Abstracts

RATIONALE: Vigabatrin (VGB), a specific and irreversible inhibitor of GABA-transaminase, has been shown to raise cerebral GABA levels in animals and humans. It has been proven an effective treatment for infantile spasms (IS), especially in patients with tuberous sclerosis (TS) who show typically rapid responses with rates up to 90%. Prompted by delayed responses in some of our patients, we asked if concomitant anti-convulsant medication had any effect. METHODS: Records of all the patients with TS and IS enrolled in a Phase II clinical trial for the use of VGB in IS were reviewed with respect to presence of concomitant medication, response to VGB and latency of this response. RESULTS: Ten patients were identified, all of whom eventually responded with cessation of infantile spasms. Of the four on VGB monotherapy, three responded within five days, but the fourth took four months. Of the six who entered the trial on phenobarbital (PB) one responded in the first week, the others taking between two weeks and seven months. Median time to response was three days in the group not on PB and 34 days in the group on PB. Three of these patients (50%) did not respond until after the PB was withdrawn. CONCLUSIONS: This is an important clinical observation for those treating IS with VGB as many of these patients are likely to have received PB as initial therapy. Our data suggest that PB interfered with the efficacy of VGB. If PB was not withdrawn in three of our patients our response rate may have been only 50% instead of the 100% remission seen in our patients. The mechanism of this interference is unknown, but may involve a previously described block of the Cl- channel in the presence of PB at GABA A receptor.