Monoterapia con clonazepam para el tratamiento de personas con epilepsia de diagnóstico reciente
Resumen
Antecedentes
La epilepsia es uno de los trastornos neurológicos más comunes en todo el mundo, con una prevalencia ajustada por edad de 4 a 8 por cada 1000 habitantes y una incidencia ajustada por edad de 44 por cada 100 000 años‐persona en los países desarrollados. La monoterapia representa la mejor opción terapéutica en personas con epilepsia de diagnóstico reciente.
Ésta es una versión actualizada de la revisión Cochrane original publicada en el número 11, 2019.
Objetivos
Evaluar la eficacia y la tolerabilidad del clonazepam oral utilizado como monoterapia para la epilepsia de diagnóstico reciente, en comparación con placebo o un fármaco anticonvulsivo diferente.
Métodos de búsqueda
Para la última actualización de esta revisión se hicieron búsquedas en las siguientes bases de datos el 14 de septiembre de 2021: el Registro Cochrane de estudios (CRS Web) y en MEDLINE (Ovid) (1946 al 13 de septiembre de 2021). La CRS Web incluye ensayos controlados aleatorizados (ECA) o cuasialeatorizados de PubMed, Embase, ClinicalTrials.gov, la Plataforma de registros internacionales de ensayos clínicos (ICTRP) de la Organización Mundial de la Salud, el Registro Cochrane central de ensayos controlados (Cochrane Central Register of Controlled Trials; CENTRAL) y los registros especializados de los Grupos de Revisión Cochrane, incluido el de epilepsia.
Criterios de selección
Se incluyeron ECA o ensayos cuasialeatorizados que compararon clonazepam oral utilizado como tratamiento de monoterapia versus placebo o un fármaco anticonvulsivo diferente (comparador activo) en personas de cualquier edad con epilepsia de diagnóstico reciente, de acuerdo con la definición clínica práctica propuesta por la International League Against Epilepsy (ILAE).
Obtención y análisis de los datos
Se consideraron los siguientes desenlaces: proporción de participantes sin convulsiones al mes, a los tres, seis, 12 y 24 meses después de la aleatorización; proporción de participantes que respondieron al tratamiento (los que presentaron una reducción de al menos un 50% en la frecuencia de las convulsiones desde el inicio hasta el final del tratamiento); proporción de participantes con eventos adversos emergentes del tratamiento (EAET) durante el período de tratamiento o que dieron lugar a su interrupción durante el período de tratamiento; proporción de abandonos/retiros debido a los efectos secundarios, falta de eficacia u otras razones; y mejoría en la calidad de vida, según lo evaluado con escalas de calificación validadas y fiables.
Dos autores de la revisión seleccionaron de forma independiente todos los títulos y resúmenes de las publicaciones identificadas mediante las búsquedas para evaluar su elegibilidad. De forma independiente se extrajeron los datos de los ensayos y se verificó su exactitud. Cualquier desacuerdo entre los dos autores sobre la extracción de los datos se resolvió mediante discusión y consenso. Se examinaron los ensayos y se evaluó la calidad metodológica de los estudios incluidos. Se utilizó el método GRADE para evaluar la certeza de la evidencia.
Resultados principales
En la versión anterior de la revisión ya se habían incluido dos ensayos controlados aleatorizados, con un total de 115 participantes. Un estudio comparó clonazepam con carbamazepina como monoterapia para los participantes con epilepsia psicomotora de diagnóstico reciente (una afección que corresponde a lo que ahora se denomina epilepsia del lóbulo temporal mesial). Un estudio (publicado como resumen) comparó clonazepam con etosuximida como monoterapia en niños con crisis de ausencia. Sobre la base de los datos disponibles y los detalles sobre la metodología proporcionada, se consideró que ambos estudios estaban en riesgo de sesgo incierto o alto en los dominios evaluados (a excepción del dominio de informe selectivo [sesgo de notificación]; se consideró que un estudio estuvo en riesgo bajo de sesgo y el otro en riesgo alto).
