Lamotrigine add‐on therapy for drug‐resistant focal epilepsy

Abstract

Background

This is an updated version of a Cochrane Review last updated in 2020.

Epilepsy is a common neurological disorder, affecting 0.5% to 1% of the population. In nearly 30% of cases, epilepsy is resistant to currently available drugs. Pharmacological treatment remains the first choice to control epilepsy. Lamotrigine is a second‐generation antiseizure medication. When used as an add‐on (in combination with other antiseizure medications), lamotrigine can reduce seizures, but with some adverse effects.

Objectives

To evaluate the benefits and harms of add‐on lamotrigine, compared with add‐on placebo or no add‐on treatment in people with drug‐resistant focal epilepsy.

Search methods

For this update, we searched the Cochrane Register of Studies (CRS Web) and MEDLINE (Ovid) on 3 October 2022 with no language restrictions. CRS Web includes randomised and quasi‐randomised controlled trials from PubMed, Embase, ClinicalTrials.gov, the World Health Organization International Clinical Trials Registry Platform (ICTRP), Cochrane Central Register of Controlled Trials (CENTRAL), and the Specialised Registers of Cochrane Review Groups, including Epilepsy.

Selection criteria

We included randomised controlled trials (RCTs) that investigated add‐on lamotrigine versus add‐on placebo or no add‐on treatment in people of any age with drug‐resistant focal epilepsy. We used data from the first period of eligible cross‐over trials.

Data collection and analysis

For this update, two review authors independently selected trials and extracted data. Our primary outcome was 50% or greater reduction in seizure frequency. Our secondary outcomes were treatment withdrawal, adverse effects, cognitive effects, and quality of life. Primary analyses were by intention‐to‐treat. We performed sensitivity best‐ and worse‐case analyses to account for missing outcome data. We calculated pooled risk ratios (RRs) with 95% confidence intervals (95% Cls) for dichotomous outcomes.

Main results

We identified no new studies for this update, so the results and conclusions of the review are unchanged.

We included five parallel‐group studies in adults or children, eight cross‐over studies in adults or children, and one parallel study with a responder‐enriched design in infants. In total, these 14 studies enroled 1806 eligible participants (38 infants, 199 children, 1569 adults). Baseline phases ranged from four to 12 weeks and treatment phases ranged from eight to 36 weeks. We rated 11 studies (1243 participants) at low overall risk of bias and three (697 participants) at unclear overall risk of bias due to lack of information on study design. Four studies (563 participants) reported effective blinding.

Lamotrigine compared with placebo probably increases the likelihood of achieving 50% or greater reduction in seizure frequency (RR 1.80, 95% CI 1.45 to 2.23; 12 trials, 1322 participants (adults and children); moderate‐certainty evidence). There is probably little or no difference in risk of treatment withdrawal for any reason among people treated with lamotrigine versus people treated with placebo (RR 1.11, 95% CI 0.91 to 1.37; 14 trials; 1806 participants; moderate‐certainty evidence).

Lamotrigine compared with placebo is probably associated with a greater risk of ataxia (RR 3.34, 99% Cl 2.01 to 5.55; 12 trials; 1525 participants; moderate‐certainty evidence), dizziness (RR 1.76, 99% Cl 1.28 to 2.43; 13 trials; 1768 participants; moderate‐certainty evidence), nausea (RR 1.81, 99% CI 1.22 to 2.68; 12 studies, 1486 participants; moderate‐certainty evidence), and diplopia (RR 3.79, 99% Cl 2.15 to 6.68; 3 trials, 944 participants; moderate‐certainty evidence). There is probably little or no difference in the risk of fatigue between lamotrigine and placebo (RR 0.82, 99% CI 0.55 to 1.22; 12 studies, 1552 participants; moderate‐certainty evidence).

Authors' conclusions

Lamotrigine as an add‐on treatment for drug‐resistant focal seizures is probably effective for reducing seizure frequency. Certain adverse effects (ataxia, dizziness, diplopia, and nausea) are probably more likely to occur with lamotrigine compared with placebo. There is probably little or no difference in the number of people who withdraw from treatment with lamotrigine versus placebo.

The trials were of relatively short duration and provided no long‐term evidence. In addition, some trials had few participants.

Further trials are needed to assess the long‐term effects of lamotrigine and to compare lamotrigine with other add‐on drugs.

Plain language summary

Lamotrigine add‐on therapy for drug‐resistant focal epilepsy

Key messages

1. Lamotrigine is an antiseizure medication used as an adjunctive treatment for focal seizures in people with drug‐resistant epilepsy. Focal epilepsy is characterised by seizures arising from a specific area of the brain.
3. Lamotrigine is probably effective for reducing seizure frequency in children and adults, but further trials are needed to assess the long‐term effects of lamotrigine and to compare lamotrigine with other add‐on drugs.

What is epilepsy and how is it treated?

Epilepsy is a brain disorder that causes recurring seizures. Approximately one‐third of people with epilepsy continue to have seizures despite using antiseizure medicines. Older antiseizure medicines can have many unwanted effects, so it is important to develop effective new therapies. Several new medicines have been developed as 'add‐on' treatments (to be used in combination with other medicines). One of these new medicines is called lamotrigine.

What did we want to find out?

We wanted to find out if add‐on lamotrigine was better than add‐on placebo (dummy treatment) or no add‐on treatment for reducing the frequency of seizures and withdrawal from treatment, and for improving cognition (ability to learn) and quality of life. We also wanted to know if lamotrigine had any unwanted effects.

What did we do?

We searched for studies that examined add‐on lamotrigine compared with add‐on placebo or no add‐on treatment in people of any age who had focal epilepsy that had not responded to previous treatments (drug‐resistant focal epilepsy). We compared and summarised the results of the studies and rated our confidence in the evidence, based on factors such as study methods and sizes.

What did we find?

We found 14 studies that enroled a total of 1806 people (adults, children, and infants).

Lamotrigine, used in combination with other antiseizure medicines in people with drug‐resistant focal epilepsy, is probably more effective than placebo for achieving a 50% or greater reduction in seizure frequency. In addition, people using add‐on lamotrigine are probably no more likely to stop treatment than those using add‐on placebo. However, adding lamotrigine to usual treatment probably leads to more unwanted effects such as unsteadiness (ataxia), dizziness, double vision (diplopia), and nausea.

Further high‐quality research is needed to fully evaluate the benefits and harms of add‐on lamotrigine compared with add‐on placebo or no add‐on treatment, and compared with other new antiseizure medicines.

What are the limitations of the evidence?

We have moderate confidence in the evidence, mainly because the studies reported few events (seizures, treatment withdrawal, unwanted effects).

How up to date is this evidence?

This review updates our previous review. The evidence is current to 3 October 2022.

Summary of findings

Summary of findings 1. Summary of findings: lamotrigine versus placebo for drug‐resistant focal epilepsy.

Lamotrigine versus placebo for drug‐resistant focal epilepsy
Patient or population: participants with drug‐resistant focal epilepsy Settings: outpatient setting Intervention: add‐on lamotrigine Control: add‐on placebo
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Assumed risk (placebo)	Corresponding risk (lamotrigine)
50% or greater reduction in seizure frequency (ITT analysis) Follow‐up (range): 8–36 weeks	157 per 1000	283 per 1000 (223 to 350)	RR 1.80 (95% CI 1.45 to 2.23)	1322 (12 studies)	⊕⊕⊕⊝ Moderatea	Lamotrigine compared to placebo probably increases the likelihood of achieving a 50% or greater reduction in seizure frequency.
Treatment withdrawal Follow‐up (range): 8–36 weeks	159 per 1000	176 per 1000 (144 to 217)	RR 1.11 (95% CI 0.91 to 1.37)	1806 (14 studies)	⊕⊕⊕⊝ Moderatea	Lamotrigine compared to placebo probably has little or no effect on the risk of treatment withdrawal.
Ataxia Follow‐up (range): 8–36 weeks	45 per 1000	150 per 1000 (90 to 250)	RR 3.34 (99% CI 2.01 to 5.55)	1525 (12 studies)	⊕⊕⊕⊝ Moderatea	Lamotrigine compared to placebo probably increases the risk of ataxia.
Dizziness Follow‐up (range): 8–36 weeks	128 per 1000	255 per 1000 (194 to 337)	RR 1.76 (99% CI 1.28 to 2.43)	1768 (13 studies)	⊕⊕⊕⊝ Moderateb	Lamotrigine compared to placebo probably increases the risk of dizziness.
Fatigue Follow‐up (range): 8–36 weeks	113 per 1000	93 per 1000 (62 to 138)	RR 0.82 (99% CI 0.55 to 1.22)	1552 (12 studies)	⊕⊕⊕⊝ Moderatea	Lamotrigine compared to placebo probably has little or no effect on the risk of fatigue.
Nausea Follow‐up (range): 8–36 weeks	83 per 1000	150 per 1000 (101 to 222)	RR 1.81 (99% CI 1.22 to 2.68)	1486 (12 studies)	⊕⊕⊕⊝ Moderatea	Lamotrigine compared to placebo probably increases the risk of nausea.
Diplopia Follow‐up (range): 8–36 weeks	61 per 1000	233per1000 (132 to 410)	RR 3.79 (2.15 to 6.68)	944 (3 studies)	⊕⊕⊕⊝ Moderatea	Lamotrigine compared to placebo probably increases the risk of diplopia.
*The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; RR: risk ratio.
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect. Very low certainty: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

^a Downgraded once for imprecision: number of events (< 400) was below the optimal information size.
^b Downgraded once for inconsistency: we detected significant statistical heterogeneity (P < 0.10).

Background

This is an update of a Cochrane Review first published in 2001 and last updated in 2020 (Ramaratnam 2001; Ramaratnam 2016; Panebianco 2020).

