Abstract
Short-lasting unilateral neuralgiform headache with conjunctival injection and tearing (SUNCT) syndrome was first described in 1978, fully characterised in 1989 by Sjaastad (1) and recognised as a subgroup of the trigeminal autonomic cephalalgias in the second edition of the International Classification of Headache Disorders.
Even though SUNCT is defined as a primary headache, sporadic ‘symptomatic SUNCT’ cases have been described in the literature since 1991, associated with posterior fossa lesions and vascular malformations (2,3), craneosynostosis (4), pituitary adenomas (5), tumours of the cavernous sinus (6), intraorbital masses (7), paranasal sinusitis (8) and viral meningitis (9), among others. Formally, because these cases fail to fulfil criterion E for SUNCT (‘symptoms not attributed to another disorder’), they should in fact be called ‘SUNCT mimics’ and classified as secondary headaches.
We report the first known case of a cavernous sinus dural fistula mimicking SUNCT: a 46-year-old Caucasian male, without previous headache or other neurological symptoms, presented with a 3-month history of recurrent lancinating episodes (about 10 per day) of acute cephalalgia in the left periorbital area, together with ipsilateral blurred vision, lacrimation and conjunctival injection, lasting for approximately 2 minutes. The patient did not identify any triggers for the episodes.
The interictal physical examination showed left eye blurred vision, a very mild left palpebral ptosis, eyelid oedema and infero-medial left eye deviation. The patient was treated symptomatically with indomethacin (75 mg/day), carbamazepine (400 mg/day) and prednisone (60 mg/day) without benefit. Lamotrigine (50 mg/day) was then added with some improvement.
Brain magnetic resonance imaging (MRI) demonstrated mild bulging, intracavernous voids (T2FSE sequence) and signs of arterialisation (3D TOF MR angiography sequence) of the left cavernous sinus, consistent with a cavernous sinus dural fistula.
He was submitted to a digital subtraction angiography (diagnostic and therapeutic) that confirmed the diagnosis and showed that the dural fistula was fed by meningohypophyseal branches of both internal carotid arteries (ICA) and by terminal meningeal branches of both internal maxillary arteries (the right arteries crossing the midline). Venous drainage took place to inferior and superior petrosal sinus, to the right cavernous sinus and to the left sphenoparietal sinus, with retrograde drainage into frontal cortical veins implicating a high risk of bleeding (type 2 A + B on the Cognard dural fistula classification system) (Figure 1).
Digital subtraction angiography: on the left the diagnostic procedure and on the right after the embolization.
Current treatment options for cavernous sinus dural arteriovenous fistula is primarily endovascular. The transvenous route is the preferred one as it has higher cure rates and lower incidence of complications (10). Transarterial embolisation with microparticles is found to be frequently inefficient and carries a higher risk of complication, due to the anastomoses between external carotid artery and ICA.
In our case, it was possible to markedly decrease the flow after transvenous intracavernous deployment of multiple platinum coils but, as this was not completely successful, treatment was complemented by the injection of microparticles into some of the arterial feeders. The shunt couldn’t however be completely excluded, as some of the smaller arterial feeders were impossible to embolise.
The patient achieved full symptomatic improvement (pain episodes disappeared and vision became normal) and remains stable after 1 year of follow-up. Although the procedure was not able to completely exclude the shunt, the flow was greatly decreased and retrograde drainage into cortical veins was stopped. We believe that this fact, along with a probable spontaneous thrombosis of the remaining shunt, might have been responsible for the complete symptom resolution.
Although most of our patient’s symptoms could be explained by the presence of a cavernous sinus dural fistula, we would expect a continuously progressive clinical set, instead of a paroxystic one, which resembles a ‘SUNCT syndrome’.
The pathophysiological mechanism of SUNCT is not completely understood. Currently, a central component (involving the trigeminal-autonomic system and the posterior hypothalamus) is believed to predominate in the pain mechanism of primary SUNCT forms. On the other hand, it has been hypothesised that symptomatic SUNCT cases can be attributed to the irritation of pain-sensitive structures and activation of the trigeminal nerve endings in places such as the sella turca and the cavernous sinus, supporting a role for peripheral structures in its pathophysiology. In fact, the activation of the trigeminal fibres by the different lesions is the proposed mechanism by some of the previously reported cases of secondary SUNCT syndromes (6) and could hence be a possible explanation for our own case (11).
This case highlights the importance of performing a brain MRI and angio-MRI in all SUNCT patients, in order to rule out vascular pathology and other structural lesions.
To our knowledge, no other cases of an ipsilateral cavernous sinus dural fistula mimicking SUNCT have been previously described.