Abstract
Introduction
Sjaastad and coworkers first described in 1978 a patient with headache attacks associated with conjunctival injection appearing in clusters (1). SUNCT syndrome, an abbreviation for short-unilateral neuralgiform headache attacks with conjunctival injection and tearing, was completely characterized by Sjaastad in 1989 (2). SUNCT is not a frequently encountered disease, with a prevalence estimate closely resembling the 0.5 per 1000 of paroxysmal haemicrania (3).
According to the International Classification of Headache Disorders, SUNCT is one of the trigeminal autonomic cephalalgias (4). There are five criteria that must be met to diagnose SUNCT. There must be at least 20 attacks of unilateral orbital, supraorbital or temporal stabbing or pulsating pain lasting from 5 to 240 seconds, accompanied by ipsilateral conjunctival injection and lacrimation. Each attack must occur with a frequency of 3 to 200 per day and must not be attributed to another disorder (4).
SUNCT syndrome is often refractory to optimal medical management. We present the first case of a patient who was successfully treated by gamma knife radiosurgery after failing medical treatment.
Case report
A 76-year-old male came to clinical attention in April 2001, with a 4-month history of regular paroxysmal facial pain. He described an electric shock sensation occurring with a frequency of 3 to 10 times per day, departing from the right nostril and irradiating throughout the distribution of V2. The pain highest intensity lasted 30 seconds to 2 minutes approximately. These symptoms were triggered when he spoke, ate or touched his cheek. There were no associated autonomic symptoms. His neurological examination was normal at that time. The patient was diagnosed with trigeminal neuralgia and was given a prescription of carbamazepine 100 mg daily. This was to be started only in case of recurrence of his symptoms, considering that there were no attacks at time of the visit. Nine months later, the patient had started the medication and consulted again because of increased pain on medication. His dosage was increased to 100 mg three times a day and the patient was suggested to increase it up to 800 mg daily. A brain MRI was done and revealed only scattered white matter anomalies of probable vascular origin, which were deemed unrelated to his clinical condition. The patient was relieved temporarily, but the medication was changed to oxcarbazepine by his family doctor in the following year. Six years later, the patient presented, during an attack, with episodic right-sided orbital and supraorbital electrical shocks associated with ipsilateral lacrimation and nasal congestion. Oxcarbazepine had been ceased after years of control from 2004 to February 2007, but restarted by the patient himself when symptoms recurred a few months before this visit. These attacks lasted approximately 15 seconds and recurred every 30 to 60 seconds. His family doctor had recently ceased oxcarbazepine and started indomethacin, which led to an increase in the intensity and frequency of pain attacks. The neurological exam showed right conjunctival injection, which had not been noticed by the patient, with miosis and ptosis. The clinical picture was now compatible with SUNCT syndrome, with associated Horner syndrome. Oxcarbazepine was restarted, indomethacin was stopped and lamotrigine was added. A repeat MRI remained unremarkable. Despite modest pain control, oxcarbazepine and lamotrigine were discontinued and replaced by gabapentin because of intolerable side effects, consisting of balance difficulties, anorexia, weight loss and hyponatraemia. However, attacks kept increasing in frequency and intensity. A prednisone trial led to no improvement. At that point, he was referred to our department for consideration of stereotactic radiosurgery. After carefully considering the risks and benefits of the procedure, the patient accepted.
The patient was 82 years old at the time of radiosurgery. A Leksell G frame was installed under local anaesthesia, with mild sedation. A 1-mm thick axial contrast-enhanced magnetization prepared rapid gradient echo (MPRAGE) MRI sequence was obtained for treatment planning. Because of similarities between SUNCT and cluster headache, it was elected to treat both the trigeminal nerve and the sphenopalatine ganglion (5). The trigeminal nerve was targeted at the root entry zone with an 80 Gy maximum dose, using a single 4-mm isocentre. The sphenopalatine ganglion, which was localized with the assistance of the neuroradiologist, was treated with a maximum dose of 80 Gy using a single 8-mm isocentre.
The patient reported pain relief a few hours following the treatment, with complete pain disappearance 2 weeks after. After 1 year, the gabapentin dose was progressively reduced. The patient discontinued gabapentin 28 months after radiosurgery. On the latest follow-up, 39 months after the procedure, he remains completely pain free and off medication. He did not suffer from any side effect or complication from the procedure. The Horner syndrome resolved partially after radiosurgery.
Discussion
SUNCT syndrome is a facial pain syndrome that leads to important disability for afflicted patients and is often refractory to medical therapy (6–9). It can be difficult to correctly diagnose it, with some clinical features overlapping other facial pain diseases, such as trigeminal neuralgia and cluster headache. It is classified as a trigeminal autonomic cephalalgia. The clinical criteria for SUNCT are well established and were mentioned earlier (4).