Description of the condition

Epilepsy is characterised by recurrent and unprovoked seizures; these seizures constitute a transient sign and symptom of abnormal, excessive electrical activity in the cerebral cortex (Fisher 2005). Epilepsy is one of the most common serious neurological conditions worldwide, affecting approximately 50 million people, and with significant psychosocial and physical morbidity. Management of the condition requires expertise and good knowledge of the available pharmacological options (Lyer 2014). The total annual cost in Europe is approximately EUR 15.5 billion (Mula 2013). Seizures affect the lives of 10% of the global population and result in epilepsy in 1% to 2% of people around the world (Falco‐Walter 2020). Most people will become seizure free after a few years; however, up to 30% of cases are refractory to treatment with adequate doses of antiseizure medications (ASMs), which are often given in combination (Cockerell 1995, Hakami 2021, Kwan 2010). Depending on where seizures start in the brain, they can be classified into focal onset, generalised onset, and unknown onset. In focal epilepsy, seizures start in a specific area of the brain; the abnormal electrical activity can remain localised or spread to other parts of the brain during the seizure. Most people with drug‐resistant epilepsy have focal onset seizures. Focal epilepsies (unifocal or multifocal) are seizure disorders that originate within a neuronal network limited to one hemisphere. This is in contrast to generalised epilepsy, where seizures rapidly engage bilateral distributed networks. There are many types of focal seizures, including focal aware seizures, focal impaired awareness seizures, focal motor seizures, focal non‐motor seizures, and focal to bilateral tonic‐clonic seizures. The interictal electroencephalogram (EEG) typically shows focal epileptiform discharges, but diagnoses are primarily based on clinical findings, supported by EEG findings (Commission 1989; Scheffer 2017).

Description of the intervention

Although many new ASMs have entered the market since 1993, up to 30% of people with epilepsy remain refractory to current treatments. Thus, there is a concerted effort to identify and develop new therapies (Barker‐Haliski 2014, Bresnahan 2020). Pharmacological treatment remains the first‐line approach to controlling epilepsy (Loscher 2002), although there have been advances in vagal stimulation and surgery (Panebianco 2015, Panebianco 2022a; West 2019). As standard drugs (e.g. carbamazepine, phenytoin, valproate, gabapentin, pregabalin) cannot eliminate seizures in all people and are not without adverse effects (Panebianco 2018, Panebianco 2021, Panebianco 2022b), there has been renewed interest in developing novel ASMs since the beginning of the 21st century. Lamotrigine is a second‐generation ASM that is widely used as an adjunctive treatment for focal, secondarily generalised, and tonic‐clonic seizures in people with drug‐resistant epilepsy or bipolar disorder (Hakami 2021; Yamamoto 2012).

How the intervention might work

The U.S. Food and Drug Administration (FDA) approved lamotrigine in 1994 as an add‐on for use in focal‐onset seizures. It was approved for monotherapy in 1998. Lamotrigine is effective against a broad spectrum of seizure types and has a favourable metabolic profile. It is frequently prescribed in the USA as both an immediate and an extended‐release agent (Moore 2012). Lamictal (GlaxoSmithKline) is considered the reference drug (Girolineto 2012). In vitro pharmacological studies have suggested that the main mechanism of action of lamotrigine is to inhibit voltage‐sensitive sodium channels, thereby stabilising neuronal membranes and consequently modulating presynaptic transmitter release of excitatory amino acids such as glutamate and aspartate (Leach 1995). Research has demonstrated the efficacy of lamotrigine as both an antiseizure drug and a mood stabiliser (Vajda 2013).

Why it is important to do this review

In this review, we summarised evidence from randomised controlled trials (RCTs) that investigated the efficacy and tolerability of lamotrigine in people with drug‐resistant focal epilepsy. Our ultimate objective was to aid decision‐making for clinicians considering lamotrigine as add‐on treatment in this population. As ASMs may impair people's cognitive abilities, we included outcomes related to cognitive effects in this review. In addition, we chose to include quality of life (QOL) outcomes to assess the global impact of this drug on people's well‐being.

Objectives

To evaluate the benefits and harms of add‐on lamotrigine, compared with add‐on placebo or usual treatment with no add‐on treatment in people with drug‐resistant focal epilepsy.

Methods

Criteria for considering studies for this review

Types of studies

We included parallel‐group and cross‐over randomised controlled trials (RCTs). Double‐blind, single‐blind, and unblinded trials were eligible.

Types of participants

Individuals of any age with drug‐resistant focal epilepsy (focal aware seizures, focal impaired awareness seizures, focal motor seizures, focal non‐motor seizures, and focal to bilateral tonic‐clonic seizures) that had not responded to at least two ASMs.

Types of interventions

In eligible studies, the treatment group received lamotrigine in addition to conventional antiseizure treatment, and the control group received placebo in addition to conventional antiseizure treatment or conventional antiseizure treatment only (i.e. no add‐on treatment).

Types of outcome measures

Primary outcomes

Our primary outcome was 50% or greater reduction in seizure frequency during the treatment period, compared to the prerandomisation baseline frequency. We chose this outcome as it is commonly reported in this type of study and can be calculated for all studies that provide baseline seizure data.

Secondary outcomes

1. Treatment withdrawal: we chose to report the proportion of participants who had their treatment withdrawn during the treatment period as a measure of global effectiveness. Clinicians may choose to withdraw treatment because of adverse effects, lack of efficacy, or both, and the individual directly contributes to this decision. However, in studies of relatively short duration (i.e. the type of studies we expected to include in this review), adverse effects are likely to be the main reason for withdrawal.

2. Common adverse effects: the proportion of participants who experienced ataxia, dizziness, fatigue, or nausea, which we considered to be the most common and important adverse effects of ASMs.

3. Other adverse effects: the proportion of participants who experienced the five most common adverse effects in a study, if different from those stated above.

4. Cognitive effects: the difference between intervention and control group means in cognitive assessments used in individual studies.

5. Quality of life (QOL): the difference between intervention and control group means in QOL assessments used in individual studies.

Search methods for identification of studies

Electronic searches

For this update, we searched the following databases on 3 October 2022.

1. Cochrane Register of Studies (CRS Web), using the strategy outlined in Appendix 1

2. MEDLINE (Ovid, 1946 to 30 September 2022), using the strategy outlined in Appendix 2

CRS Web includes RCTs and quasi‐RCTs from PubMed, Embase, ClinicalTrials.gov, the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP), the Cochrane Central Register of Controlled Trials (CENTRAL), and the Specialised Registers of Cochrane Review Groups, including Epilepsy. In MEDLINE (Ovid), the coverage end date always lags a few days behind the search date. We imposed no language restrictions.

Searching other resources

We checked reference lists of reviews and retrieved articles for additional studies, and performed citation searches on key articles.

We contacted experts in the field to enquire about unpublished and ongoing trials, and we contacted study authors and manufacturers of lamotrigine (GlaxoSmithKline) for additional information.

Data collection and analysis

Selection of studies

For this update, two review authors (MP and RB) independently assessed trials for inclusion, extracted data from the included trials, and assessed the risk of bias of the included trials. They resolved any disagreements by discussion with a third review author (AM).

Data extraction and management

We extracted the following data from each trial using a data extraction form.

1. Methods and trial design

   1. Method of randomisation

   2. Method of allocation concealment

   3. Method of blinding

   4. Whether any participants had been excluded from reported analyses

   5. Duration of baseline period

   6. Duration of treatment period

   7. Dose(s) of lamotrigine tested

   8. Information on sponsorship and funding

2. Participant and demographic information

   1. Total number of participants allocated to each treatment group

   2. Age and sex

   3. Number with focal and generalised epilepsy

   4. Seizure types

   5. Seizure frequency during the baseline period

   6. Number of background drugs

3. Outcomes: number of participants who experienced each outcome per randomised group (see Types of outcome measures). We contacted trial authors for missing information where necessary.

Assessment of risk of bias in included studies

Two review authors (MP and RB) independently assessed the risk of bias of each trial using the original Cochrane risk of bias tool (RoB 1), as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreements by discussion. We completed a risk of bias table for each included study in RevMan 5 (RevMan 2014). We rated all included studies as being at low, high, or unclear risk of bias in each of the following domains.

1. Random sequence generation

2. Allocation concealment

3. Blinding of participants and personnel

4. Blinding of outcome assessment

5. Incomplete outcome data

6. Selective outcome reporting

7. Other sources of bias

Measures of treatment effect

We analysed the primary outcome of seizure reduction and the secondary outcome of treatment withdrawal as dichotomous outcomes and presented the treatment effects as risk ratios (RRs) with 95% confidence intervals (CIs). We also analysed adverse effects as dichotomous outcomes and calculated RRs with 99% CIs. We had planned to present cognitive effects and QOL as continuous outcomes, reporting the mean difference (MD) if all studies contributing to an outcome had used the same measurement scale, or the standardised mean difference (SMD) if studies had used different measurement scales for the same outcome. However, due to the limited amount of data available for these outcomes, we presented them in a narrative discussion.

Unit of analysis issues

We included eight cross‐over studies, analysing data from the first treatment period only. We analysed parallel and cross‐over studies in separate subgroups.

Dealing with missing data

We sought missing data by contacting the study authors. We carried out intention‐to‐treat (ITT), best‐case, and worst‐case analyses of the primary outcome to account for any missing data (see Data synthesis). We presented all analyses in the main report.

Assessment of heterogeneity

We assessed clinical heterogeneity by comparing the distribution of important participant factors (e.g. age, seizure type, duration of epilepsy, number of ASMs taken at time of randomisation) and trial factors (e.g. randomisation concealment, blinding, losses to follow‐up) across the included trials. We examined statistical heterogeneity using the Chi² test and the I² statistic. When we found no significant heterogeneity (P < 0.10 in the Chi² test), we used a fixed‐effect model. Had we found substantial heterogeneity (I² > 50%), we would have used a random‐effects model for the analysis.

Assessment of reporting biases

We requested protocols for all included studies to ensure the a priori outcomes corresponded with the outcomes reported in the publication. We had planned to investigate any suspected outcome reporting bias using the ORBIT matrix system (Kirkham 2010). Had we included more than 10 studies in a single meta‐analysis, we would have created funnel plots to investigate publication bias.

Data synthesis

We used a fixed‐effect meta‐analysis to synthesise the data. We measured the effect of each intervention on our prespecified primary and secondary outcomes, if data were available. We expected to evaluate the following comparisons.

1. Usual treatment plus lamotrigine versus usual treatment plus placebo

2. Usual treatment plus lamotrigine versus usual treatment (no add‐on treatment)

Our preferred estimator for all binary outcomes was the Mantel‐Haenzsel RR. For the outcomes 50% or greater reduction in seizure frequency and treatment withdrawal, we calculated 95% CIs. For individual adverse effects, we calculated 99% CIs to make an allowance for multiple testing.

Our analyses included all participants in the treatment groups to which they had been allocated following randomisation.

