association has occurred only by chance. Six years after her first symptoms, our patient has not required surgical intervention for either the syrinx or the Chiari I, because she has responded well to medical treatment and no other neurological symptoms have developed. Her condition is still being followed in our out-patient clinic. It is hoped that this report may contribute to additional understanding of the mechanisms underlying the PH–tic association.