Objective

Method

Quality assessment

Participants

2

RCT, cross‐over single blind, participants blinded.

Each arm of at least 1 month duration IR CBZ versus CR CBZ.

Outcome measures to assess cognitive side effects:

‐ 15‐word Test: Verbal memory

‐ Complex Figure Test: Nonverbal memory

‐ Stroop Word Colour Test: Attention

‐ WISC‐R maze: Concentration

‐ Perceptual Speed: Visual scanning ability

Tests to assess performance related to fluctuations in serum CBZ level:

‐ Tapping Task: Motor speed

‐ Corsi’s Block Tapping Test: Memory span

‐ Computerised Visual Searching Task: Visual‐spatial information processing

No follow up beyond completion

Compliance not assessed

Randomisation stated but method unclear

Concealment of allocation unclear

Blinding: tablets administered in same tablet form and dose frequency

Not analysed as intention‐to‐treat

11 patients with a diagnosis of epilepsy. Partial in 8, generalised in 2, multifocal in 1

5 male, 6 female

Age (mean) 32.1

Age (range) 16‐58

Mean age of onset of epilepsy 11.6 years (range birth‐25)

Patients treated with CBZ for at least 1 year

Non‐medication control group for cognitive testing standards was matched to epilepsy group for sex, age, educational level

Interventions/protocol

Results

Other

Patients randomised to receive either CR or IR CBZ in same tablet form and dose frequency

400mg CBZ given bd at 9.00AM and 7.00PM

At a non‐specified day during the study period, the psychological tests were administered 6 hourly

The non‐medication control group followed the same test scheme for 1 day

Summary:

CR CBZ produced overall increases in performance over IR in the following tests:

‐ 15 word test (direct and delayed recall) for verbal memory

‐ Complex figure test (recall) for non‐verbal memory

‐ Stroop Colour Word Test for attention

‐ Perceptual speed for visual scanning ability

CR CBZ produced superior results in all tests assessing differences in performance relating to fluctuations in serum level:

‐ Tapping task (dominant and non‐dominant hands) for motor speed

‐ Corsi’s Block Tapping Test for memory span

‐ Computerised Visual Searching Task for visual‐spatial information processing

Objective

Method

Quality assessment

Participants

2

RCT crossover

56‐day treatment arms, no washout period

Double blind including participant, but unclear whether outcome assessor or the clinician responsible for patients care is blinded

Relevant outcomes include seizure frequency recorded continuously in a diary and adverse events assessed periodically during interim follow‐up appointments

No follow up beyond study completion reported.

Compliance was assessed

Randomisation stated, method unclear

Concealment of allocation unclear

Participants blinded. Study is double blind but unclear whether outcome assessor or the clinician responsible for patients care is blinded

Not analysed as ‘intention to treat’

Patients diagnosed with partial or generalised epilepsy prescribed monotherapy with CBZ 200 mg tablets either tds or qds in a stable regimen for at least 3 months

No more than 3 seizures in each of the 3 months prior to enrolment

101 patients enrolled, 87 completed both arms

Both groups were statistically comparable. Mean age group 1 was 34, group 2 was 32 years

Interventions/protocol

Results

Other

During the 21‐day run‐in period patients continued their usual regimen of IR CBZ

Day 22: patients were randomised to receive either IR or CR CBZ at their usual daily dose. Group 1: IR CBZ given every 12 hours plus placebo. Group 2: CR given every 12 hours plus placebo

Day 78: patients were crossed over to alternate arm, no washout period mentioned

Mean CBZ dose 1084 mg (range 400‐2000)

No change in pattern of seizure frequency observed overall during both arms. 74% during CR and 77% during IR, the seizure rate did not exceed that at baseline

Mean monthly seizure rate 0.53 during CR, 0.41 during IR. Statistical significance is not reached

Adverse events:

‐ CR: 4 patients reported 6 adverse events, dizziness, (2) diplopia, headache, nausea and vomiting

‐ IR: 5 patients reported 5 adverse events, dizziness, drowsiness, hand tremor, stomach cramps and vomiting.