For the efficacy outcome (50% or greater reduction in seizure frequency), we first undertook the primary (ITT) analysis, assuming that participants who had not completed follow‐up or who had inadequate seizure data were non‐responders. In this primary analysis, we included data from studies that had performed ITT analysis. To test the effect of this assumption, we undertook the following sensitivity analyses.

1. Worst‐case analysis: participants not completing follow‐up or with inadequate seizure data were assumed to be non‐responders in the lamotrigine group and responders in the placebo group.

2. Best‐case analysis: participants not completing follow‐up or with inadequate seizure data were assumed to be responders in the lamotrigine group and non‐responders in the placebo group.

Subgroup analysis and investigation of heterogeneity

We performed subgroup analyses based on age of participants (adults and children) and dose of lamotrigine.

Sensitivity analysis

We intended to carry out sensitivity analyses if we identified any relevant issues during the review in terms of study quality, characteristics of participants, interventions, or outcomes.

Summary of findings and assessment of the certainty of the evidence

We created summary of findings tables for the following outcomes using GRADEpro software (GRADEpro 2014)

1. 50% or greater reduction in seizure frequency

2. Treatment withdrawal

3. Adverse effects (each presented separately)

One review author (RB) rated the certainty of the evidence using the GRADE approach. In GRADE, evidence from RCTs begins with a 'high certainty' rating but can be downgraded to 'moderate certainty', 'low certainty', or 'very low certainty' for limitations related to risk of bias, inconsistency, indirectness, imprecision, or publication bias (Schünemann 2023).

Results

Description of studies

Results of the search

The search carried out on 3 October 2022 identified 42 records from the databases listed in Search methods for identification of studies.

After removing duplicates, we screened the titles and abstracts of 33 records, eliminating 29 that we considered to be clearly irrelevant. We retrieved and reviewed the full‐text articles of the remaining four records. We contacted the authors of these trials for more information, provided their contact details were available. Following this, we excluded all four studies (see Figure 1 and Characteristics of excluded studies for reasons).

1.

Study flow diagram for 2023 update.

Subsequent searches identified no new relevant studies. Therefore, we included no new studies in this update.

Included studies

The previous version of this review included 14 RCTs that investigated the use of add‐on lamotrigine compared to add‐on placebo in 1806 people with uncontrolled focal seizures (38 infants, 199 children, 1569 adults). For further information on each trial, see the Characteristics of included studies table.

There were five parallel‐group studies (Baulac 2010; Duchowny 1999; Matsuo 1993; Naritoku 2007; Schachter 1995); eight cross‐over studies (Binnie 1989; Boas 1996; Jawad 1989; Loiseau 1990; Messenheimer 1994; Schapel 1993; Schmidt 1993; Smith 1993); and one parallel‐group responder‐enriched study, in which all participants received adjunctive lamotrigine during an open‐label phase, and those who had a 40% or greater reduction in the frequency of focal seizures during the last four weeks were randomly assigned to double‐blind treatment for up to eight weeks with continued lamotrigine or placebo (Piña‐Garza 2008). Duchowny 1999 recruited only children, and Piña‐Garza 2008 enroled only infants aged one to 24 months; the remaining studies included mostly adults. Naritoku 2007 used an extended‐release formulation of lamotrigine, while the remaining trials used immediate‐release formulations. In general, participants had at least three to four focal seizures a month despite being on stable regimens of two or three appropriate ASMs given in adequate doses.

Exclusion criteria in almost all studies were: intellectual disabilities, progressive neurological disease, major psychiatric problems, associated pseudo seizures, newly‐diagnosed epilepsy, status epilepticus in the 24 weeks preceding the trial, associated systemic diseases, abnormal laboratory investigations not explained by enzyme induction by ASMs, a history of non‐compliance, failure to keep reliable records of seizures or adverse effects, irregular clinic visits, recent use of any other investigational ASM, abuse of alcohol or other prescription or non‐prescription drugs, chronic use of medication (especially antipsychotic drugs), pregnancy or risk of pregnancy, and breastfeeding. For the cross‐over studies, participants were not randomised to a single dose, but took a range of doses, depending on their clinical response and the concurrent administration of other ASMs. Three studies did not permit use of valproate (Matsuo 1993; Messenheimer 1994; Schachter 1995), two excluded people on valproate monotherapy (Schapel 1993; Smith 1993), while others used lower dosages of lamotrigine for people on valproate. One parallel‐group study tested doses of 300 mg and 500 mg of lamotrigine per day (Matsuo 1993), whereas the others tested a range of doses between 75 mg and 600 mg per day (median between 200 mg/day and 400 mg/day). The length of the treatment period varied from eight weeks to 24 weeks.

Thirteen studies were published as full‐text articles, and Schmidt 1993 was published as an abstract only. The manufacturer of lamotrigine (GlaxoSmithKline) sponsored 13 studies as part of the prelicensing programme. Pfizer Inc sponsored Baulac 2010.

Excluded studies

We excluded four studies in this update: two were not RCTs, and two did not investigate add‐on lamotrigine (see the Characteristics of excluded studies table).

Ongoing studies

We identified one ongoing study with no published results (NCT03689114; see the Characteristics of ongoing studies table).

Risk of bias in included studies

Figure 2 and Figure 3 show the review authors' judgements about each risk of bias item across all included studies and for each included study.

2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

3.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

We rated 11 trials (1243 participants) at low risk of selection bias because they used a computer‐generated randomisation schedule or random number tables, and because they dispensed sequentially numbered packages to each participant (Binnie 1989; Boas 1996; Duchowny 1999; Jawad 1989; Loiseau 1990; Matsuo 1993; Messenheimer 1994; Schachter 1995; Schapel 1993; Schmidt 1993; Smith 1993). Three trials (563 participants) did not provide sufficient information on the method of random sequence generation, so were at unclear risk of selection bias (Baulac 2010; Naritoku 2007; Piña‐Garza 2008).

Blinding

We rated four studies (704 participants) at low risk of performance and detection bias because they blinded participants, parents, and investigators (Binnie 1989; Duchowny 1999; Loiseau 1990; Schachter 1995). Nine studies (1021 participants) did not provide details of the method of blinding, so were at unclear risk of bias for this domain (Baulac 2010; Boas 1996; Jawad 1989; Matsuo 1993; Messenheimer 1994; Naritoku 2007; Piña‐Garza 2008; Schapel 1993; Schmidt 1993). We rated Smith 1993 (81 participants) at high risk of bias because participants and investigators were able to identify the lamotrigine treatment.

Incomplete outcome data

We rated 13 studies (1725 participants) at low risk of attrition bias because they had minimal missing data, and because they either used ITT analysis or had missing data that did not affect the outcome estimate (Baulac 2010; Binnie 1989; Boas 1996; Duchowny 1999; Jawad 1989; Loiseau 1990; Matsuo 1993; Messenheimer 1994; Naritoku 2007; Piña‐Garza 2008; Schachter 1995; Schapel 1993; Schmidt 1993). Smith 1993 (81 participants) was at unclear risk of attrition bias because participants who were withdrawn or who dropped out did not complete the health‐related quality of life (HRQOL) measure at the time of discontinuation.

Selective reporting

We requested the protocols for all included studies to ensure a priori methods and outcomes were as reported in the publication; however, no protocols were available. We rated all studies at low risk of reporting bias because they reported all expected outcomes.

Other potential sources of bias

Baulac 2010 reported responder rates as percentages without providing numbers of responders, so we judged the study at unclear risk of other bias. For the remaining studies, we detected no other sources of bias.

Effects of interventions

See: Table 1

For the cross‐over trials, we analysed data from the first treatment phase (before crossover). We obtained these data from the sponsor, GlaxoSmithKline.

Primary outcome

50% or greater reduction in seizure frequency

Twelve studies (1322 participants; adults and children) reported our primary outcome (Baulac 2010; Binnie 1989; Boas 1996; Duchowny 1999; Jawad 1989; Loiseau 1990; Matsuo 1993; Messenheimer 1994; Naritoku 2007; Schapel 1993; Schmidt 1993; Smith 1993). A Chi² test for responses to lamotrigine indicated no significant heterogeneity between trials (Chi² = 11.02, df = 11, P = 0.44; I² = 0%), so we used a fixed‐effect model to measure efficacy.

Lamotrigine (any dose) added to regular ASM therapy, compared with placebo added to the same therapy, probably increases the likelihood of achieving 50% or greater reduction in seizure frequency (RR 1.80, 95% CI 1.45 to 2.23; moderate‐certainty evidence; Analysis 1.1).

1.1. Analysis.

Comparison 1: Add‐on lamotrigine versus add‐on placebo, Outcome 1: ≥ 50% reduction in seizure frequency (intention‐to‐treat analysis)

The analysis that only included cross‐over trials (8 trials, 382 participants) produced a similar result (RR 2.58, 95% CI 1.44 to 4.61; Analysis 1.1). Our worst‐case analysis showed no effect (RR 0.97, 95% CI 0.82 to 1.15; Analysis 1.2), while our best‐case analysis showed a stronger effect favouring lamotrigine (RR 2.88, 95% CI 2.36 to 3.50; Analysis 1.3).

1.2. Analysis.

Comparison 1: Add‐on lamotrigine versus add‐on placebo, Outcome 2: ≥ 50% reduction in seizure frequency (worst‐case analysis)

1.3. Analysis.

Comparison 1: Add‐on lamotrigine versus add‐on placebo, Outcome 3: ≥ 50% reduction in seizure frequency (best‐case analysis)

Our subgroup analyses according to dose of lamotrigine suggested that a daily dose of 300 mg has little or no effect on the likelihood of achieving 50% or greater seizure reduction (RR 1.23, 95% CI 0.57 to 2.67; 1 study; 144 participants; Analysis 1.1), while 500 mg per day has a positive effect (RR 2.13, 95% CI 1.08 to 4.20; 1 study, 145 participants; Analysis 1.1). Evidence from one study (199 participants) suggested that lamotrigine reduces seizure rate in children (RR 2.64, 95% CI 1.59 to 4.38; Analysis 1.1).

We could not calculate responder rates for two studies: Schachter 1995 did not provide the baseline seizure counts, and the primary endpoint of Piña‐Garza 2008 was exit due to treatment failure.