Some authors claim that SUNCT could be a variant trigeminal neuralgia (7). Our patient was initially diagnosed with trigeminal neuralgia, and a diagnosis of SUNCT was established later in his evolution. Transformation or evolution of a trigeminal neuralgia to SUNCT has been reported previously (10,11). A case of coexistence of both entities has been published, before final transformation into SUNCT (11). In our patient, it would be appropriate to consider the successive diagnosis of both entities: trigeminal neuralgia initially, which successfully resolved with medical therapy, followed by appearance of SUNCT a few years later. It has been suggested that some cases of trigeminal neuralgia are misdiagnosed SUNCT (10). A transformation seems less likely in our case, because the V1 ophthalmic branch was not affected initially. There is also a clear period of time during which our patient was pain free. The differential diagnosis to consider is a trigeminal neuralgia with autonomic symptoms. Considering the previous argument mentioned and the fact that lacrimation is generally the only autonomic symptom in patients with trigeminal neuralgia with autonomic symptoms, we do not think that this was the appropriate diagnosis in our patient (10). In addition, our patient did not have any accompanying autonomic symptoms initially. Finally, trigeminal neuralgia usually responds to carbamazepine, a drug to which our patient was refractory.
Generally, the neurological examination is normal in SUNCT patients. However, many cases with either allodynia or hyperaesthesia and post-traumatic SUNCT with sensory deficit in V1 have been described (12–16). There are reports of associated persistent ipsilateral ptosis and miosis for which no other lesion can be identified (9,13,17,18).
Magnetic resonance imaging usually reveals no significant anomaly in SUNCT patients. Articles reporting scattered cerebral white matter lesions have been published, but such findings are frequent in older patients and no formal relation with SUNCT can be identified (9,19).
In a recent report from Leone and Bussone on the pathophysiology of trigeminal autonomic cephalalgias, the source of autonomic symptoms was hypothesized to be either the trigeminal discharge itself, dysregulation of autonomic centres, compression of pericarotid sympathetic fibres due to vasodilatation or perivascular oedema, or evoked by parasympathetic overactivity during attacks (5).
As with other facial pain syndromes, surgical therapy can be attempted to relieve SUNCT patients who are refractory to medical treatment. The minimally invasive options are always sought first. Local blocks of the greater occipital nerve have been published with successful results (20,21). Occipital nerve and hypothalamic deep brain stimulation are starting to enter the therapeutic arsenal of trigeminal autonomic cephalalgias (22–24). In view of the autonomic component of the syndrome, hypothalamic deep brain stimulation has been proposed (25–27). There are only a few cases published so far. In one case, prolonged stimulation resulted in long-lasting pain relief without drugs (27). The second case showed a 63% decrease in mean attacks per day and a 80% reduction at 12 months (26).
Considering that the sphenopalatine ganglion is a parasympathetic relay for the autonomic fibres of the facial nerve towards the lacrimal and nasal mucosal glands, and knowing that the zygomaticotemporal nerve, a sensitive branch from the trigeminal nerve, passes through the sphenopalatine canal, this area makes sense as a target for temporary or permanent destructive procedures (28–31). Gamma knife radiosurgery is a minimally invasive procedure that is now increasingly being used to treat trigeminal neuralgia patients (32–35). It has also been proposed for refractory cluster headache and sphenopalatine neuralgia, with lower success rates than for trigeminal neuralgia (29,36–38). To improve the outcome of the procedure, some authors have suggested adding the sphenopalatine ganglion as a target in addition to the trigeminal nerve (28,30,31,39). There has been only one other publication reporting the effect of gamma knife radiosurgery for SUNCT patients (6). In this report, the authors described two patients who had medically refractory SUNCT. The first patient had glycerol rhizotomy with pain relief lasting 2 weeks. He then underwent radiosurgery targeting the trigeminal nerve with 90 Gy. He had pain relief for 2 months, but had pain recurrence with anaesthesia dolorosa complicating the treatment. The second patient did not have any pain relief or side effect after radiosurgery. The trigeminal nerve was the only target in those patients. Our patient had radiosurgery targeting both the ipsilateral trigeminal nerve and sphenopalatine ganglion with a maximum dose of 80 Gy. The dual targets might explain why the procedure was a success compared with what was previously reported.
Conclusion
Our case illustrates that gamma knife radiosurgery can be successful to relieve pain attacks in medically refractory SUNCT patients. Because of the autonomic component of the SUNCT syndrome, we believe targeting both the trigeminal nerve and sphenopalatine ganglion will yield superior results than the single trigeminal nerve target.