No adverse events were clinically significant or required discontinuation of therapy.

96% of patients were compliant at any study visit.

Objective

Method

Quality assessment

Participants

2

RCT double blind, crossover

CBZ monotherapy at individualised doses for 1‐month study period following 2‐month dose finding phase

No washout period stated

Compliance was assessed prior to the study and non‐compliant patients excluded

Relevant outcomes include seizure frequency recorded continuously in a seizure diary, and side effects assessed by direct enquiry according to a form

Randomised, method unclear

Concealment of allocation unclear

CBZ formulations were indistinguishable in taste and physical appearance

Outcome assessor blinded to treatment

Not analysed as intention to treat

48 patients, 21 male, 27 female

Treated with CBZ monotherapy for 3 months or greater with inadequate seizure control or intermittent side effects

Exclusion criteria: oto‐vestibular disease or poor pre‐study compliance

Age: mean 34.2, range 18‐64 years

No withdrawals

2 generalised, 46 partial seizures

Duration of epilepsy mean: 17.7 range: 1.5‐44

CBZ daily dose (mg) mean: 1.125, range: 400‐2.400

Interventions/protocol

Results

Other

Each period of cross over consisted of a 2‐month optimal dose finding phase. The dose was altered to reach the highest best tolerated in patients with seizures or lowest effective dose in patients with side effects

The 1‐month maintenance phase was used for statistical analysis during which the dose was not altered

Seizure frequency was recorded throughout the maintenance phase in a seizure diary

Side effects were checked by direct enquiry according to a form assessing the most common effects. In addition, at the end of each period a global evaluation of tolerability was assessed

CBZ total daily dose (mg) significantly (P=0.001) higher with CR CBZ: 1558.3±735.7 versus IR 1310.4±481.7

CR CBZ had significantly reduced number of administrations. 38 patients managed a bd regimen compared to 15 on IR (P=0.001)

Mean monthly seizure frequency was significantly reduced with CR: 6.3±9.8 versus 9.3±15.6 (P=0.013)

6 patients reported intermittent side effects with CR whereas 26 reported with IR (P=0.001)

Global evaluation of tolerability significantly better in CR CBZ group (P=0.001).

Objective

Method

Quality assessment

Participants

2

RCT crossover, double blind

2 weeks each arm

Dose was according to existing CBZ therapy. Patients were assigned to 800, 1200 or 1600mg/day

Relevant outcomes:

Incidence of adverse events

Seizure frequency

No follow up beyond study completion

Compliance was assessed

Randomisation stated, method unclear

Concealment of allocation: Identical capsules used and covered in powder and packaged in identical blister packs. Placebo tablets appeared the same and were packaged in the same blister packs

Double blind: participant and outcome assessor

Not analysed as intention to treat

Adult patients with a diagnosis of epilepsy prescribed IR CBZ at a stable and therapeutic dose for at least 30 days

24 patients were included. 1 patient was excluded from pharmacokinetic analysis due to blood sampling problems

Mean age: 36.1±8.1 (range 21‐54) years

13 F 11 M

CBZ dose (mg)

800: 9 patients

1200: 9 patients

1600: 6 patients

Interventions/protocol

Results

Other

CBZ dose was determined according to patients pre‐study dose: either 800, 1200 or 1600mg was administered. If a change was required the dose was kept the same for 30 days prior to starting the study

IR CBZ was divided into 4 doses/day

CR CBZ was divided into 2 doses/day

Placebo tablets were used during the CR CBZ arm

The occurrence of seizures and adverse events was recorded throughout the study periods. Method not stated

No patient was withdrawn from the study because of increased seizure frequency or adverse events

Mean number of seizures during the study periods:

CR CBZ: 2.8±6.2

IR CBZ: 1.6±3.9

However, 2 patients reported over half the total number of seizures

1 adverse event, somnolence, was reported during the IR CBZ treatment period

Compliance was assessed by pill counts and was of a high level and similar for both groups

Concomitant use of additional AEDs was allowed but the dosing remained constant.