Secondary outcomes

Treatment withdrawal

Fourteen studies (1806 participants) reported treatment withdrawal. Add‐on lamotrigine compared to add‐on placebo probably has little or no effect on treatment withdrawal (RR 1.11, 95% CI 0.91 to 1.37; moderate‐certainty evidence; Analysis 2.1).

2.1. Analysis.

Comparison 2: Add‐on lamotrigine versus add‐on placebo, Outcome 1: Treatment withdrawal

In the parallel studies in adults, 136 participants withdrew from the lamotrigine group and 77 withdrew from the control group (Baulac 2010; Matsuo 1993; Naritoku 2007; Schachter 1995). In the parallel study in children, 14 participants withdrew from the lamotrigine group and 18 withdrew from the control group (Duchowny 1999). In the cross‐over studies in adults, 19 participants withdrew from the lamotrigine group and 10 withdrew from the control group (Binnie 1989; Boas 1996; Jawad 1989; Loiseau 1990; Messenheimer 1994; Schapel 1993; Schmidt 1993; Smith 1993). In the parallel study in infants, 11 infants withdrew from the control group and 16 withdrew from the control group (Piña‐Garza 2008). See Analysis 2.1 for the meta‐analyses of these data.

Insufficient data were available to undertake the planned dose‐response subgroup analyses.

Adverse effects

In addition to ataxia, dizziness, fatigue, and nausea, some studies reported somnolence, diplopia, and headache among the five most common adverse effects. Lamotrigine compared to placebo probably increases the risk of ataxia (RR 3.34, 99% CI 2.01 to 5.55; 12 studies, 1525 participants; Analysis 3.1), dizziness (RR 1.76, 99% CI 1.28 to 2.43; 13 studies, 1768 participants, Analysis 3.2), nausea (RR 1.81, 99% CI 1.22 to 2.68; 12 studies, 1486 participants, Analysis 3.4), and diplopia (RR 3.79, 99% CI 2.15 to 6.68; 3 studies, 944 participants; Analysis 3.6). Lamotrigine probably has little or no effect on the risk of fatigue (RR 0.82; 99% CI 0.55 to 1.22; 12 studies, 1552 participants, Analysis 3.3). The evidence for these five outcomes was of moderate certainty. Our analyses also suggested that lamotrigine has little or no effect on somnolence (RR 1.39, 99% CI 0.96 to 2.00; 13 studies, 1768 participants; Analysis 3.5) or headache (RR 1.13, 99% CI 0.88 to 1.45; 5 studies, 1386 participants; Analysis 3.7).

3.1. Analysis.

Comparison 3: Add‐on lamotrigine versus add‐on placebo, Outcome 1: Ataxia

3.2. Analysis.

Comparison 3: Add‐on lamotrigine versus add‐on placebo, Outcome 2: Dizziness

3.4. Analysis.

Comparison 3: Add‐on lamotrigine versus add‐on placebo, Outcome 4: Nausea

3.6. Analysis.

Comparison 3: Add‐on lamotrigine versus add‐on placebo, Outcome 6: Diplopia

3.3. Analysis.

Comparison 3: Add‐on lamotrigine versus add‐on placebo, Outcome 3: Fatigue

3.5. Analysis.

Comparison 3: Add‐on lamotrigine versus add‐on placebo, Outcome 5: Somnolence

3.7. Analysis.

Comparison 3: Add‐on lamotrigine versus add‐on placebo, Outcome 7: Headache

Cognitive effects and quality of life

Two studies (54 participants) incorporated measures of cognitive functions (Banks 1991; Smith 1993). They reported no differences between the treatment groups in any of the tests used. However, participants receiving lamotrigine showed a marginal reduction in general cerebral efficiency as assessed by the third segment of the Stroop colour word test, which is a test of concentration and distractibility (Table 2). Meta‐analysis of the two studies was not possible because each measured a different cognitive function.

1. Cognitive outcomes.

Outcome	Study	Number tested	Mean score (on a 0–100 scale; higher is better)
			Lamotrigine	Placebo
Stroop time	Smith 1993	41	93.98	98.39
Stroop error	Smith 1993	44	2.18	2.41
Stroop colour word (total score)	Banks 1991	10	32.4 (SD10.9)	35.6 (SD 9.42)
Number cancellation: task AC	Smith 1993	44	51.36	49.7
Number cancellation: task AE	Smith 1993	43	3.6	3.04
Number cancellation: task BC	Smith 1993	42	48.21	48.54
Number cancellation: task C	Smith 1993	42	38.19	39.29
Critical flicker fusion	Smith 1993	40	30.44	30.37
Choice reaction time	Smith 1993	40	0.675	0.669
Digit symbol (scaled score)	Banks 1991	10	5 (SD 2.45)	6.6 (SD 2.71)
Rey complex figure recall percentile	Banks 1991	10	22 (SD 17.51)	30.5 (SD 27.33)
Trail making part B percentile	Banks 1991	10	26 (SD 30.35)	30.5 (SD 32.09)

SD: standard deviation.

Table 3 shows the results of the HRQOL assessments. Smith 1993 (54 participants) incorporated an HRQOL measure containing previously validated measures of physical, social, and psychological functioning and a novel measure of seizure severity. There were no differences between the treatment groups in the physical and social components. Participants in the lamotrigine group reported greater improvements on the seizure severity scale compared to the participants in the control group.

2. Health‐related quality of life outcomes (Smith 1993).

Outcome category	Outcome	Number tested	Mean score (on a 0–100 scale; higher is better)		Clinical relevance
			Lamotrigine	Placebo
Psychological	Depression	54	4.24	4.26	No significant difference
	Happiness	51	3.8	1.96	Higher scores in lamotrigine group; P = 0.003
	Mood	50	24.36	26.8	No significant difference
	Self‐esteem	50	30.06	29.16	No significant difference
	Mastery	50	20.02	18.78	Higher scores in lamotrigine group; P = 0.003
	Anxiety	54	6.87	6.83	No significant difference
Physical (Nottingham Health Profile)	Energy	53	0.68	0.68	No significant difference
	Pain	53	0.6	0.69	No significant difference
	Emotional reaction	53	1.96	1.96	No significant difference
	Sleep	53	0.89	0.76	No significant difference
	Social isolation	53	0.92	0.94	No significant difference
	Physical mobility	53	0.96	0.91	No significant difference
Seizure severity	Percept	53	25.19	25.47	No significant difference
	Ictal	53	19.47	20.53	Less severe seizures in lamotrigine group; P = 0.017
	Caregivers	53	20.35	21.80	Less severe seizures in lamotrigine group; P = 0.035

Discussion

Summary of main results

We identified no additional eligible studies in this update of the review.

In all studies except Schmidt 1993, the baseline phase ranged from four to 12 weeks and the treatment phase from eight to 36 weeks. Eleven of the 14 included trials described adequate methods of concealment of randomisation, and only four described adequate blinding. Thirteen studies were sponsored by the manufacturer of lamotrigine.

Lamotrigine is probably more effective than placebo for achieving a 50% or greater reduction in seizure frequency when added to conventional ASMs in people with drug‐resistant focal epilepsy. We were unable to examine dose effects in planned subgroup analyses, but the results from Matsuo 1993 suggested increased efficacy with an increased dose. One study recruited only children (aged two to 16 years; Duchowny 1999), and one study recruited infants (Piña‐Garza 2008). We have no evidence from this review to indicate whether lamotrigine is more or less effective in infants and children than in adults.

For a drug to be an attractive option, it needs to have a favourable adverse‐effect profile, have little effect on cognition, and have positive effects on QOL, in addition to reducing seizures. Our findings show that certain adverse effects (ataxia, dizziness, diplopia, and nausea) are probably more likely to occur with lamotrigine. Researchers should routinely and regularly enquire about adverse effects using a standardised check‐list thesaurus rather than only recording the adverse effects volunteered by the patient. Lamotrigine compared with placebo probably has little or no effect on withdrawal from treatment.

The results of this review apply only to the add‐on use of lamotrigine compared with placebo (and not compared with other add‐on drugs). It is important to close this gap in the evidence to aid clinicians who have to make evidence‐based choices between an ever‐increasing number of ASMs. Indirect comparisons can be made with results of other reviews, but such comparisons require cautious interpretation.

Overall completeness and applicability of evidence

Only two studies evaluated the effects of add‐on lamotrigine therapy on cognition and QOL. The results of these studies suggested that lamotrigine was probably not associated with any significant cognitive decline. Regardless, the limited data available and heterogenous measurement scales for QOL precluded us from drawing any conclusions about the effects of add‐on lamotrigine on these outcomes.

Caution is required when translating the results of clinical trials into everyday practice. The individuals in trials are a highly selected population who may be better motivated, and are closely followed and monitored. Participants who are uncooperative and non‐compliant, who are likely to have adverse effects and fewer benefits, are excluded. The results of this review cannot be extrapolated to people with generalised epilepsies, about whom there is a great paucity of data.

We cannot ascertain the safety of lamotrigine during pregnancy and lactation from our findings.

The duration of the studies included in this review was insufficient to detect changes in cognition, social problems, or long‐term adverse effects. There is a need for trials that recruit more participants and that use reliable, validated measures and longer follow‐up. This review was not sufficiently sensitive to detect rare but serious adverse effects of ASMs, such as psychosis, Steven Johnson's syndrome, or aplastic anaemia. Rare phenomena such as habituation and tolerance may not be evident in short‐term trials.

This review did not examine the economic aspects of lamotrigine therapy.

Quality of the evidence

We judged 11 studies at low overall risk of bias and three at unclear overall risk of bias, mainly due to the lack of information regarding study design. Only four trials reported effective methods of blinding. We rated 13 studies at low risk of bias for incomplete outcome data, as they undertook ITT analyses. As a result, we did not downgrade the certainty of the evidence for risk of bias.

We judged the evidence for all outcomes included in the summary of findings table as moderate certainty. The main reason for downgrading was the low number of reported events (below the optimal information size). For one secondary outcome (dizziness), we downgraded the certainty of the evidence for inconsistency (statistical heterogeneity).

Potential biases in the review process

Our choice of primary outcome (50% or greater reduction in seizure frequency) as a measure of efficacy could be criticised, as seizure freedom would be a more relevant clinical measure. However, seizure freedom is rarely achieved in studies that recruit people with drug‐resistant epilepsy.

Although we requested protocols for all studies, most were conducted at a time when trial preregistration was uncommon. This could lead to potential bias through omitted information.

