Five RCTs, two small pilot studies and six non-comparative studies were used to assess efficacy. The number of patients per intervention group is reported in the 'Results' section.
Only the results for efficacy and adverse events are reported here. The review also included information on the pharmacodynamic and pharmacokinetic properties of topiramate.
Children with partial seizures: there was one RCT involving 86 children aged 2 to 16 years with six or more partial seizures over the 8-week baseline period; treatment duration was 16 weeks. Topiramate was initiated at 25 mg/day and increased to the target dose of 125 to 400 mg/day over 8 weeks. Compared with placebo, topiramate significantly reduced the average monthly partial seizure frequency; the reductions were 33.1% and 10.5% with topiramate and placebo, respectively (P=0.034). It also reduced secondarily generalised seizures (40 children, 20 in each treatment group); the reductions were 31.6% and 10.6% with topiramate and placebo, respectively (P not reported in the study). Topiramate significantly increased the proportion of global assessments reported as marked improvement in comparison with placebo: 29% versus 11% (P=0.019).
Children and adults with generalised tonic-clonic seizures: there were two RCTs involving 160 patients with at least three poorly controlled, primary generalised tonic clonic seizures over the 8-week baseline period; treatment duration was 20 weeks. The topiramate dose was titrated over 8 weeks to a target dose, which ranged from 5.2 to 9.3 mg/kg per day according to defined weight ranges of 25 to 43 kg or more. The results differed. One RCT (80 patients, drop-out rate 11%) found that topiramate significantly reduced seizure frequency in comparison with placebo (56.7% reduction versus 9%, P=0.019), but found no significant difference in global assessment of improvement (55% versus 56%, P=0.09). The other RCT (80 patients, drop-out rate 25%, imbalance in baseline seizure frequency) found no significant difference in reduction in seizure frequency between the treatments (57.1% versus 32.2%, P=0.124), but found that topiramate significantly increased global assessment of improvement (48% versus 33%, P=0.026).
A post-hoc analysis of the 31 children aged 16 years or younger found a seizure reduction of at least 50% in 47% of those with topiramate versus 21% of those with placebo (P not reported in the study).
Lennox-Gastout Syndrome: there was one RCT involving 98 adults and children (mean age 11.2 years; age range: 2 to 42); treatment duration 11 weeks. The median topiramate dose was 5.8 mg/kg per day during the 8-week maintenance period. Topiramate significantly reduced drop attacks in comparison with placebo; the reduction was 14.8% with topiramate versus 5.1% with placebo (P=0.041). Topiramate significantly improved global parental evaluation of seizure severity (P=0.037).
Children with the West Syndrome: there were two small pilot studies involving 21 children aged 3 to 25 months. One study (11 children) in children who had failed at least two courses of conventional therapy found that spasm frequency was reduced by 50% or more in 9 of the 11 children and that 5 children became spasm free with no electroencephalogram (EEG) evidence of typical hypsarrhythmia. The other study (10 newly diagnosed children) found that seizure frequency was reduced by 50% or more in 5 of the 10 children and that one child became spasm free, but no EEG changes were seen.
Children and adults with newly diagnosed epilepsy of all types: there was one RCT of 613 patients (age range: 6 to at least 64 years) that compared topiramate (100 and 200 mg) with either carbamazepine (600 mg) or valproic acid (1,250 mg). There was no significant difference between the three treatments in efficacy or tolerability (no details were presented in the review). A subgroup analysis (119 children aged 6 to 16 years) found that the proportion of children who were seizure free after at least 6 months of treatment was 63% with topiramate 100 mg, 59% with topiramate 200 mg, 39% with carbamazepine, and 53% with valproic acid (P was not reported). The distribution of seizure type was uneven between the treatment groups.
Non-comparative studies: there were six studies, each with at least 30 children; the mean treatment duration ranged from 6 to 14.8 months. The response rates with topiramate (2 to 11 mg/kg per day) were variable, but were similar to the rates in controlled trials. The response ranged from 25 to 81% in children with partial seizures, and from 0 to 82% in those with generalised seizures.
Adverse events: the withdrawal rate due to adverse events was 4.8% (8 out of 106 children) in four RCTs; higher rates were reported in non-comparative studies. The withdrawal rates reported in the latter were 41.4%, 39%, 11%, 7.2% and 4.5%. Seven different neuropsychiatric adverse events and weight loss (199 children) were more common (greater than 5%) with topiramate than with placebo. The most frequent adverse effects reported were weight loss, memory, aggressive behaviour, concentration or attention problems, nervousness, fatigue, anorexia and somnolence.