Carbamazepine was used in migraine prophylaxis based on its efficacy against trigeminal neuralgia. One placebo-controlled, randomized, double-blind, crossover trial suggested a significant benefit: either marked or complete improvement was reported by 26 of 45 (58%) patients on carbamazepine and by 5 of 48 (10%) on placebo. However, this trial was inadequately described in several important respects (46). Another trial, comparing carbamazepine with clonidine and pindolol, suggested that carbamazepine had a weaker effect on headache frequency than either comparator treatment, although differences from clonidine were not statistically significant (2). Significantly more patients reported adverse events (AEs) with carbamazepine than with placebo or pindolol; there was no significant difference in this respect between carbamazepine and clonidine.

Clonazepam was studied in one placebo-controlled crossover trial (58) of 34 patients. Those completing 4 weeks’ treatment (1 mg daily) had their mean headache days per month reduced by 50% compared with an 8% reduction for patients receiving placebo (p <0.05). The effect was less at 2 mg daily. Drowsiness was a problem.

Gabapentin’s mode of action in migraine is unclear (66). It interacts with the α₂δ-subunit of the calcium channel and increases the concentration and probably the synthesis of brain γ-aminobutyric acid (GABA). Gabapentin binds to gabapentin-binding protein—a novel, membraneassociated protein in the outer layers of the cerebral cortex (61). It penetrates the blood-brain barrier but does not interact with GABA receptors (17). Gabapentin (600 to 1800 mg) was effective in both episodic and chronic migraine in a 12-week open-label study (35). Gabapentin was not effective in one placebo-controlled double-blind study (65). In a second randomized, placebo-controlled, double-blind trial (36), gabapentin 1800 to 2400 mg was superior to placebo in reducing the frequency of migraine attacks. The responder rate was 36% for gabapentin and 14% for placebo (p = 0.02). The most common AEs were dizziness or giddiness and drowsiness. Some trials reported relatively high patient withdrawal rates due to AEs associated with gabapentin (18).

Lamotrigine blocks voltage-sensitive sodium channels, inhibiting neuronal release of glutamate (31,32,64), essential to the propagation of spreading depression (49). In an open study, 10 patients with migraine with aura responded to lamotrigine. Lamotrigine as combination therapy was studied in one prospective, open-label trial of 65 patients, most of whom had chronic migraine (67). Only 35 patients were sufficiently compliant with treatment to warrant inclusion in the analysis; 12 dropped out because of AEs. The primary endpoint was reduction in severe headache frequency; 17 (48.6%) responded, at a mean dose of 55 mg/day. Those with migraine with aura had a better response rate (12/18 or 67%), including 4 of 8 whose headaches were chronic. Another open-label study found that lamotrigine significantly reduced both the frequency and duration of aura (67).

Chen et al. (5) reported two patients with migraine with persistent auralike visual phenomena for months to years. After lamotrigine treatment for 2 weeks, both had resolution of the visual symptoms.

Steiner et al. (57) compared lamotrigine to placebo in a double-blind, randomized, parallel-group migraine prophylaxis trial. Lamotrigine was initially begun at the full dose of 200 mg/day, but, following a high incidence of skin rashes, a slow dose-escalation was introduced: 25 mg/day for 2 weeks, 50 mg/day for 2 weeks, and then 200 mg/day. Attack rates were reduced from baseline means of 3.6 per month on lamotrigine and 4.4 on placebo to 3.2 and 3.0, respectively, during the last month of treatment. In this study lamotrigine was ineffective for migraine prophylaxis. There were more AEs on lamotrigine than on placebo, most commonly rash.

Tiagabine was effective (14) in an open-label clinical trial of 41 patients who had been previously treated with divalproex sodium and who discontinued for AEs or lack of efficacy. Tiagabine was started at 4 mg at bedtime for 1 week and then increased to 4 mg twice a day. Five patients experienced a remission, and 33 of 41 patients had at least a 50% reduction in their attacks. The mean dose of tiagabine was 10 mg/day. Fourteen AEs were reported by 12 patients. No placebo-controlled, double-blind trials are available.

Topiramate is a structurally unique anticonvulsant that is a derivative of the naturally occurring monosaccharide D-fructose and contains a sulfamate functionality. Topiramate is rapidly and almost completely absorbed. The blood plasma concentration increases linearly as a function of dose over the pharmacologically relevant range (13,60). It is not extensively metabolized and is eliminated predominantly unchanged in the urine. The average elimination half-life is approximately 21 hours (13). Topiramate readily enters the central nervous system parenchyma; in rats, the concentration in whole brain was approximately one-third that in blood plasma 1 hour after oral dosing.

Topiramate influences the activity of some types of voltage-activated Na⁺ and Ca²⁺ channels, GABA_A receptors, and the α-amino-3-hydroxy-5-methylisoxazole-4-proprionic acid (AMPA)/kainate subtype of glutamate receptors. Topiramate also inhibits some isozymes of carbonic anhydrase (CA) and exhibits selectivity for CA II and CA IV (11,50).

The effects of topiramate on voltage-activated NA⁺ channels, voltage-activated calcium channels, GABA_A receptors, and AMPA/kainate receptors are unique. They are all regulated by protein phosphorylation (29,45,51,64). One or more subunit of each complex is phosphorylated by protein kinase A, protein kinase C, and possibly CA²⁺/CaM-activated kinases. Topiramate may bind to the membrane channel complexes at phosphorylation sites in the inner loop and thereby allosterically modulate ionic conductance through the channels.

Storer and Goadsby (59) found that topiramate inhibited the activation of trigeminocervical neurons in response to stimulation of the superior sagittal sinus. Its inhibition is a plausible mechanism of the action of migraine or cluster headache preventive medicines.