All studies but one were sponsored by GlaxoSmithKline, the manufacturers of lamotrigine, and this could be a potential source of bias.

Agreements and disagreements with other studies or reviews

The findings and conclusions of our update regarding the efficacy and safety profile of lamotrigine have not changed since the previous version of this review, which supported the use of add‐on lamotrigine as an effective treatment for drug‐resistant focal epilepsy (Panebianco 2020).

Authors' conclusions

Implications for practice.

In people with drug‐resistant focal epilepsy, add‐on lamotrigine is probably effective for reducing seizure frequency. The lowest daily dose tested in the trials included in this review was 75 mg for people on sodium valproate monotherapy, 100 mg in the balanced group receiving enzyme inducing antiseizure medications (ASMs) and valproate, and 200 mg in people receiving enzyme‐inducing ASMs.

However, the trials reviewed were of relatively short duration, and some trials had relatively few participants.

Implications for research.

Further evaluation of lamotrigine is required to assess the following effects in the long term on larger samples of the population.

1. Effects on seizures

2. Adverse effects

3. Effects on cognition

4. Effects on quality of life

5. Health economic effects

Future research should cover the following scenarios.

1. Add‐on lamotrigine compared to other add‐on treatments in drug‐resistant focal epilepsy

2. Lamotrigine for childhood and generalised epilepsies

3. Lamotrigine as monotherapy compared with standard antiseizure medications in focal epilepsy and in generalised epilepsy

What's new

Date	Event	Description
11 December 2023	New search has been performed	Searches updated 3 October 2022, no new relevant studies were identified.
11 December 2023	New citation required but conclusions have not changed	Conclusions are unchanged.

History

Protocol first published: Issue 1, 2000
Review first published: Issue 3, 2000

Date	Event	Description
9 March 2020	New citation required but conclusions have not changed	Conclusions remain unchanged. The term 'partial' has been replaced by 'focal', in accordance with the most recent classification of epilepsies of the International League Against Epilepsy (Scheffer 2017).
9 March 2020	New search has been performed	Searches updated 09 March 2020; no new studies were identified.
28 May 2015	New search has been performed	Searches updated 28 May 2015. No new relevant studies were identified.
28 May 2015	New citation required but conclusions have not changed	Conclusions remain unchanged.
6 January 2010	New search has been performed	Searches updated 6th January 2010. Two new studies have been included (Naritoku 2007 and Piña-Garza 2008); the conclusions are unchanged.
10 September 2008	Amended	Converted to new review format.
25 April 2007	New search has been performed	Searches updated 25th April 2007. One new conference abstract (Carignani 2006) has been added to the 'Studies Awaiting Classification' section. This will be assessed for inclusion at a later date.
16 November 2005	Amended	We re‐ran our search on 31 March 2005. One new study (Carignani 2004) has been added to the 'studies awaiting assessment' section.

Appendices

Appendix 1. Cochrane Register of Studies (CRS Web)

1. MESH DESCRIPTOR Lamotrigine EXPLODE ALL AND CENTRAL:TARGET

2. (Lamotrigin* OR Elmendos OR Epilepax OR "GW 273293" OR Lamictal OR Lamictin OR Lamitor OR Lamitrin OR Lamogine OR Lamotrine OR LTG):AB,KW,KY,MC,MH,TI AND CENTRAL:TARGET

3. #1 OR #2

4. MESH DESCRIPTOR Epilepsies, Partial EXPLODE ALL AND CENTRAL:TARGET

5. ((partial or focal) and (seizure* or epilep*)):AB,KW,KY,MC,MH,TI AND CENTRAL:TARGET

6. (secondar* and (generalized or generalised) and seizure*):AB,KW,KY,MC,MH,TI AND CENTRAL:TARGET

7. #4 OR #5 OR #6

8. #3 AND #7

9. (monotherap* NOT (adjunct* OR "add‐on" OR "add on" OR adjuvant* OR combination* OR polytherap*)):TI AND CENTRAL:TARGET

10. #8 NOT #9

11. >09/03/2020:CRSCREATED AND CENTRAL:TARGET

12. #10 AND #11

Appendix 2. MEDLINE (Ovid) 1946‐

This strategy includes the Cochrane Highly Sensitive Search Strategy for identifying randomized trials (Lefebvre 2022).

1. exp Lamotrigine/

2. (Lamotrigin$ or Elmendos or Epilepax or "GW 273293" or Lamictal or Lamictin or Lamitor or Lamitrin or Lamogine or Lamotrine or LTG).mp.

3. 1 or 2

4. exp Epilepsies, Partial/

5. ((partial or focal) and (seizure$ or epilep$)).mp.

6. (secondar$ and generali?ed and seizure$).mp.

7. 4 or 5 or 6

8. exp controlled clinical trial/ or (randomi?ed or placebo or randomly).ab.

9. clinical trials as topic.sh.

10. trial.ti.

11. 8 or 9 or 10

12. exp animals/ not humans.sh.

13. 11 not 12

14. 3 and 7 and 13

15. (monotherap$ not (adjunct$ or "add‐on" or "add on" or adjuvant$ or combination$ or polytherap$)).ti.

16. 14 not 15

17. limit 16 to ed=20200306‐20221003

18. 16 not (1$ or 2$).ed.

19. 18 and (2020$ or 2021$ or 2022$).dt.

20. 17 or 19

21. remove duplicates from 20

Data and analyses

Comparison 1. Add‐on lamotrigine versus add‐on placebo.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 ≥ 50% reduction in seizure frequency (intention‐to‐treat analysis)	12		Risk Ratio (M‐H, Fixed, 95% CI)	Subtotals only
1.1.1 Cross‐over studies	8	382	Risk Ratio (M‐H, Fixed, 95% CI)	2.58 [1.44, 4.61]
1.1.2 Parallel‐group studies (lamotrigine 300 mg)	1	144	Risk Ratio (M‐H, Fixed, 95% CI)	1.23 [0.57, 2.67]
1.1.3 Parallel‐group studies (lamotrigine 500 mg)	1	145	Risk Ratio (M‐H, Fixed, 95% CI)	2.13 [1.08, 4.20]
1.1.4 Parallel‐group studies (children)	1	199	Risk Ratio (M‐H, Fixed, 95% CI)	2.64 [1.59, 4.38]
1.1.5 Parallel‐group studies (adults; lamotrigine extended‐release)	1	243	Risk Ratio (M‐H, Fixed, 95% CI)	1.70 [1.16, 2.50]
1.1.6 Parallel‐group studies (lamotrigine 300 mg to 600 mg)	1	282	Risk Ratio (M‐H, Fixed, 95% CI)	1.13 [0.74, 1.75]
1.1.7 Any dose of lamotrigine, adults or children	12	1322	Risk Ratio (M‐H, Fixed, 95% CI)	1.80 [1.45, 2.23]
1.2 ≥ 50% reduction in seizure frequency (worst‐case analysis)	12	1322	Risk Ratio (M‐H, Fixed, 95% CI)	0.97 [0.82, 1.15]
1.3 ≥ 50% reduction in seizure frequency (best‐case analysis)	12	1322	Risk Ratio (M‐H, Fixed, 95% CI)	2.88 [2.36, 3.50]

Comparison 2. Add‐on lamotrigine versus add‐on placebo.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Treatment withdrawal	14	1806	Risk Ratio (M‐H, Fixed, 95% CI)	1.11 [0.91, 1.37]
2.1.1 Parallel‐group studies (adults)	4	1187	Risk Ratio (M‐H, Fixed, 95% CI)	1.17 [0.90, 1.50]
2.1.2 Parallel‐group studies (children)	1	199	Risk Ratio (M‐H, Fixed, 95% CI)	0.80 [0.42, 1.52]
2.1.3 Cross‐over studies	8	382	Risk Ratio (M‐H, Fixed, 95% CI)	1.82 [0.89, 3.72]
2.1.4 Parallel‐group studies (infants)	1	38	Risk Ratio (M‐H, Fixed, 95% CI)	0.69 [0.45, 1.06]

Comparison 3. Add‐on lamotrigine versus add‐on placebo.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Ataxia	12	1525	Risk Ratio (M‐H, Fixed, 99% CI)	3.34 [2.01, 5.55]
3.1.1 Parallel‐group studies (adults)	3	944	Risk Ratio (M‐H, Fixed, 99% CI)	3.40 [1.67, 6.90]
3.1.2 Parallel‐group studies (children)	1	199	Risk Ratio (M‐H, Fixed, 99% CI)	5.15 [0.72, 36.64]
3.1.3 Cross‐over studies	8	382	Risk Ratio (M‐H, Fixed, 99% CI)	2.98 [1.38, 6.41]
3.2 Dizziness	13	1768	Risk Ratio (M‐H, Fixed, 99% CI)	1.76 [1.28, 2.43]
3.2.1 Parallel‐group studies (adults)	4	1187	Risk Ratio (M‐H, Fixed, 99% CI)	1.72 [1.11, 2.67]
3.2.2 Parallel‐group studies (children)	1	199	Risk Ratio (M‐H, Fixed, 99% CI)	4.33 [1.27, 14.79]
3.2.3 Cross‐over studies	8	382	Risk Ratio (M‐H, Fixed, 99% CI)	1.41 [0.83, 2.38]
3.3 Fatigue	12	1552	Risk Ratio (M‐H, Fixed, 99% CI)	0.82 [0.55, 1.22]
3.3.1 Parallel‐group studies (adults)	3	971	Risk Ratio (M‐H, Fixed, 99% CI)	0.81 [0.46, 1.42]
3.3.2 Parallel‐group studies (children)	1	199	Risk Ratio (M‐H, Fixed, 99% CI)	1.89 [0.54, 6.63]
3.3.3 Cross‐over studies	8	382	Risk Ratio (M‐H, Fixed, 99% CI)	0.65 [0.34, 1.23]
3.4 Nausea	12	1486	Risk Ratio (M‐H, Fixed, 99% CI)	1.81 [1.22, 2.68]
3.4.1 Parallel‐group studies (adults)	3	905	Risk Ratio (M‐H, Fixed, 99% CI)	1.68 [1.02, 2.78]
3.4.2 Parallel‐group studies (children)	1	199	Risk Ratio (M‐H, Fixed, 99% CI)	5.67 [0.81, 39.69]
3.4.3 Cross‐over studies	8	382	Risk Ratio (M‐H, Fixed, 99% CI)	1.67 [0.85, 3.29]
3.5 Somnolence	13	1768	Risk Ratio (M‐H, Fixed, 99% CI)	1.39 [0.96, 2.00]
3.5.1 Parallel‐group studies (adults)	4	1187	Risk Ratio (M‐H, Fixed, 99% CI)	1.58 [0.93, 2.68]
3.5.2 Parallel‐group studies (children)	1	199	Risk Ratio (M‐H, Fixed, 99% CI)	1.37 [0.67, 2.81]
3.5.3 Cross‐over studies	8	382	Risk Ratio (M‐H, Fixed, 99% CI)	1.06 [0.51, 2.17]
3.6 Diplopia	3	944	Risk Ratio (M‐H, Fixed, 99% CI)	3.79 [2.15, 6.68]
3.7 Headache	5	1386	Risk Ratio (M‐H, Fixed, 99% CI)	1.13 [0.88, 1.45]

