Abstract
Short-lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) is characterized by severe and frequent daily pain attacks causing transient physical disability for the patients during the headache period. Currently there is no option for abortive treatment of the attacks, mainly due to the short-lived nature and frequency of the repeated headaches, while highly efficacious therapy is also unavailable for short-term prevention. We report rapidly suppressed headache attacks with orally administered methylprednisolone in eight headache periods of three patients with idiopathic, episodic SUNCT syndrome. The remission was maintained until the period was over in all cases. Although the mechanism of methylprednisolone action is unclear, it is probably based on the anti-inflammatory effects of the drug.
Introduction
Trigeminal autonomic cephalalgia (TAC) is a collective term that refers to a group of headaches characterized by episodic or chronic, short-lasting, unilateral head or face pain with accompanying autonomic features (1,2). TACs include cluster headache, paroxysmal hemicrania and short-lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) (1, 2).
The pathophysiology of TACs involves activation of trigeminovascular nociceptive pathways (1) with cranial autonomic activation, which is due to central disinhibition of the trigeminal autonomic reflex by the hypothalamus (3–8). Despite the fact that TACs share common pathophysiological pathways, there are dissimilarities among the headache syndromes not only in the basic clinical features (e.g. in SUNCT the duration of pain and autonomic symptoms is the shortest, the incidence of daily attacks is the highest), but also in response to therapy (1).
Most of the current treatment options for TACs are rather empirical than pathophysiology-based, and little is known concerning the mechanism of action of the pharmacological agents used as acute and prophylactic therapy. In cluster headache and in paroxysmal hemicrania there are some accomplished and well-tolerated, therefore recommended, acute and preventive medications. However, in the rare SUNCT syndrome we have only case reports with frequently disappointing therapeutic effectiveness. Abortive therapy of SUNCT is meaningless because of its characteristic short and frequent pain attacks (9,10). Intravenous lidocaine has been suggested for short-term prevention (9,10), but this treatment requires hospitalization, and complete pain relief was only in a fraction of the treated patients reported (11–13). Among all medications attempted in SUNCT, lamotrigine was most effective for long-term prevention (12–17), but patients need to be particularly vigilant during the initiation of therapy because of possible skin reactions, which may even progress to Stevens–Johnson syndrome. The risk can be reduced by slow introduction and titration; however, during this period, which might last for a few weeks, the patient has no efficient symptom relief. In cluster headache corticosteroids are very effective in suppressing the headache period and are therefore recommended for short-term prevention (10,18,19). In SUNCT syndrome only a few corticosteroid-treated cases have been reported with partial effectiveness (20–23).
During the past few years, only three patients with idiopathic, episodic SUNCT syndrome were referred to our Outpatient Headache Department. All patients had been unsuccessfully treated during their previous headache periods. Based on the success of cluster headache corticosteroid treatment, we tried methylprednisolone therapy for short-term prevention in all three patients to minimize their disability during the headache period.
Case histories
Patient 1
A 60-year-old man was examined for frequent headache paroxysms (15–40/day) lasting 30–150 s with few minutes-long refractory periods between the attacks. Although most of his attacks were spontaneous, he could sometimes trigger the headaches with forced and repeated eye closure. The left-sided orbital/supraorbital pain [intensity 10/10 on the visual analogue scale (VAS)] was associated with ipsilateral cranial autonomic symptoms such as conjunctival injection, tearing, eyelid oedema, nasal congestion and rhinorrhoea. The headaches occurred in cluster periods, once or twice every year, lasting for 30–60 days for 13 years. There was no chronic disorder in his medical history and he was not on regular medication. He did not smoke and there was no history of other headaches. Neurological and ophthalmological examinations, routine blood tests and 3T brain magnetic resonance imaging (MRI) studies were normal. Previous treatments with non-steroid anti-inflammatory drugs (NSAIDs; diclofenac, ibuprofen, indomethacin, naproxen), carbamazepine, gabapentin and pregabalin had not proved effective. He took methylprednisolone in a daily dose of 1 mg/kg for 2 weeks, then in reduced dose (16 mg dose reduction/week) for 4 weeks. The therapy had been repeated four times during the last 3 years. The medication was started on the first day of the headache bout in each period, and the steroid therapy lasted for 4–6 weeks. To decrease the possible side-effects of corticosteroid treatment, the histamine H2 receptor antagonist famotidine in a daily dose of 80 mg and potassium chloride in a daily dose of 1000 mg were co-administered with the methylprednisolone therapy. His pain intensity and the number of daily headaches decreased dramatically within 48 h, then he became attack-free in all treated periods. Treatment-related side-effects did not occur and repeated blood tests showed only transient elevation of serum hepatic transaminases and white blood cell count.
