Positive changes from Baseline indicate an improvement (i.e., a reduction) in seizure frequency per week.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 4 were missing a baseline, 3 had a baseline of 0, and 1 was missing treatment period seizure data. The result was a sample size of 247. Of these 247 subjects 226 continued into the maintenance period.

Positive changes from Baseline indicate an improvement (i.e., a reduction) in seizure frequency per week.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 4 were missing a baseline, 2 had a baseline of 0, and 1 was missing treatment period seizure data. The result was a sample size of 248. Of these 248 subjects 227 continued into the maintenance period.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 1 was missing treatment period seizure data leaving a sample size of 254. Of these 254 subjects 233 continued into the maintenance period.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 1 was missing treatment period seizure data leaving a sample size of 254. Of these 254 subjects 233 continued into the maintenance period.

Positive changes from Baseline indicate an improvement (i.e., a reduction) in seizure frequency per week.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 4 were missing a baseline and 1 was missing treatment period seizure data. The result was a sample size of 250. Of these 250 subjects 229 continued into the maintenance period.

Positive changes from Baseline indicate an improvement (i.e., a reduction) in seizure frequency per week.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 4 were missing a baseline and 1 was missing treatment period seizure data. The result was a sample size of 250. Of these 250 subjects 229 continued into the maintenance period.

The responder rate is defined as the number of responders. A responder is a patient with a 50% or greater change (reduction) in partial seizure frequency per week.

Note: Rates were reported as percentages.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 4 were missing a baseline, 3 had a baseline of 0, and 1 was missing treatment period seizure data. The result was a sample size of 247. Of these 247 subjects 226 continued into the maintenance period.

For subjects with up to 24 weeks in the evaluation phase the denominator for each subject is their number of days in the evaluation phase.

Intention to Treat (ITT) subjects with <= 24 Weeks of Exposure and treatment period seizure data

For subjects with greater than 24 weeks in the evaluation phase the denominator for each subject is their number of days in the evaluation phase.

Intention to Treat (ITT) Subjects with > 24 Weeks of Exposure and treatment period seizure data

For subjects with up to 24 weeks in the evaluation phase the denominator for each subject is their number of days in the evaluation phase.

Intention to Treat (ITT) Subjects with <= 24 Weeks of Exposure and treatment period seizure data

For subjects with greater than 24 weeks in the evaluation phase the denominator for each subject is their number of days in the evaluation phase.

Intention to Treat (ITT) subjects with > 24 weeks of exposure and treatment period seizure data

The measure description is the product limit adjusted percent of subjects seizure free starting from the beginning of the Maintenance Period.

The up-titration period is the up to 6 week period of increasing dose prior to the Maintenance Period. The Maintenance Period is the period of stable dosing, subsquent to the up-titration period, which could last from 42 to 48 weeks.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 1 was missing treatment period seizure data leaving a sample size of 254. Of these 254 subjects 233 continued into the maintenance period.

Type I Seizure is a partial onset Seizure (see International League Against Epilepsy definitions).

Type II Seizure is a Generalized Seizure (see International League Against Epilepsy definitions).

Type III Seizure is a Unknown Seizure Type (see International League Against Epilepsy definitions).

A subject could experience more than one seizure type.

255 subjects were Intention to Treat (ITT), i.e. all subjects with at least 1 dose of study medication. Of these 255 ITT subjects 1 was missing treatment period seizure data.

There are 7 categories, 3 for improvement (Marked improvement, Moderate improvement, Slight improvement), 3 for worsening (Slight worsening, Moderate worsening, Marked worsening), and 1 for no change (No change).

Intention to Treat (ITT) subjects for whom the assessment was performed

There are 7 categories, 3 for improvement (Marked improvement, Moderate improvement, Slight improvement), 3 for worsening (Slight worsening, Moderate worsening, Marked worsening), and 1 for no change (No change).

Intention to Treat (ITT) subjects for whom the assessment was performed

There are 7 categories, 3 for improvement (Marked improvement, Moderate improvement, Slight improvement), 3 for worsening (Slight worsening, Moderate worsening, Marked worsening), and 1 for no change (No change).

Intention to Treat (ITT) subjects >= 8 years old for whom the assessment was performed

The Leiter-R AM battery has 10 subtests. The raw scores of the subtests are converted into scaled scores. Six composite scores are constructed from the 10 subtest scaled scores. The Memory Screen is one of them. It is composed of 2 subtests the Associated Pairs and Forward Memory. The sum of the Associated Pairs and Forward Memory subtest scaled scores are converted into a Memory composite score normally distributed with a mean and standard deviation of 100 (±15). Higher scores and positive changes from baseline are better. The range of the Memory Screen composite score is 44 to 155.

At baseline there were 98 subjects with valid Memory Screen composite scores (mean=85.5, standard deviation=18.7). Of these 98 subjects, 87 had valid scores at Visit 5 (week 24) and 80 at Visit 7 (week 48).

This score is obtained from a total raw score which is the sum of a battery of individual questions. It is adjusted for a child's age, has an expected mean of 100 and standard deviation of 15, and can be categorized as: (1) Accelerated Performance (>= 115), (2) Within Normal Limits (85-114), (3) Mildly Delayed Performance (70-84), and (4) Significantly Delayed Performance (<=69). Changes from baseline are then further categorized where 'Improved' is any positive category change, 'Stable' is no category change, and 'Worsened' is any negative category change, from baseline.

The Bayley Scale of Infant Development (BSID) II assessment was performed only at selected sites where a neuropsychologist was available. Both a valid baseline and Visit 5 (week 24) assessment had to be present.

This score is obtained from a total raw score which is the sum of a battery of individual questions. It is adjusted for a child's age, has an expected mean of 100 and standard deviation of 15, and can be categorized as: (1) Accelerated Performance (>= 115), (2) Within Normal Limits (85-114), (3) Mildly Delayed Performance (70-84), and (4) Significantly Delayed Performance (<=69). Changes from baseline are then further categorized where 'Improved' is any positive category change, 'Stable' is no category change, and 'Worsened' is any negative category change, from baseline.

The Bayley Scale of Infant Development (BSID) II assessment was performed only at selected sites where a neuropsychologist was available. Both a valid baseline and Visit 7 (week 48) assessment had to be present.

This score is obtained from a total raw score which is the sum of a battery of individual questions. It is adjusted for a child's age, has an expected mean of 100 and standard deviation of 15, and can be categorized as: (1) Accelerated Performance (>= 115), (2) Within Normal Limits (85-114), (3) Mildly Delayed Performance (70-84), and (4) Significantly Delayed Performance (<=69). Changes from baseline are then further categorized where 'Improved' is any positive category change, 'Stable' is no category change, and 'Worsened' is any negative category change, from baseline.

The Bayley Scale of Infant Development (BSID) II assessment was performed only at selected sites where a neuropsychologist was available. Both a valid baseline and Visit 5 (week 24) assessment had to be present.

This score is obtained from a total raw score which is the sum of a battery of individual questions. It is adjusted for a child's age, has an expected mean of 100 and standard deviation of 15, and can be categorized as: (1) Accelerated Performance (>= 115), (2) Within Normal Limits (85-114), (3) Mildly Delayed Performance (70-84), and (4) Significantly Delayed Performance (<=69). Changes from baseline are then further categorized where 'Improved' is any positive category change, 'Stable' is no category change, and 'Worsened' is any negative category change, from baseline.

The Bayley Scale of Infant Development (BSID) II assessment was performed only at selected sites where a neuropsychologist was available. Both a valid baseline and Visit 7 (week 48) assessment had to be present.