A Newsletter from the Cornell Headache Service : Fall 1997, Vol. 1, Issue 1
A Newsletter from the Cornell Headache Service : Fall
1997, Vol. 1, Issue 1
The "Migraine Generator" | |
|
|
This is a PET study in a patient with a spontaneous migraine. There is an area of increased cerebral blood flow in the rostral brainstem which persists even after the headache is successfully treated with sumatriptan. Could this be the region of the brain that initiates a migraine? Weiller C, et al. Brainstem activation in spontaneous human migraine attacks. Nature Medicine 1995; 1:658-660.
|
Divalproex sodium (Depakote) has emerged as a first line medication in headache treatment. Silberstein reviews the clinical experience of valproate and offers sensible clinical guidelines for it's use. To date 4 double-blind placebo controlled studies (n= 238) have confirmed the efficacy of valproate as prophylaxis in migraine. Other headache disorders such as chronic daily headache and cluster headache have shown a lesser response. The consensus seems to be that at least half the migraine subjects treated enjoy a 50% reduction in headache frequency. Furthermore headache intensity and duration is also diminished. Efficacy correlated to serum valproate levels when it was studied but no specific conclusions could be drawn about dose and pain relief. Adverse effects were the same as those experienced in the epilepsy population and included anorexia, nausea, vomiting, dyspepsia, asymptomatic serum hepatic transaminase elevations, tremor, sedation, weight gain, alopecia and rash.
Recommendations for treating migraine with divalproex are similar to those used in epilepsy with some exceptions:
1) The suggested starting dose is 250mg and increased slowly
to 500-750mg/day in divided amounts but ultimate doses of 2000mg/day were reported.
2) Strict adherence to "therapeutic" levels (50-100 ug/mL) is not needed.
3) Follow-up valproate levels to measure compliance, toxicity, and drug interactions should be done only when needed.
4) Obtain an initial CBC and liver function profile but routine monitoring is not necessary in healthy, asymptomatic patients.
5) In the event of a mild transaminitis reduce the valproate dose. However, for a transaminitis 2-3x normal stop and restart valproate when the LFTs return to normal.
6) Do not use valproate for migraine in children under 10 or pregnant women with headache.
7) Silberstein anecdotally recommends zinc and multivitamins to control valproate associated hair loss.
Cutrer and Moskowitz describe a specific GABAergic mechanism by which valproate might exert its antimigraine effect. In the Moskowitz model trigeminal pain can be studied in guinea pigs using an intracisternal injection of capsaicin to activate trigeminal c-fiber afferents. Quantification of the "pain response" can then be determined by measuring the expression of the c-fos gene/ protein in the trigeminal nucleus caudalis. In the animals pretreated with valproate there was a 52% reduction in c-fos protein expression (p<.05). Or in other words a 52% reduction in trigeminal or "pain" activation.
This reduced response could be blocked by bicuculline (GABAA antagonist) but not by phaclofen (GABAB antagonist) demonstrating the importance of the GABAA receptor. They also observed 42% reduced c-fos expression in subjects pretreated with allopregnanolone, a progesterone metabolite modulating GABAA (p<.01).
In an earlier study (Limmroth et al.Cephalalgia 1995;15 (suppl 14): 102) Moskowitz showed that valproate can block neurogenic inflammation within the meninges via GABAA receptors as well. Although GABA receptors in brainstem and cortex may also be implicated, valproate seems to exert it's effect on migraine through modulation of the of GABAA receptors in the peripheral meningeal trigeminal vascular system. As with the 5-HT receptors this finding suggests new pharmacological strategies for migraine based upon modulation of the GABAA receptor.
COMMENTS:
Valproate has been a valuable addition in headache treatment. Already it enjoys wide applications as an antiepileptic drug for multiple seizure types and for a range of psychiatric conditions including mania, and schizoaffective disorder. As such in headache patients with comorbid epilepsy, mania or anxiety disorders it should be considered as the initial prophylactic choice.
Silberstein S, Divalproex sodium in headache: literature review and clinical guidelines. Headache 1996;36:547-555.
Cutrer FM, and Moskowitz MA, The action of valproate and neurosteroids in a model of trigeminal pain. Headache 1996; 36:579-585.