9 patients received concomitant AEDs

Objective

Method

Quality assessment

Participants

2

RCT crossover

Baseline period for 2 months, each treatment arm 10 weeks long

Daily CBZ dose kept the same as prior to study

Seizure frequency is the single relevant outcome, recorded by experienced nurses both day and night. Participants remained in the institution for the duration of the study

No follow up beyond study completion

Randomisation stated but method unclear

Concealment of allocation: IR CBZ, CR CBZ and placebo tablets looked identical

Participants blinded but unclear whether outcome assessor or clinician responsible for participants care is blinded

Mentally retarded patients px CBZ at a therapeutic serum level for at least 2 months and with at least 4 seizures per month despite therapy were eligible

21 patients enrolled, 1 withdrawn

Mean age: 24.9±10.3 (range 6‐38) years

11 F, 9 M

18 patients experienced secondary generalised, and 2 primary generalised seizures

Mean CBZ dose: 780.0±370.8 mg

Interventions/protocol

Results

Other

2 10‐week arms followed 2‐month baseline period where usual CBZ treatment was given: IR was divided into 3 daily doses, CR was divided into 2 daily dosages with 1 placebo tablet

Tablets were taken at the same times during both arms

No washout period

Seizure frequency was recorded by experienced nurses both day and night. Participants remained in the institution for the duration of the study

1 patient was withdrawn due to appendicitis

Mean total number of epileptic seizures were equivalent:

CR: 44.0

IR: 42.7

Over time there was a significant trend for less seizures during CR CBZ (P=0.01)

During last 2 weeks of therapy, seizure frequency was significantly lower during CR CBZ (P=0.02)

There were no significant differences between the mean totals for different types of seizure

This trial involved mentally retarded patients, only 2 of which were prescribed CBZ monotherapy

Many patients were prescribed additional non‐antiepileptic medications

Objective

Method

Quality assessment

Participants

1

RCT, crossover

Study period: 8 weeks, 4 weeks in each arm

Patients stabilised on maximally tolerated doses of CBZ monotherapy for 3 months minimum. Previous attempts at increasing the dose resulted in neurotoxic adverse events

Relevant outcomes:

Seizure frequency

Incidence of cognitive adverse effects

No follow up beyond study completion

Compliance was assessed by a tablet count following the study

Randomisation stated, method unclear

Participant and outcome assessor blinded

Concealment of allocation is unclear

Not analysed as intention to treat

Adult patients with an existing diagnosis of epilepsy

25 patients were included. 9 reported generalised seizures, 16 complex partial

12 M, 13 F

Age range: 18‐53 years

4 patients were excluded from analysis: 3 did not comply with the protocol and 1 experienced inadequate seizure control during the IR CBZ phase

Mean CBZ dose: 1076mg

Range: 600‐2000mg

Interventions/protocol

Results

Other

Patients continued to take their usual CBZ dose for 3‐month baseline period during which seizure frequency was recorded

CR CBZ: taken twice daily with 2 placebo tablets in all patients

IR CBZ: qds in 4, tds in 8 and bd in 1. Placebo tablets were given if required

No mention of appearance of tablets

Seizure frequency charts were completed throughout the trial

Compliance was assessed following each treatment arm

Cognitive function tests were performed at 1, 4 and 8 hours following the morning dose at baseline and following each treatment arm

There were no differences in cognitive function between baseline and after 4 weeks treatment with IR CBZ. Reaction times using Leeds Psychomotor Tester were shorter after CR CBZ:

Mean time±SD after 1 hour:

IR CBZ: 0.51±0.19s

CR CBZ: 0.46±0.15s (P<0.01)

Mean time±SD after 4 hours:

IR CBZ: 0.49±0.15s

CR CBZ: 0.45±0.1s (P<0.05)

Adverse event scores were lower with CR CBZ at 1 hour:

8.7±7 with IR CBZ compared to 6.6±6.3 with CR CBZ  (P<0.05)

Seizure frequency was higher with CR CBZ:

2.8±1.2 with IR CBZ compared to 3.8±0.9 with CR CBZ (P<0.01)

However, control was not statistically worse than baseline

Compliance was good following tablet counts

Patients were prescribed CBZ monotherapy

Concomitant medication is not stated

Objective

Method

Quality assessment

Participants

1

RCT, parallel

The length of the study period is not stated but final measurements are taken on the 20^th day. It can be assumed this is the length of the study period

For both CR and IR CBZ, the target dose administered was 200mg tds

Relevant outcomes:

Incidence of adverse effects

No follow up beyond study completion

Compliance is mentioned in the method but has not been reported in the results

Randomisation stated, method unclear

Concealment of allocation: drugs given in plastic containers to ensure compliance but no mention of drug appearance

This study can be considered unblinded

Not analysed as intention to treat

Adult patients with a new diagnosis of partial seizures with no previous prescription of AED therapy

20 patients were included. No withdrawals are mentioned

IR CBZ group:

Mean age: 20.32±8.28

(range 16‐34) years

3 F, 7 M

CR CBZ group:

Mean age: 22.48±9.23

(range 18‐35) years

2 F, 8 M

Interventions/protocol

Results

Other

CBZ doses in both groups given in increments:

100mg bd for 2 days

200mg bd for 2 days

200mg tds

The drugs were given in plastic containers to ensure compliance

There is no mention of placebo tablets or the appearance of the tablets themselves

The occurrence of adverse events has been reported but there is no mention of how they were assessed

4 patients reported adverse events in the IR CBZ group:

Sedation was present in 2

Diplopia/ataxia in 1

Skin rash in 1

2 patients reported adverse events in the CR CBZ group:

Sedation in 1

Diplopia/ataxia in 1

This study is unblinded and of small sample size

Patients were not prescribed any other concurrent medication

Discrepancies between results reported in text and that stated in tables

Objective

Method

Quality assessment

Participants

1           

RCT, crossover

Study period: 2 month baseline followed by 3 months in each arm

Adult patients with epilepsy with few or no seizures and with subjectively moderate to severe adverse events caused by IR CBZ were entered into the trial

Relevant outcomes:

Seizure frequency

Incidence of adverse effects

No follow up beyond study completion

Compliance was not assessed

Randomisation performed by a computer program

Participant and outcome assessor blinded

Concealment of allocation was adequate: IR CBZ and CR CBZ tablets looked identical and were supplied in identical containers with appropriate labels

Not analysed as intention to treat

21 adult patients with epilepsy with few or no seizures and with subjectively moderate to severe adverse events caused by IR CBZ. 1 withdrew due to ataxia

9 M, 11 F

Age: Mean: 43.35±15.7 (range 20‐69) years

Duration of epilepsy:

Mean: 17.7±14.4 (range 0.5‐47)

4 patients interrupted treatment during IR CBZ due to adverse events but tolerated SR CBZ. 1 patient interrupted treatment during both arms due to adverse events

Mean CBZ dose: 682mg (range 300‐1100mg)

Interventions/protocol

Results

Other

Patients continued to take their usual CBZ dose for 2‐month baseline period

Total daily dose and dosing frequency was kept unchanged throughout study. There was no mention of placebo tablets

Adverse events were assessed at monthly visits by the study investigators using the Questionnaire on Systemic Toxicity (STRS) and Neurotoxicity Rating Scales (NTRS). Mean values for each arm were used in the analysis

Seizure frequency charts were completed throughout the trial and registered at each monthly visit

1 patient withdrew during IR CBZ due to ataxia and did not continue

STRS mean total scores:

IR CBZ:   12.1±19.0

CR CBZ:  8.3±18.6 (P=0.09)

NTRS mean total scores:

IR CBZ:   80.3±66.2

CR CBZ:  47.2±39.5 (P=0.04)

NTRS + STRS mean total scores:

IR CBZ:   92.5±68.8

CR CBZ:  55.5±46.4 (P=0.04)

Statistically significant reductions in subscores for the occurrence of GI problems, disturbance of vision, speech and motor function, dizziness and headache

Seizure frequency:

2.2 per patient during first month of IR CBZ

1.2 per patient during first month of CR CBZ.

