Main Entry Criteria. Children ≥6 years old with newly diagnosed epilepsy

Comparator. Topiramate (100 and 200 mg/day), carbamazepine (600 mg/day), and valproate (1,250 mg/day)

Number of Patients. 613

Primary Outcome Variable. Time to exit, time to first seizure, proportion of patients who are seizure free at 6 months

Results. There was no difference in treatments; fewer patients on topiramate 100 mg/day withdrew because of adverse events.

Main Entry Criteria. Children aged 3 to 16 years with at least two previously untreated generalized tonic-clonic or partial seizures

Comparator. Carbamazepine, phenobarbitone, phenytoin, or valproate

Number of Patients. 167

Primary Outcome Variable. Time to first seizure and time to achieve 1-year remission

Results. No significant differences were observed between treatment groups for all primary outcome measures. Phenobarbital arm was discontinued because of 60% dropout
rate in the first ten patients enrolled. Children on phenytoin were more likely to drop out compared to those on carbamazepine and valproate.

Main Entry Criteria. Children with two or more generalized tonic-clonic or partial seizures in the previous 6 months

Comparator. Carbamazepine or valproate

Number of Patients. 260

Primary Outcome Variable. Seizure counts

Results. No significant differences were observed between treatment groups or for different seizure types.

Main Entry Criteria. Patients with epilepsy diagnosed in previous 3 months

Comparator. Topiramate (100 or 200 mg/day) or investigator’s choice of carbamazepine (600 mg/day) or valproate (1,250 mg/day)

Number of Patients. 631

Primary Outcome Variable. Time to exit, time to first seizure, proportion of seizure-free patients in last 6 months of therapy

Results. No differences in efficacy measures were observed between treatment groups. Topiramate 100 mg/day had fewest withdrawals because of adverse events

Main Entry Criteria. Patients older than 2 years and those newly diagnosed with partial epilepsy

Comparator. Carbamazepine or lamotrigine

Number of Patients. 618

Primary Outcome Variable. Proportion of patients who are seizure free during last 16 weeks of the study

Results. No difference was observed in efficacy, and fewer patients on lamotrigine withdrew because of adverse events.

Main Entry Criteria. Patients with newly diagnosed epilepsy

Comparator. Carbamazepine or lamotrigine

Number of Patients. 260

Primary Outcome Variable. Side Effect and Life Satisfaction (SEALS) inventory

Results. Patients on carbamazepine had significantly lower scores in the SEALS inventory. Fewer patients on carbamazepine completed the study.

Main Entry Criteria. Elderly patients with newly diagnosed epilepsy (mean age 77 years)

Comparator. Carbamazepine or lamotrigine

Number of Patients. 150

Primary Outcome Variable. Time to first seizure, being seizure free during the last 16 weeks of study

Results. There was no difference in time to first seizure. A significantly higher percentage of patients on lamotrigine remained seizure free. Significantly more patients on lamotrigine completed the study. Patients on carbamazepine were twice as likely to withdraw from the study.

Main Entry Criteria. Patients with newly diagnosed and previously untreated partial seizures

Comparator. Carbamazepine (600 mg/day) or vigabatrin (2 g/day)

Number of Patients. 459

Primary Outcome Variable. Time to withdrawal due to seizures or adverse effects, time to 6-month remission of seizures, time to first seizure

Results. There was no difference in time to withdrawal. Time to first seizure was significantly longer in patients on carbamazepine. There was no significant difference in time to achieve 6 months of being seizure free.

Main Entry Criteria. Patients with newly diagnosed epilepsy

Comparator. Carbamazepine or lamotrigine

Number of Patients. 260

Primary Outcome Variable. Seizure counts, number of patients who are seizure free during last 24 weeks of treatment

Results. There were no differences in any measure of efficacy for any seizure type. Fewer patients on lamotrigine withdrew because of adverse events.

Main Entry Criteria. Patients >16 years old referred to an epilepsy treatment program with at least two previously untreated tonic-clonic or partial seizures

Comparator. Carbamazepine, phenytoin, phenobarbitone, or valproate

Number of Patients. 243

Primary Outcome Variable. Time to first seizure and time to enter 1 year of remission

Results. There were no significant differences between treatment groups for any measure of efficacy. Overall, 27% remained seizure free after starting a drug and 75% had at least 1 year of remission. Phenobarbital was more likely to be withdrawn than the other drugs.