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Baulac 2010.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, parallel‐group study Number of treatment arms: 3 (1 placebo, 1 lamotrigine, 1 pregabalin) Length of baseline period: 6 weeks Length of treatment period: 17 weeks, which included initial 5 weeks dosage titration for lamotrigine and 6 weeks maintenance at 300 mg/day and additional treatment period of 6 weeks with dose escalation to 400 mg/day for those with continuing seizures. Double‐blind treatment period was followed by an open‐label study or a 2‐week taper phase.
Participants	Setting: 97 centres in Europe, Canada, and Australia Number of participants: 546 people screened, 434 randomised Sex (males/females): lamotrigine group: 54.6%/45.4%; placebo group: 39.3%/60.7% Age: lamotrigine group: mean 39.4 years; placebo group: mean 39.1 years History of epilepsy: not reported Inclusion criteria Age ≥ 18 years Bodyweight ≥ 40 kg Diagnosis of focal seizures (as defined by the International League Against Epilepsy Classification of Seizures) Minimum seizure frequency of 4 focal seizures during the 6‐week baseline period and no 28‐day period free of focal seizure, despite treatment with ≥ 3 AEDs from ≥ 2 different AED classes, each at or above the lowest recommended dose or the lowest adequate plasma concentration given for a minimum of 3 months Exclusion criteria Previous treatment with pregabalin Previous treatment with lamotrigine within 6 months before entering baseline History of rash with lamotrigine Previous treatment with valproic acid products within 2 months of baseline Previous treatment with gabapentin, felbamate, or vigabatrin < 6 weeks prior to screening History of status epilepticus in the last year, significant psychiatric disorder, or use of concomitant medication that could interfere with response to study medications or affect seizure frequency Pregnancy, planned pregnancy, or lactation
Interventions	Group I (n = 141): placebo Group II (n = 141): lamotrigine 300 mg/day after dose titration over 5 weeks, and if seizures occurred during 6‐week maintenance, further dose escalation to 400 mg/day from week 12 to 17 Group III (n = 152): pregabalin. The participants randomised to pregabalin were not included in this review.
Outcomes	Seizure frequency Adverse events, including changes in physical and neurologic examinations, 12‐lead ECGs, and clinical laboratory tests (haematology, blood chemistry, pregnancy, and urinalysis).
Notes	This study was sponsored by Pfizer Inc.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation not specified.
Allocation concealment (selection bias)	Unclear risk	No details in the publication.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No details regarding blinding of participants, study personnel, and outcome assessors. Regarding the medications, blinding was maintained by administering the same numbers of capsules per day per group.
Incomplete outcome data (attrition bias)	Low risk	35 withdrew from placebo group and 40 from lamotrigine group. The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	Protocol unavailable to check a priori outcomes, but it appears all expected and prespecified outcomes were reported.
Other bias	Unclear risk	Responder rates reported as percentages without numbers. We contacted a trial author to ask about the actual number of responders in each group, but we have not obtained this information to date.

Binnie 1989.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 8 weeks Length of treatment period: treatment I and II: 12 weeks each; washout: 6 weeks, including taper period
Participants	Setting: single site in the Netherlands Number of participants: 34 adults (16 were randomised to lamotrigine and 18 to placebo during the first treatment phase). Sex (males/females): not reported Age: mean 31.1 years (SD 10.26), range 16–51 years History of epilepsy: age of onset: mean 14.3 years (SD 10.7), range 1 to 40 years; duration: mean 22.8 years (SD 11), range 6 to 49.5 years Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures ≤ 4 other AEDs Exclusion criteria: not reported
Interventions	Lamotrigine 200 mg/day (participants on valproate received lower doses) versus placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated sequentially‐numbered sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Low risk	Participants and parents were blinded. An unblinded investigator with knowledge of the medication and plasma concentrations instructed the blinded investigators about dispensing the trial medications. Identical tablets and packaging used.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. No participant withdrew from the study during the first treatment phase.
Selective reporting (reporting bias)	Low risk	All outcomes stated in methods section of paper were reported in the results. There was no protocol available to check a priori outcomes.
Other bias	Low risk	No other sources of bias identified.

Boas 1996.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 12 weeks Length of treatment period: treatment I: 12 weeks; washout I: 4 weeks; treatment II: 12 weeks; washout II: 4 weeks
Participants	Setting: 4 sites in Denmark Number of participants: 30 allocated to lamotrigine and 26 to placebo during the first treatment phase Sex (males/females): 27 men/29 women Age: range 16–65 years History of epilepsy: not reported Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures ≤ 4 other AEDs Exclusion criteria: not reported
Interventions	Add‐on lamotrigine (dose varied from 75 mg/day to 400 mg/day; participants on valproate received lower doses) versus add‐on placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated sequentially‐numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No details provided regarding blinding of participants, study personnel, and outcome assessors. All treatments (tablets) and packaging were identical. Prepacked coded medication was dispensed by pharmacy.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. 10 participants withdrew from the study (8 randomised to lamotrigine and 2 to placebo). The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	All outcomes stated in methods section of paper were reported in the results. There was no protocol available to check a priori outcomes.
Other bias	Low risk	No other sources of bias identified.

Duchowny 1999.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, parallel‐group study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 8 weeks Length of treatment period: treatment phase: 18 weeks (including 6‐week titration); follow‐up: 1–6 weeks, including 1‐week taper
Participants	Setting: 40 sites in USA and France Number of participants: 199 children, 98 allocated to lamotrigine and 101 to placebo Sex (males/females): 103 boys/96 girls Age: range 2–16 years (27% aged < 6 years, 60% aged 6–12 years, 11% aged > 12 years) History of epilepsy: not reported Inclusion criteria Age ≥ 2 years and < 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures ≤ 2 other AEDs Exclusion criteria: not reported
Interventions	Group I: add‐on lamotrigine. Median dose ranged from 2.7 mg/kg/day to 12.9 mg/kg/day, depending upon concurrent use of other AEDs. Participants on valproate received lower doses. Group II: add‐on placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were randomised with a blocked randomisation scheme to treatment with add‐on lamotrigine or matched placebo in bottles labelled with pregenerated participant numbers.
Blinding (performance bias and detection bias) All outcomes	Low risk	Treatment assignments were unknown to all study‐site personnel, patients and sponsors. Lamotrigine and matching placebo were provided as berry‐flavoured, chewable, dispersible caplets or tablets in strengths of 5 mg, 25 mg, and 100 mg.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. 2 enroled participants withdrew before randomisation. 14 participants allocated to lamotrigine and 18 participants allocated to placebo withdrew during the treatment phase. The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	All outcomes stated in methods section of paper were reported in the results. There was no protocol available to check a priori outcomes.
Other bias	Low risk	No other sources of bias identified.

Jawad 1989.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 8 weeks Length of treatment period: treatment I: 12 weeks; washout I: 6 weeks; treatment II: 12 weeks; washout II: 6 weeks
Participants	Setting: single site in the UK Number of participants: 12 allocated to lamotrigine and 12 to placebo in the first treatment phase Sex (males/females): not reported Age: range 16–60 years History of epilepsy: not reported Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizure) ≤ 2 other AEDs Exclusion criteria: not reported
Interventions	Add‐on lamotrigine (median daily dose was 250 mg; participants on valproate received lower doses; unblinded investigator wrote prescriptions based on plasma concentration) versus add‐on placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated by sequentially numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No details provided regarding blinding of participants, study personnel, and outcome assessors. Identical tablets and packaging used.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. One participant who was allocated to lamotrigine withdrew from the study (the reason for exclusion was reported) and none withdrew from the placebo group.
Selective reporting (reporting bias)	Low risk	All outcomes stated in methods section of paper were reported in the results. There was no protocol available to check a priori outcomes.
Other bias	Low risk	No other sources of bias identified.

Loiseau 1990.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 4 weeks Length of treatment period: treatment I: 8 weeks; washout I: 4 weeks; treatment II: 8 weeks; washout II: 4 weeks
Participants	Setting: single site in France Number of participants: 11 randomised to lamotrigine and 14 to placebo in the first treatment phase Sex (males/females): not reported Age: mean 34.2 years (SD 12.41), range 20–54 years History of epilepsy: duration: mean 17.4 years (SD 10.81); range 3–45 years Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizure) ≤ 2 other AEDs Exclusion criteria: not reported
Interventions	Add‐on lamotrigine (median daily dose was 300 mg; participants on valproate received lower doses) versus add‐on placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated by sequentially numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Low risk	Neurologists, participants, and parents were blinded. Investigators were blinded. All treatments (tablets) and packaging were identical. Prepacked coded medication dispensed by pharmacy.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. 2 participants withdrew from the study (1 receiving lamotrigine and 1 receiving placebo). The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	All outcomes stated in methods section of paper were reported in the results. There was no protocol available to check a priori outcomes.
Other bias	Low risk	No other sources of bias identified.