Patient 2
A 75-year-old non-smoker man had suffered from repeated headache periods for 11 years. The periods lasted for 20–40 days and had appeared not more than once per year during periods of weather transition in the spring or autumn. He complained of stabbing and burning pain in the left orbital, and periorbital regions along with tearing and conjunctival injection. The intensity of his headaches measured by VAS was 10/10. The duration of pain attacks was 45–120 s and the headaches occurred 20–70 times every day without triggering refractory periods between the attacks. There was no reported nausea or vomiting, and he could trigger the attacks with speaking, eating and face grimacing. The patient did not have any other type of headache. His medical history consisted of well-controlled type II diabetes, hypertension and cardiac atrial fibrillation. Neurological, ophthalmological and otorhinolaryngological examinations were normal; a routine blood test showed only an increased International Normalized Ratio level, which was explained by chronic anticoagulant therapy. 3T brain MRI images demonstrated small signal abnormalities in the hemispheric white matter, but the brainstem was intact and there was no sign of trigeminal nerve vascular or non-vascular compression. Previous medications including NSAIDs (diclofenac, ibuprofen), clomipramine, carbamazepine, gabapentin and pregabalin did not alleviate the symptoms. He was treated with methylprednisolone (1 mg/kg every day) during his last two headache periods. The therapy started on the first day of the attacks in each period, and both the intensity and frequency of the attacks improved within 3 days and then he became pain-free. The corticosteroid treatment was tapered after 2 weeks by 4 mg every day and was finally discontinued within 3 weeks. The remission was maintained during the treatment period and the headaches did not relapse afterward. To decrease the possible side-effects of corticosteroid treatment, the histamine H2 receptor antagonist famotidine at a daily dose of 80 mg and potassium chloride at a daily dose of 1000 mg were co-administered with the methylprednisolone therapy. The dose of his regular oral antidiabetic drug was increased during the treatment period. The patient tolerated the treatments well during both periods, and control blood tests revealed transiently and mildly elevated hepatic transaminases, white blood cell count and glucose levels.
Patient 3
A 27-year-old man had reported his first headache period 8 years before his neurological examination. The short-lasting (5–45 s), frequent (up to 100/day), severe (10/10 in intensity by VAS) and unilateral headaches developed in his right temple, eye and upper teeth. There was no refractory period between the attacks. The pain was sharp and stabbing. There was no nausea, light or noise sensitivity. The headaches were associated with ipsilateral conjunctival injection, tearing and nasal congestion. The pain could be triggered by touch of the forehead, temple and chin, as well as by chewing, drinking cold water and brushing of teeth. The headache period occurred twice per year (spring and autumn) and lasted for 2 or 3 weeks. He was a non-smoker and past medical history revealed only Gilbert syndrome. Ear, nose, teeth and throat examinations were negative, and he did not have any pathological neurological signs. Routine laboratory tests showed only a mildly increased bilirubin level, while brain 3T MRI was normal. The patient was not on a daily medication, and previously he had used different NSAIDs (diclofenac, ibuprofen, indomethacin, nimesulide), carbamazepine and amitriptyline for his headache without any effect. He had taken methylprednisolone at a 1 mg/kg daily dose on the second day of his last headache period and his pain attacks had completely disappeared within 48 h. To decrease the possible side-effects of the corticosteroid treatment, the histamine H2 receptor antagonist famotidine at a daily dose of 80 mg and potassium chloride at a daily dose of 1000 mg were co-administered with the methylprednisolone therapy. After 2 weeks' corticosteroid therapy the daily dose was reduced by 8 mg every day until it was discontinued. The patient reported no treatment-related side-effects and the control blood test parameters were within normal limits.
Discussion
In this study we treated three patients with oral administration of methylprednisolone in daily doses of ≤ 1 mg/kg. All patients fulfilled the International Headache Society classification criteria of episodic SUNCT syndrome. The headache bouts were refractory to all previously used medications, including different NSAIDs, antiepileptic and tricyclic antidepressant drugs. We have treated a total of eight SUNCT episodes with oral methylprednisolone in these patients, and the headache periods were completely suppressed in all cases.