The overall incidence of trigeminal neuralgia is 4-5/100,000. The inial treatment usually begins with medication including carbamazepine, phenytoin, baclofen. Frequently surgery is considered for refractory cases. There are currently 3 percutaneous procedures available: radiofrequency thermal rhizotomy, retrogasserian glycerol injections, trigeminal-ganglion balloon micro compression. These techniques cause controlled injury to the trigeminal nerve and/or ganglion and presumably inactivate the pain mechanism. Micro-vascular decompression is often considered a procedure of last resort. Pain is alleviated by relieving vascular compression on the nerve or ganglion via a small retromastoid craniectomy. A new report by Young and colleagues heralds a new radiosurgical technique for idiopathic and tumor related trigeminal pain.
The use of gamma radiosurgery for trigeminal neuralgia began with Lars Leksell in 1953 (Acta Chir Scand 137:311-314, 1971). In that initial report 2 patients with up to 18 years follow up were successfully treated by external radiation directed to the trigeminal ganglia. In an attempt to reevaluate their limited success Kondziolka and colleagues (J Neurosurg 1996;84:940-945) initiated a multicenter study using stereotactic high-resolution MRI to better define radiosurgical technique, dose and efficacy-safety. Fifty patients at 5 centers underwent radiotreatment aimed at the trigeminal root entry zone and proximal nerve. The dose ranged from 60-90 Gy. The median follow up was 18 months (range 11-36 months). Twenty eight patients (56%) were pain free, 16 (32%) obtained good control (>50-90% relief) and 6 (12%) were treatment failures. The minimally effective radiosurgical dose was determined to be 70 Gy. Only 3 (6%) developed post radiosurgical facial parathesia.
Young and colleagues participated in this multicenter study and report their own uniform experience. They studied 60 patients with a mean age of 64.3 years. Forty nine patients had idiopathic trigeminal neuralgia, 9 had skull based tumors, 2 were diagnosed with multiple sclerosis and all were deemed medically intractable. Twenty two patients had previously undergone single or multiple surgical treatments. The median follow up was 16.3 months (range 6-36 months). Forty three of the 51 (72%) patients who underwent 70 Gy of trigeminal root entry zone directed radiotreatment had complete relief. Six of the 51 (10%) had a good response defined as >50-90% reduction in pain. The mean latency to pain relief was 14 days. Eleven of the 51 (18%) were treatment failures. Seven of these 11 (64%) treatment failures had previous trigeminal neuralgia surgery. Looked at another way, of the 22 patients with prior surgery 7 (32%) failed radiosurgery compared with only 3 of 29 (10%) failures in the patients without prior intervention. The 9 patients with skull based tumors had radiotreatment specifically designed to treat the tumor. Eight of 9 (88%) had complete pain relief. Taken together 49 (82%) of the 60 total patients with trigeminal neuralgia treated with radiosurgery experienced complete or nearly complete pain relief. Almost no adverse effect were reported. In fact only a single patient with a Schwannoma and preexisting facial sensory loss had a mild increase in facial numbness.
COMMENT:
This report by Young confirms the earlier Kondziolka study establishing gamma radiosurgery as safe and effective treatment for trigeminal neuralgia. It is worth stressing that better outcomes were seen in the group naive to other surgical treatments. However as Maciewicz and Scriviani state in the follow up editorial (Neurology 1997;48:565-566) the report lacks rigorous design. They point out that the current study fails to systematically document pain and affective response before and after treatment, randomize patients to different treatment groups and blind study participants. Longer term follow up is also needed to confidently assess the full efficacy and complication profile.
Surgical studies of other techniques also lack in rigorous study design. But despite the shortcomings one can still compare outcomes. Baker, Jannetta and colleagues (NEJM 1996; 334:1077-1083) reviewed their 20 year experience in 1185 patients who underwent microvascular decompression. The median follow-up was 6.2 years. At 10 years post-op 70% of patients were pain free without medicine and additional 4% had occasional pain requiring short term medication. Predictors of symptom recurrence included female sex, symptoms for greater than 8 years, venous compression of the trigeminal-root entry zone, and lack of immediate post-operative response. Previous trigeminal ablative procedure did not lessen a patients likelihood of response. Major complications were few and included death 0.2%, brainstem stroke 0.1%, ipsilateral hearing loss 1.0% and facial numbness 1.7%. Total complication rate including post-op chemical meningitis was 24.9%.
Taha and colleagues (J Neurosurg 1995;83:989-993) report 75% success rate in a prospective 15 year follow up of 154 patients treated with percutaneous stereotactic thermal rhizotomy. The response correlated to the severity of sensory loss affected in the region. Recurrence rate was 60% in patients with mild hypalgesia, 25% in those with dense hypalgesia and 20% in patients with analgesia as the target lesion. The facial sensory loss was only described as dysesthetic in 23% of patients. Other adverse effects included absent or depressed corneal reflex in 29 patients (18.8%) and subsequent keratitis 3 patients (0.65%). Trigeminal motor weakness occurred in 22 patients (14.3%) but resolved by 1 year in 19 of them.