Matsuo 1993.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, parallel‐group study Number of treatment arms: 3 (1 placebo, 1 lamotrigine 300 mg, and 1 lamotrigine 500 mg) Length of baseline period: 12 weeks Length of treatment period: treatment phase: 24 weeks; taper and follow‐up: 3 weeks.
Participants	Setting: multicentre study from the USA Number of participants: 216 adults, 73 randomised to placebo, 71 to lamotrigine 300 mg/day, 72 to lamotrigine 500 mg/day Sex (males/females): 67 men/149 women Age: mean 33 years, range 18–63 years History of epilepsy: duration: mean 21.9 years; age at onset: mean 11 years Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizure ≤ 3 other AEDs Exclusion criteria Use of valproate
Interventions	Group I: add‐on lamotrigine 300 mg Groups II: add‐on lamotrigine 500 mg Group III: add‐on placebo.
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Randomisation concealment: allocated sequentially numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No details provided regarding blinding of participants, study personnel, and outcome assessors. Identical tablets and packaging used.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. 25 participants withdrew from the study (7 receiving lamotrigine 300 mg, 12 receiving lamotrigine 500 mg, and 6 receiving placebo). The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	Protocol unavailable to check a priori outcomes, but it appears all expected and prespecified outcomes were reported.
Other bias	Low risk	No other sources of bias identified.

Messenheimer 1994.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 8 weeks Length of treatment period: treatment I 14 weeks (including 2 weeks blinded tapering); follow‐up period for treatment I: 3 weeks; treatment B: 14 weeks (including 2 weeks blinded tapering); washout: 4 weeks
Participants	Setting: multicentre study from the USA Number of participants: 98 adults, 46 randomised to lamotrigine, 52 to placebo in the first treatment phase Sex (males/females): 46 men/52 women Age: mean 35 years, range 18–64 years History of epilepsy: age at onset: mean 12 years; duration: mean 23.1 years Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizure) ≤ 3 other AEDs Exclusion criteria Use of valproate
Interventions	Add‐on lamotrigine (median dose 400 mg/day) versus add‐on placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated by sequentially‐numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	Investigators were blinded. No more information provided regarding blinding of neurologists, participants, and parents. All treatments (tablets) and packaging were identical. Prepacked coded medication dispensed by pharmacy.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. 6 participants withdrew from the study (2 receiving lamotrigine and 4 receiving placebo). The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	All outcomes stated in methods section of paper were reported in the results. There was no protocol available to check a priori outcomes.
Other bias	Low risk	No other sources of bias identified.

Naritoku 2007.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, parallel‐group study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 8 weeks Length of treatment period: screening phase of up to 2 weeks; 7‐week, double‐blind escalation phase during which lamotrigine XR (Extended Release) was introduced and titrated to its target dose; and 12‐week, double‐blind maintenance phase during which dosage of study medication and concomitant AED were maintained
Participants	Setting: international multicentre study (North and South America, Europe, and Asia) Number of participants: 244 adults, 121 randomised to lamotrigine, 123 to placebo Sex (males/females): lamotrigine group: 47%/53%; placebo group: 53%/47% Age: range 16–16 years; lamotrigine group: mean 35.8 years (SD 12.7); placebo group: mean 37.5 years (SD 14.4) History of epilepsy: ≥ 8 focal seizures in 8 weeks with ≥ 1 focal seizure during each 4‐week period of the baseline phase Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures) Stable regimen of 1 or 2 AEDs for ≥ 4 weeks before starting the baseline phase Exclusion criteria Primary generalised seizures Status epilepticus during or within 24 weeks before the start of the baseline phase Chronic treatment with ≥ 3 AEDs Current or previous use of lamotrigine Current use of felbamate or adherence to a ketogenic diet Pregnancy or planned pregnancy during the study or within 3 weeks after the last dose of study medication
Interventions	Group I: lamotrigine XR (Extended Release); dosage escalated gradually up to 200 mg/day in participants receiving valproate, 300 mg/day in those receiving valproate and an enzyme inducing AED, and up to 500 mg/day in those receiving enzyme inducing AEDs without valproate. Group II: identical placebo
Outcomes	Seizure frequency Adverse events Withdrawals from study Participants in the USA had the following additional assessments Profile of Mood States (POMS) Center for Epidemiologic Studies‐Depression Scale (CES‐D) Research version of the Neurological Disorders Depression Inventory‐Epilepsy (NDDI‐E) Quality of Life in Epilepsy‐31‐P (QOLIE‐31‐P) Liverpool Adverse Experience Profile (AEP) Seizure Severity Questionnaire (SSQ) Epworth Sleepiness Scale (ESS)
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation not specified.
Allocation concealment (selection bias)	Unclear risk	No details in the publication.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No information provided.
Incomplete outcome data (attrition bias)	Low risk	24 participants withdrew from treatment group and 16 from placebo group. The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	There was no protocol available to check a priori outcomes, but it appears all expected and prespecified outcomes were reported.
Other bias	Low risk	No other sources of bias identified.

Piña‐Garza 2008.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of treatment period: responder‐enriched design in which all participants received adjunctive lamotrigine during an open‐label phase (wherein dose was escalated to achieve optimal response); those who had a ≥ 40% reduction in the frequency of focal seizures during the last 4 weeks of the optimisation period were randomly assigned to double‐blind treatment for up to 8 weeks with continued lamotrigine or placebo.
Participants	Setting: international multicentre study (North and South America, Europe, and Asia) Number of participants: 57 participants, 38 randomised to lamotrigine and 19 to placebo Sex (males/females): not reported Age: lamotrigine group: median 13.5 months; placebo group: median 14.2 months History of epilepsy: age at onset: median 3 months; duration: median 9.1 months in lamotrigine group and 8.5 months in placebo group Inclusion criteria Age 1–24 months Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures) with ≥ 4 focal seizures (with or without generalisation) per month Uncontrolled by 1 AED ≥ 40% reduction in seizure frequency during an initial open label phase Exclusion criteria Progressive myoclonic epilepsy Progressive neurologic disease, seizures unrelated to epilepsy or resulting from drug withdrawal Use of felbamate, adrenocorticotropic hormone, previous use of lamotrigine 2 AEDs as maintenance treatment Hepatic dysfunction Functioning vagus nerve stimulation Ketogenic diet Participants were withdrawn from the study if they met any of the following escape criteria. 50% increase in monthly focal seizure frequency compared with seizure frequency during the last 4 weeks of the open‐label optimisation period Doubling of the highest consecutive 2‐day focal seizure count observed during the open‐label optimisation period Onset of a new and more severe seizure type Clinically significant worsening of non‐focal seizures that were also observed during the historical baseline phase or the open‐label optimisation period Need to use any therapeutic intervention in addition to study medication to control seizures Status epilepticus
Interventions	Group I: continued on lamotrigine. The maximum maintenance dose was 5.1 mg/kg/day for infants on non‐enzyme‐inducing AEDs or valproate and 15.6 mg/kg/day for those on enzyme‐inducing AEDs. Group II: lamotrigine dose tapered and changed to placebo
Outcomes	Percentage of participants who had treatment failures during the double‐blind phase Cumulative percentage of participants who met escape criteria as a function of days on double‐blind study medication
Notes	The protocol was amended midway through the study to randomly assign all patients with ≥ 40% reduction in seizure frequency, instead of planned inclusion of participants with 40%–80% reduction in seizure frequency. 43 participants who had more than 80% reduction in seizure frequency before the protocol amendment were not included in the double‐blind study. This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation not specified.
Allocation concealment (selection bias)	Unclear risk	Details not reported in the publication.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No information provided.
Incomplete outcome data (attrition bias)	Low risk	11 patients (8 in the lamotrigine group and 3 in placebo group) completed the double‐blind phase, 25 (9 in the lamotrigine group and 16 in placebo group) met escape criteria, and 2 (both in the lamotrigine group) prematurely withdrew because of protocol violations. The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	All expected and prespecified outcomes were reported. Protocol was not available.
Other bias	Low risk	No other sources of bias identified.

Schachter 1995.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, parallel‐group study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 4 weeks Length of treatment period: treatment phase: 24 weeks; taper and follow‐up: 3 weeks
Participants	Setting: 34 sites in the USA Number of participants: 446 adults, 334 randomised to lamotrigine and 112 to placebo Sex (males/females): 236 men/210 women Age: mean 35 years, range 18–64 years History of epilepsy: duration: mean 21 years; age at onset: median 2 years in the lamotrigine group and 11.5 in the placebo group Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures) ≤ 3 other AEDs Exclusion criteria Concomitant valproate therapy
Interventions	Group I: add‐on lamotrigine (dose up to 500 mg/day) Group II: add‐on placebo
Outcomes	Withdrawals from treatment Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated by sequentially numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Low risk	Neurologists, participants and parents were blinded. Investigators were blinded. Identical tablets and packaging used.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. 73 participants withdrew from the study (53 receiving lamotrigine and 20 receiving placebo). The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	All outcomes stated in methods section of paper were reported in the results. There was no protocol available to check a priori outcomes.
Other bias	Low risk	No other sources of bias identified.

Schapel 1993.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 12 weeks Length of treatment period: treatment I and II: 12 weeks each; washout I and II: 4 weeks each, including 1 week taper
Participants	Setting: multicentre study from Australia Number of participants: 41 participants, 21 randomised to lamotrigine and 20 to placebo Sex (males/females): 21 males/20 females Age: median 28 years, range 17–63 years History of epilepsy: age at onset: mean 10.4 years (SD 9.6) Inclusion criteria Age ≥ 17 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures) ≤ 3 other AEDs Exclusion criteria Valproate monotherapy
Interventions	Add‐on lamotrigine (median daily dose was 300 mg; participants receiving valproate received lower doses) versus add‐on placebo
Outcomes	50% responder rate Withdrawal from study for any reason Adverse effects
Notes	Banks 1991 is linked to this study and investigated cognitive functions: concentration and attention, general cerebral efficiency, and mnestic functions (immediate, short term and new learning ability). This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated by sequentially numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No details provided regarding blinding of participants, study personnel, and outcome assessors. All treatments and packaging were identical. Prepacked coded medication dispensed by pharmacy.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. None withdrew from the study.
Selective reporting (reporting bias)	Low risk	Protocol unavailable, but it appears all expected and prespecified outcomes were reported.
Other bias	Low risk	No other sources of bias identified.

