Abstract
Short-lasting unilateral neuralgiform headache with conjunctival injection and tearing (SUNCT) syndrome represents a serious therapeutic and diagnostic challenge, since it is usually refractory to most drugs and lacks biological markers. Response to intravenous lidocaine administration has been reported in some patients while it has failed in others. We report a patient with SUNCT syndrome who showed a clear-cut and consistent response to intravenous lidocaine therapy, which proved superior to placebo in a single-blinded fashion. Intravenous lidocaine should be considered in all patients with SUNCT syndrome. Response to this therapy could represent a diagnostic tool.
Short-lasting unilateral neuralgiform headache with conjunctival injection and tearing (SUNCT) syndrome is an infrequent form of unilateral headache accompanied by cranial autonomic features (1). Pain is severe, disabling, and typically refractory to therapy. A positive response to intravenous (i.v.) lidocaine has been reported in some patients (2, 3), although it has not been confirmed by other authors (4).
We here report on a SUNCT patient who showed a clear-cut and consistent response to i.v. lidocaine, which proved superior to placebo in a single-blinded fashion.
Case report
A 70-year-old man was admitted because of recent-onset, progressive headaches. He had started 1 month previously with attacks of periocular left pain, associated with lacrimation and conjunctival injection, which repeated 5–10 times a day with no particular regularity and lasted for up to 5 min according to his report. With a suspicion of cluster headache and then paroxysmal hemicrania, he had received corticosteroids (60 mg daily for 3 weeks, progressive tapering), topiramate (200 mg daily for 2 weeks) and indomethacin (150 mg daily for 1 week); all of these drugs failed. Transdermal fentanyl was started (initial dose 25 µg/h, increased to 50 µg/h after 3 days), but pain attacks continued. Ophthalmological evaluation ruled out glaucoma and any other intrinsic eye disease. Brain magnetic resonance imaging was normal.
His past medical history was unremarkable and he had no history of headaches. Both general and neurological examinations were normal between attacks.
During admission, attacks were witnessed. They were abrupt, repeated up to 30 times a day, and consisted of severe unilateral, left periocular pain accompanied by prominent dysautonomic features (lacrimation, Horner's syndrome and conjunctival injection). Attacks occurred as single stabs (5) and their timing revealed a duration of 30–60 s. Touching the nose or sneezing triggered pain without a refractory period. He was pain free between attacks. A diagnosis of SUNCT was made and i.v. lidocaine was started at a rate of 1.3 mg kg−1 h−1. Fentanyl (50 µg/h) was continued and lamotrigine was initiated (to a dose of 200 mg daily after 1 week). The patient kept a diary of the attacks so that we could establish their number and duration.
The patient reported a prompt positive response to lidocaine, and infusion was maintained over 4 days. Approximately 10 min after infusion was initiated, the attack rate decreased significantly and he was almost pain free. To evaluate the response better, informed consent was obtained to administer placebo in a single-blinded manner (i.e. i.v. saline from a bottle that was also covered with foil). The infusion of lidocaine was stopped and changed for placebo for 2 h approximately; lidocaine was then administered again for another 2 h and then switched again to placebo. This process was repeated three times a day for 5 days at different times of the day (morning, afternoon, evening). After day 5, the patient was informed that he was to receive only lidocaine from that moment on, and that placebo was withheld.
The patient was consistently able to tell when placebo was administered 100% of the time, as reflected by reappearance of the attacks. Pain would reappear 2–3 min after lidocaine had been stopped. Perfusion had to be prolonged for 14 days, and was then tapered to stop over 1 week.
He was discharged on lamotrigine (200 mg/day) and transdermal fentanyl. He remained pain free for 2 months after discharge, and attacks reappeared. A new infusion of i.v. lidocaine was started at the same dosage, again with a positive clinical response. The pain reappeared whenever the infusion rate was halved. After 9 days of therapy, lidocaine could be discontinued because the patient was free of pain. He was discharged on lamotrigine and fentanyl. These drugs were discontinued after 6 months. During 1 year's follow-up the patient has shown no further relapses.
Discussion
The patient described here met the International Headache Society criteria for SUNCT syndrome (6), i.e. at least 30 attacks of unilateral, moderately severe, orbital or temporal, stabbing or throbbing pain, lasting for 15–120 s, associated with one of the following cranial autonomic features: conjunctival injection, lacrimation, nasal congestion, rhinorrhoea, ptosis and eyelid oedema. The attacks have to occur at a frequency of 3–100 daily during the active phase. The natural history of the condition is poorly understood. Generally, a clustering pattern is reported, with active phases erratically alternating with remissions (1,7). Diagnosis relies on clinical features, since there are no biological markers for this entity, which is often confused with other trigemino-autonomic cephalalgias, particularly trigeminal neuralgia on its first division (8).
Numerous therapies have been tried, most in single patients or small series, usually with poor results. Pareja and colleagues reported that i.v. lidocaine 100 mg bolus followed by 4 mg/min over 48 h was ineffective in two patients with SUNCT (4). In contrast, Matharu et al., in 2004, reported four cases highly responsive to i.v. lidocaine (2), and the same group later described the response of 52 patients with SUNCT to different therapies (5). Whereas oxygen and indomethacin administration failed in all patients in whom it was tried (10 and 12, respectively), and different levels of response were observed with lamotrigine, topiramate, gabapentin, carbamazepine, or with greater occipital nerve blockade, 11 patients were treated with i.v. lidocaine and all showed a favourable response.
Williams and Broadley administered lidocaine on 17 occasions to 14 patients, intravenously in three and subcutaneously in the rest. They obtained a fairly high response to lidocaine, since attacks were abolished in 76% of cases, which further supports lidocaine's effectiveness in this rather refractory primary headache (3).
Intravenous lidocaine also resulted in complete pain relief in our patient, in whom other therapies had failed. Response was clear-cut and consistent over time, and clearly superior to placebo in a single-blinded manner.
Intravenous lidocaine administration carries risks, and requires careful monitoring, particularly of ECG and blood pressure. Interestingly, headache was a side effect of lidocaine in eight of 106 patients with fibromyalgia (9). Unfortunately, there are no good alternatives to administer lidocaine by a different route, since transdermal patch application works only at a topical level.
The mechanism of action of lidocaine in SUNCT is unclear. Attacks reappeared soon after the drug was stopped and switched to placebo, which may be consistent with a lidocaine half-life of 10 min after i.v. bolus administration; with continuous i.v. administration a half-life of about 1.5–2 h may be achieved.
Unresolved issues regarding i.v. lidocaine administration in SUNCT patients include length of therapy, which probably differs from one patient to another, and whether lidocaine response could also be employed for diagnostic purposes in a similar manner to the indo-test for paroxysmal hemicrania.
Intravenous lidocaine should be tried in patients who meet SUNCT criteria to alleviate their pain attacks, and as a potentially useful diagnostic tool.