Main Entry Criteria. Patients diagnosed with complex partial or secondarily generalized tonic-clonic seizures

Comparator. Carbamazepine or valproate

Number of Patients. 480

Primary Outcome Variable. Number of seizures, time to first seizure, seizure-rating score, composite score that combined seizure control with adverse events

Results. There was no difference between carbamazepine and valproate in secondarily generalized tonic-clonic seizures. For complex partial seizure, carbamazepine was significantly better in reducing the total number of seizures and number of seizures per month. Time to first seizure was significantly lengthened with carbamazepine. The seizure-rating score and composite score was significantly improved with carbamazepine.

Main Entry Criteria. Patients with partial and secondarily generalized tonic-clonic seizures

Comparator. Carbamazepine, phenobarbital, phenytoin, or primidone

Number of Patients. 622

Primary Outcome Variable. Seizure counts, retention in study

Results. In the study, the group on carbamazepine and phenytoin retained significantly higher number of patients. There were significantly greater number of seizure-free patients on carbamazepine compared to those on phenobarbital or primidone

Main Entry Criteria. Patients with untreated generalized tonic-clonic or partial seizures

Comparator. Carbamazepine (400 mg/day) or carbamazepine (200 mg/day) combined with valproate (300 mg/day)

Number of Patients. 130

Primary Outcome Variable. Seizure counts, clinimetric epilepsy scales, neuropsychological tests

Results. There were no differences between the treatment groups.

Main Entry Criteria. Patients with more than one seizure per week previously treated with phenytoin, phenobarbital, and/or primidone

Comparator. Carbamazepine or placebo added to current medications in a crossover design

Number of Patients. 23

Primary Outcome Variable. Seizure counts and adverse effects

Results. 12 patients had a ≥50% reduction in seizure frequency with carbamazepine; none of the patients on placebo had a reduction in seizure frequency.

Main Entry Criteria. Patients with peripheral neuropathic pain reduced by spinal cord stimulation

Comparator. Phase I consisted of carbamazepine (600 mg/day) or placebo, phase II consisted of sustained-release morphine (90 mg/day) or placebo.

Number of Patients. 43

Primary Outcome Variable. Numeric analog scale rating pain

Results. There is a significant delay in pain increase with carbamazepine compared to that with the placebo. No difference in pain increase is seen with sustained-release morphine.

Main Entry Criteria. Patients with severe diabetic peripheral neuropathy

Comparator. Carbamazepine or the combination of nortriptyline and fluphenazine

Number of Patients. 16

Primary Outcome Variable. Visual analog scale

Results. Both treatment groups had significant improvements in pain measures over baseline. There was no significant difference in response between treatment groups. Adverse effects were more frequent in patients receiving nortriptyline and fluphenazine.

Main Entry Criteria. Bipolar patients with manic or mixed episodes

Comparator. Extended-release carbamazepine or placebo

Number of Patients. 204

Primary Outcome Variable. Young Mania Rating Scale, Clinical Global Impressions, Hamilton Rating Scale for Depression (HAM-D)

Results. The mean carbamazepine dose was 756.44 mg; significantly greater responder rate (≥50% decrease in Young Mania Rating Scale) was observed in patients treated with carbamazepine; patients with mixed episodes showed significant improvements in HAM-D score at 21 days. The most common adverse events were dizziness, nausea, and somnolence.

Main Entry Criteria. Patients with at least two episodes of bipolar disorder in the previous 3 years who were in remission at entry to the study; none of the patients had received lithium or carbamazepine therapy for longer than 6 months in their lifetime

Comparator. Carbamazepine or lithium

Number of Patients. 94

Primary Outcome Variable. Recurrence of a bipolar episode

Results. Lithium was more effective in patients with a previous untreated hypomanic episode and in all patients with a prior hypomanic episode without manic episodes. More patients on lithium dropped out of the study.

Main Entry Criteria. Patients meeting the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for bipolar disorder