Based on previous reports, there is no curative or consistently effective drug treatment for SUNCT syndrome. Therapy of acute attacks is not a useful concept as the attacks are very short in duration. Furthermore, abortive therapy may result in overmedication and toxicity. Therefore, preventive therapy is suggested to suppress attacks and to maintain remission over the expected duration of the headache period. The prophylactic pharmacological therapy works much more effectively in episodic SUNCT than in chronic patients, and success of the treatment depends on the pharmacological agent and its dose. Various classes of medication used in other headache syndromes have been tried for the treatment of SUNCT and have had poor effectiveness. These include NSAIDs (indomethacin, nimesulide), analgesics (paracetamol, opiates), 5-HT agonists (triptans, ergotamine), β-blockers (propranolol, timolol), α-adrenoreceptor agonist (clonidine), tricyclic antidepressants (amitriptyline, nortriptyline, desipramin), calcium channel blockers (nifedipine, flunarizine, diltiazem), antiepileptic drugs (phenytoin, valproic acid, clonazepam), lithium and baclofen (12,16,20,24–26).
In short-term prevention intravenous and subcutaneous lidocaine provided the patients some pain relief up to total pain abolition (10–13). Although cardiac adverse events did not occur, hospitalization and cardiac monitoring were necessary (13). Lamotrigine given 125–400 mg/day is the first-line recommendation for long-term prevention, and has a good effect in achieving complete remission or attack frequency reduction in most SUNCT patients (10,12–15,25). The main problem with lamotrigine is that the dose titration is very slow because of the possible skin reactions. In second-line prevention topiramate can be effective at doses of up to 300 mg/day as > 50% of patients had a good response to it (10,12,27–29). However, topiramate may cause severe side-effects and its administration is not suggested in patients with a history of renal stones, glaucoma, depression, or low body weight. Gabapentin can be effective also as a second-line agent in SUNCT in doses ranging from 800 to 3600 mg daily (10,12,26,30,31). Interestingly, this drug was more effective in patients with short-lasting unilateral neuralgiform headache attacks with cranial autonomic symptoms (SUNA) (60%) than in SUNCT (45%) (12). Carbamazepine has also been reported as having a good effect in SUNCT at doses of up to 900 mg/day (11), especially when used in combination with steroids (21,22) or topiramate (11).
Although there is no evidence-based recommendation for corticosteroid use in SUNCT syndrome, a few case reports are known where patients were treated by prednisone, prednisolone or methylprednisolone with moderate to excellent therapeutic response (20–22,32). Blockade of the greater occipital nerve with steroids and local anaesthetics were also tried in certain cases of SUNCT and were found to have moderate or good efficacy, but this treatment worked better in cluster headache (12,33,34). Conversely, corticosteroids are advised in cluster headache for short-term prevention (9,10,18,19); however, randomized, placebo-controlled trials are unavailable. The 21-day long oral prednisolone therapy is started with 1 mg/kg (maximum 60–100 mg) once daily for 5 days; thereafter the dose is decreased by 10 mg every day or every 3 days (17–19). With this method the risks of long-term steroid use (diabetes, hypertension) can be minimized. The mechanism of methylprednisolone action is unclear, although it is probably based on the anti-inflammatory effect of the drug. Methylprednisolone works via the glucocorticoid receptor by suppressing the expression of proinflammatory cytokines and also affecting the production of inflammatory mediators (35). It is known that in cluster headache the pain is a result of trigeminal nerve activation causing the release of vasodilative neurotransmitters such as substance P, serotonin (5-HT), nitric oxide, endothelins, calcitonin gene-related peptide (CGRP), vasoactive intestinal peptide (VIP) and interleukin (IL)-1, which is a cytokine acting in perivascular inflammation. It has been assumed that corticosteroids develop their influence by moderating the level of certain neurotransmitters (CGRP, VIP, histamine, 5-HT) and vasoactive inflammatory agents (IL-1), thereby decreasing the perivascular sterile inflammation (36–38).
In conclusion, orally administered methylprednisolone therapy proved effective in rapid and complete headache resolution in patients with episodic SUNCT syndrome. Symptoms did not relapse at the end of the expected headache cycle when the therapy was discontinued. Furthermore, intolerable side-effects were not manifested during the treatment periods. In case of longer headache periods, corticosteroid therapy would give time for introduction and titration of other preventive medication.