No significant differences in mean seizure frequency per month

11 patients preferred CR CBZ, 3 preferred IR CBZ, and 6 had no preference. (P=0.0176)

9 patients were prescribed additional AEDs, the dose of which was unaltered throughout

Concomitant medication is not stated    

Obj

Method

Quality assessment

Participants

2

RCT, crossover

2 weeks in each treatment arm with no washout period. IR CBZ versus CR CBZ

Single blind: outcome assessor blinded

Outcomes include clinical parameters such as total seizure incidence and total incidence of reported adverse effects. Data was recorded for these parameters throughout the study period

No follow up beyond study completion reported

Compliance not assessed

Randomised, method unclear

Concealment of allocation unclear

Observer blinded to treatment, drugs were given in identical containers

Not analysed as intention to treat

21 patients diagnosed with epilepsy and experiencing simple or complex partial seizures, on stable CBZ therapy for at least 3 months

3 withdrawals

18 evaluated: 10 females, 8 males

Mean age: 42 (±10) years

Interventions/protocol

Results

Other

CBZ given in identical containers during each 2‐week study period

Both controlled and immediate release were given at 12 hourly intervals

Dose was kept the same as before the study. Mean: 644±200mg

Throughout study period seizures and subjective adverse effects reported by patients were recorded. Dizziness, fatigue, movement disorders and visual disturbance received ‘special attention’

Patient preference was sought at the end of the study  

Wilcoxons’s signed rank and McNemar’s tests were used to test the significance of differences between the treatment periods

During immediate‐release CBZ 12 generalised, 25 complex partial and 8 simple partial, totaling 56 seizures were recorded

During controlled release CBZ, 2 generalised, 18 complex partial and 11 simple partial, totaling 31 seizures were recorded. The difference did not reach significance P=0.093

Adverse effects were reported 12 times during CR and 19 times during IR CBZ

Dizziness was reported 7 times during IR and once during CR, P=0.034.

12 patients reported no adverse effects during CR compared to 9 during IR

Only 11 patients prescribed monotherapy

Objective

Method

Quality assessment

Participants

2

RCT, crossover,

2 weeks each arm

Dose remained the same as prior to study

Relevant outcomes:

Incidence of adverse effects

Seizure frequency

No follow up beyond study completion

Compliance was assessed

Randomisation stated, method unclear

Concealment of allocation: drugs given in identical containers, no mention of appearance of drugs or inclusion of placebo tablets

This study can be considered unblinded

Analysed as intention to treat

Adult patients with a diagnosis of epilepsy prescribed CBZ monotherapy at a stable and therapeutic dose for 6 months minimum were eligible

24 patients were included. 2 excluded due to compliance issues, 1 due to protocol violation and 1 due to difficulties in analysis

Mean age: 36.9 (range 18‐62) years

Mean CBZ dose (mg) 615.0 (range 300‐1100)

9 F, 11 M

Interventions/protocol

Results

Other

CBZ dose remained the same as prior to study

IR CBZ was divided into 3 doses

CR CBZ was divided into 2 doses

The drugs were given in identical containers

There is no mention of placebo tablets or the appearance of the tablets themselves

The occurrence of seizures and adverse effects was recorded during the study periods. Method not stated. Special attention was paid to dizziness, fatigue and visual disturbance

Patient preference was determined following study completion

9 seizures during each treatment arm

Fatigue was the most common adverse effect, occurring in 4 patients in each group. There was ‘practically no difference in adverse effects between the 2 treatment periods’

50% of patients preferred CR CBZ

20% preferred IR CBZ

30% had no preference

3 patients were prescribed other non‐antiepileptic medication

This study is unblinded and of small sample size