Fujimaki and colleagues (J Neurosurg 1990; 73:212-216) studied 122 patients treated with percutaneous retrogasserian glycerol injection. Eighty patients were available for follow-up from 38 to 54 months and only 21 patients (26%) maintained lasting pain relief. Complications reported were facial hypesthesia in 50 patients (63%) with 23 (29%) describing unpleasant facial dysesthesia. Similar results for glycerol injection were reported by Burchiel (J Neurosurg 1988;69:361-366). He reported a 50% recurrence rate by 18 months.
Brown and colleagues (Neurosurgery 1993; 32:570-573) report on the results of 50 patients treated by trigeminal nerve compression. Median follow up was 3 years. Recurrence rate was seen in 13 patients (26%). Complications were minor and included mild facial numbness in 35 patients (70%) and facial dysesthesia in 10 patients (20%).
Until more rigorous study designs are employed comparisons between techniques will be relative. Nonetheless the evidence for gamma knife is still compelling. With respect to efficacy and safety radiosurgery emerges as a powerful treatment option. Even at a greater expense costing 3.5x more than the percutaneous procedures but about 50% less than microvascular decompression it may prove to be overall more cost effective. This may also be the case when compared to long term medication treatment.
Young MD et al. Gamma knife radiosurgery for treatment of trigeminal neuralgia, idiopathic and tumor related. Neurology 1997; 48: 608-614.
|
Technique |
Patients |
Follow-up |
Recurrence Rate |
Complications |
|
Microvascular decompression |
(1) 1185 |
20 year retrospective (median 6.2 years) |
(30% at 10 years, 24.9% total) |
death 0.2%, brainstem stroke 0.1%, hearing loss 1.0%, facial numbness 1.7%, meningitis 17.1% |
|
Percutaneous thermal rhizotomy |
(2) 154 |
15 year prospective |
(25% at 14 years (20-60% range*)) |
facial dysethesia 23%, jaw weakness 14.3%, corneal keratitis 0.65% |
|
Percutaneous retrogasserian glycerol |
(3) 22 |
3.2 - 4.5 years |
(74%) |
facial dysesthesia 29% |
|
Trigeminal nerve compression |
(4) 50 |
3 years median 9 months - 7.5 years |
(26%) |
facial dysesthesia 20% |
|
* recurrence rate correlated to degree of post-op hypalgesia: 60% in patients with mild; 25% in patients with dense, 20% in patients with analgesia. | ||||
|
1. Barker et al. NEJM 1996; 334: 1077-83. 2.Taha et al. J Neurosurg 1995; 83: 989-93. 3.Fujimaki et al. J Neurosurg 1990; 73: 212-16. 4.Brown et al. Neurosurgery 1993; 32: 570- | ||||
Cornell
Headache Diet Beverages
Limited caffeine sources to 8
oz/day: coffee, tea, cola type soda, chocolate milk
Alcoholic drinks
Ale, beer, Chianti, sherry,
Burgundy, vermouth
Met & Fish
Aged, cured, processed meats
containing poultry, nitrates: ham, hot dogs, salami,
sausage, bologna. Pickled herring, dried fish
Dairy
Aged, mature cheese: blue,
Boursalt, Brie types, Camembert types, cheddar, Swiss,
Roquefort, Stilton, parmesan, provolone, Romano
Bread & Cereals
Sourdough bread
Vegetables
Broad beans, Italian beans, lima,
lentils, snow peas, fava, soy beans
Fruit
Avocado, banana, citrus, figs,
raisins, papaya, passion fruit, red plums,
raspberries
Soups
Canned soups, bouillon cubes, soup
bases with autolyzed yeast or monosodium glutamate
(MSG)
Sweets
Chocolate, licorice,
molasses
Misc.
MSG, soy sauce, meat tenderizers,
Accent Yeast, yeast extracts, Brewers yeast; Nuts and seeds:
peanuts, sesame, pumpkin; Pickeled foods; Preserved
foods
Foods to Avoid
(Generally dietary
triggers are overstated, but this is a list of the most
common culprits)
|
The Headache is a quarterly publication from the Cornell Headache Service The New York Hospital-Cornell Medical Center Cornell Headache Service Editor: Designer: This issue was funded by an educational grant from Abbott Laboratories. |
|