Schmidt 1993.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: not known Length of treatment period: treatment I and II: 12 weeks each, including 2‐week tapering period; washout: 2 weeks
Participants	Setting: single site in Germany Number of participants: 23 adults, 11 randomised to lamotrigine and 12 to placebo in the initial treatment phase Sex (males/females): 11 men/12 women Age: range 16–62 History of epilepsy: not reported Inclusion criteria Age ≥ 16 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures) ≤ 2 other AEDs Exclusion criteria: not reported
Interventions	Add‐on lamotrigine (dosage varied from 50 mg to 450 mg; median dose was 300 mg) versus add‐on placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects
Notes	This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated by sequentially numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	Unclear risk	No details provided regarding blinding of participants and parents. Unblinded investigator wrote prescriptions based on plasma concentration. Identical tablets and packaging were used.
Incomplete outcome data (attrition bias)	Low risk	No participants were excluded from analysis. 1 participant receiving lamotrigine and 9 receiving placebo withdrew from the study. The reasons for exclusion were reported.
Selective reporting (reporting bias)	Low risk	Protocol unavailable, but it appears all expected and prespecified outcomes were reported.
Other bias	Low risk	No other sources of bias identified.

Smith 1993.

Study characteristics
Methods	Study design: double‐blind, placebo‐controlled, randomised, cross‐over study Number of treatment arms: 2 (1 placebo, 1 lamotrigine) Length of baseline period: 4 weeks Length of treatment period: treatment I and II: 18 weeks each; washout: 6 weeks
Participants	Setting: single site in the UK Number of participants: 81 participants, 41 randomised to lamotrigine and 40 to placebo in the initial treatment phase Sex (males/females): 33 men/48 women Age: mean 33.7 years, range 15 to 67 years History of epilepsy: duration: mean 21 years, range 4–45 years; age at onset: mean 11.8 years, range < 1– 52 years Inclusion criteria Age ≥ 15 years Drug‐resistant focal seizures (as defined by the International League Against Epilepsy Classification of Seizures) ≤ 2 other AEDs Exclusion criteria: not reported
Interventions	Add‐on lamotrigine (dose up to 400 mg/day; median daily dose was 300 mg; participants on valproate received lower doses) versus add‐on placebo
Outcomes	50% responder rates Withdrawal from study for any reason Adverse effects HRQOL
Notes	The health‐related quality of life model was completed by 40 to 54 of 81 participants. This study was sponsored by GlaxoSmithKline, the manufacturer of LTG.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random permuted blocks.
Allocation concealment (selection bias)	Low risk	Participants were allocated sequentially numbered, sealed packages containing either lamotrigine or placebo.
Blinding (performance bias and detection bias) All outcomes	High risk	Participants (46/73) and investigators (52/73) were able to identify lamotrigine treatment. Identical tablets and packaging were used. Prepacked coded medication dispensed by pharmacy.
Incomplete outcome data (attrition bias)	Unclear risk	No participants were excluded from analysis. 9 people withdrew from the study (6 receiving lamotrigine and 3 receiving placebo). The reasons for exclusion were reported. Participants who discontinued prematurely did not complete the HRQOL measure at the time of discontinuation, the exclusion of treatment failures may introduce a bias in favour of lamotrigine.
Selective reporting (reporting bias)	Low risk	Protocol unavailable, but it appears all expected and prespecified outcomes were reported.
Other bias	Low risk	No other sources of bias identified.

AED: anti‐epileptic drug; ASM: antiseizure medication; ECG: electrocardiogram; LTG: lamotrigine; HRQOL: health‐related quality of life; SD: standard deviation.

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Berg 2015	Comparative study among therapeutic equivalents of lamotrigine. Not placebo controlled.
Berg 2017	Comparative study among therapeutic equivalents of lamotrigine. Not placebo controlled.
Biton 2010	Ineligible population: participants had primary generalised epilepsy and not focal seizures.
Biton 2013	Ineligible population: participants had uncontrolled focal epilepsy and generalised tonic‐clonic seizures.
Brzakovic 2012	Ineligible population: participants had uncontrolled focal epilepsy and generalised tonic‐clonic seizures.
Carignani 2004	Published as conference abstract: details of the methods and results are not available.
Carignani 2006	Published as conference abstract: details of the methods and results are not available.
Chung 2009	Comparative study of lamotrigine versus topiramate. Not placebo controlled.
Contin 2016	Comparative study of therapeutic equivalents of lamotrigine. Not placebo controlled.
Cramer 2013	Not an RCT.
Cutillo 2021	Not an RCT.
French 2012	Not an RCT.
Frith 2015	Ineligible population: participants did not have drug‐resistant epilepsy.
Girolineto 2012	Comparative study among therapeutic equivalents of lamotrigine. Not placebo controlled.
Hammer 2008	Published as conference abstract: details of the methods and results are not available.
Hartung 2012	Ineligible population: participants had epilepsy, migraine, pain, and psychiatric disorders.
Helmstaedter 2013	Participants had all epileptic types. Lacosamide as add‐on for epilepsy and in comparison with lamotrigine and topiramate.
IRCT2013021211560N3 2013	Ineligible population: participants did not have drug‐resistant epilepsy.
Ji 2021	Not an RCT.
Kang 2012	Ineligible population: participants had all epileptic types.
Lee 2018	Ineligible population: participants had all epileptic types.
Marson 2021a	No lamotrigine in add‐on.
Marson 2021b	No lamotrigine in add‐on.
Mintzer 2018	Not an RCT (post‐hoc analyses).
Montouris 2007	Published as conference abstract: details of the methods and results are not available.
NCT00208520	Ineligible population: participants did not have drug‐resistant epilepsy.
NCT00292461	Details of the methods and results are not available.
NCT00807989	Details of the methods and results are not available.
NCT01891890	Details of the methods and results are not available.
NCT02100644	Not an RCT.
NCT02429596	Comparative study among therapeutic equivalents of lamotrigine. Not placebo controlled.
Ohtahara 2008	Comparative study of lamotrigine and zonisamide. Not placebo controlled.
Premoli 2017	No outcome measures.
Privitera 2016	Comparative study among therapeutic equivalents of lamotrigine. Not placebo controlled.
Sander 1990	Study of institutionalised people with severe epilepsy. Participants had all epileptic types.
Semah 2014	No lamotrigine in add‐on.
Sethi 2002	Ineligible population: participants did not have drug‐resistant epilepsy.
Shinnar 2015	Ineligible population: participants had primary generalised epilepsy and not focal seizures.
Stolarek 1994	Details of the results are not available.
Thangaratinam 2018	Ineligible population: participants had all epileptic types.
Ting 2015	Comparative study among therapeutic equivalents of lamotrigine. Not placebo controlled.
Tomson 2012	Published as conference abstract: details of the methods and results are not available.
Tomson 2013	Not an RCT.
Veendrick‐Meekes 2000	Published as conference abstract: details of the methods and results are not available.
Wu 2018	Ineligible population: participants had all epileptic types.
Yamamoto 2012	Not an RCT.

RCT: randomised controlled trial.

Characteristics of ongoing studies [ordered by study ID]

NCT03689114.

Study name	Efficacy and tolerability of low vs. standard daily doses of antiepileptic drugs in newly diagnosed, previously untreated epilepsy (STANDLOW)
Methods	Multicentre, randomised, single‐blind, parallel‐group trial
Participants	Aged 18 years or older with focal untreated epilepsy
Interventions	Low‐ versus standard‐dose lamotrigine
Outcomes	Primary outcomes Treatment failure (time frame: 12 months): proportion of participants experiencing a treatment failure motivated by the need to change the assigned dose or the assigned drug for seizure relapse during the follow‐up. Secondary outcomes Drug‐related adverse events (time frame: 12 months): proportion of participants experiencing a treatment failure motivated by intolerable drug‐related adverse events during the follow‐up Quality of life in epilepsy scale 31 items (QOLIE‐31), Italian version (time frame: 12 months). QOLIE‐31 total score at baseline and last visit. Maximum total score is 100 (best quality of life possible) and the minimum is 0 (worst quality of life possible) Patient healthcare satisfaction scale, 18 items (PSQ‐18; time frame: 12 months). The score of the 7 PSQ‐18 subscales (general satisfaction, technical quality, interpersonal manner, communication, financial aspects, time spent with doctor, accessibility, and convenience) at baseline and last visit. Possible scores of each subscale range from 1 (worst satisfaction) to 5 (better satisfaction). There is no total score for this scale. Healthcare resources utilization (time frame: 12 months): mean daily cost of healthcare resources consumed for the management of epilepsy during the first 12 months of the study.
Starting date	September 2018
Contact information	Ettore Beghi, Istituto di Ricerche Farmacologiche Mario Negri IRCCS, ettore.beghi@marionegri.it [mailto:ettore.beghi%40marionegri.it?subject=NCT03689114, STANDLOW, Low vs. Standard Daily Doses of Antiepileptic Drugs in Newly Diagnosed, Previously Untreated Epilepsy(STANDLOW)]
Notes

Differences between protocol and review

We made some changes to the format and content of the Methods from the original protocol, in line with current MECIR standards (MECIR 2012) and the Cochrane Style Manual. We removed the inclusion criterion that specified that only trials describing adequate methods of allocation concealment should be included. As a result, three additional studies were included in the review.

We replaced the term 'partial' with 'focal', and 'refractory' with 'drug‐resistant' in accordance with the most recent classification of epilepsies of the International League Against Epilepsy (Scheffer 2017).

We replaced the term 'anti‐epileptic drugs (AEDs)' with 'antiseizure medications (ASMs)' in accordance with the new nomenclature of the International League Against Epilepsy (Perucca 2022).

Contributions of authors

MP was primarily responsible for the writing of this update and completed data extraction and risk of bias assessments. MP also selected studies and assessed the trials for eligibility, extracted data, and assessed risk of bias.
RB assessed studies for inclusion, extracted data, and assessed risk of bias. In addition, RB was responsible for the GRADE assessment and provided guidance and manuscript feedback during the update process.
AGM provided guidance and manuscript feedback during the update process.

Sources of support

Internal sources

• New Source of support, UK

  Cochrane Methodology Review Group

External sources

• National Institute for Health Research (NIHR), UK

  This review update was funded by the National Institute for Health Research (NIHR) [Clinically effective treatments for central nervous system disorders in the NHS, with a focus on Epilepsy and Movement Disorders (SRPG project 16/114/26)]. The views expressed are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care.

Declarations of interest

MP: none known.
RB: none known.
AGM: a consortium of pharmaceutical companies (GSK, EISAI, UCB Pharma) funded the National Audit of Seizure Management in Hospitals (NASH) through grants paid to the University of Liverpool. AGM is partly funded by the Applied Research Collaboration North West Coast (ARC NWC). AGM is also the Co‐ordinating Editor of the Cochrane Epilepsy Group; however, he was not involved in the editorial process for this review update.

New search for studies and content updated (no change to conclusions)