En el estudio que comparó clonazepam con carbamazepina no se encontraron diferencias entre los grupos con respecto a la proporción de participantes que no tuvieron convulsiones al mes de la asignación al azar (razón de riesgos [RR] 1,97; intervalo de confianza [IC] del 95%: 0,99 a 3,94; 30 participantes; evidencia de certeza muy baja), tres meses después de la aleatorización (RR 1,19; IC del 95%: 0,62 a 2,29; 26 participantes; evidencia de certeza muy baja) y seis meses después de la aleatorización (RR 0,50; IC del 95%: 0,09 a 2,73; nueve participantes; evidencia de certeza muy baja). No se encontraron diferencias estadísticas entre el clonazepam y la carbamazepina en cuanto a la proporción de participantes con EAET que dieron lugar a la interrupción (RR 2,61; IC del 95%: 0,80 a 8,52; 36 participantes; evidencia de certeza muy baja) y en cuanto a los abandonos/retiros debido a los efectos secundarios, la falta de eficacia u otras razones (RR 1,56; IC del 95%: 0,61 a 4,02; 36 participantes; evidencia de certeza muy baja). El estudio no proporcionó información sobre los otros desenlaces de interés predeterminados.
El estudio que comparó clonazepam con etosuximida no presentó datos sobre la eficacia. La proporción de abandonos/retiros fue mayor en el grupo que recibió clonazepam en comparación con el grupo que recibió etosuximida (RR 3,63; IC del 95%: 1,12 a 11,74; 79 participantes; evidencia de certeza muy baja). En este estudio no se proporcionó información sobre otros desenlaces de interés.
Conclusiones de los autores
No se encontraron estudios nuevos desde la última versión de esta revisión. Solo existe evidencia limitada y de certeza muy baja proveniente de ensayos controlados aleatorizados sobre la eficacia y la tolerabilidad del clonazepam utilizado como monoterapia para el tratamiento de la epilepsia. No se encontraron diferencias en la eficacia ni la tolerabilidad en un ensayo pequeño que comparó clonazepam con carbamazepina para el tratamiento de la epilepsia del lóbulo temporal mesial. El clonazepam presentó una tolerabilidad menor que la etosuximida en un ensayo en niños con crisis de ausencia; sin embargo, no se proporcionaron datos comparativos sobre la eficacia. En la actualidad no existe evidencia suficiente para apoyar el uso del clonazepam como monoterapia para la epilepsia.
PICO
Resumen en términos sencillos
Clonazepam como medicamento antiepiléptico único para el tratamiento de personas con epilepsia de diagnóstico reciente
Antecedentes
La monoterapia (tratamiento con un solo medicamento) es la mejor opción para tratar a las personas con epilepsia de diagnóstico reciente. El clonazepam es un tipo de medicamento que puede reducir el número de ataques epilépticos; pertenece a un grupo de medicamentos conocidos como benzodiazepinas. Se realizaron búsquedas en bases de datos electrónicas con el objetivo de combinar los resultados de todos los estudios bien realizados sobre el tema para determinar la efectividad del clonazepam en la reducción del número de convulsiones en personas con epilepsia de diagnóstico reciente.
Características de los estudios
Se identificaron solo dos ensayos pequeños que compararon clonazepam con un medicamento diferente en dos síndromes epilépticos diferentes, la epilepsia del lóbulo temporal mesial (la epilepsia focal más frecuente y mejor definida con crisis que se originan en la parte interna del lóbulo temporal del cerebro) y las crisis de ausencia (crisis generalizadas que causan lapsos de conciencia). En el estudio realizado en la epilepsia del lóbulo temporal mesial, el clonazepam se comparó con la carbamazepina (un medicamento antiepiléptico utilizado para tratar la epilepsia focal). En el estudio sobre las crisis de ausencia, el clonazepam se comparó con etosuximida (un medicamento utilizado para tratar las crisis de ausencia).
Resultados clave
Se consideró que ambos estudios fueron de calidad deficiente. Los estudios no dieron seguimiento a los participantes durante el tiempo suficiente y el número total de participantes fue demasiado escaso para establecer conclusiones definitivas sobre la función del clonazepam utilizado en la monoterapia. Los resultados sobre la tolerabilidad no se comunicaron de forma consistente en todos los estudios.
No se encontraron diferencias entre el clonazepam y la carbamazepina en la proporción de participantes sin crisis epilépticas; sin embargo, este hecho no significa que el clonazepam y la carbamazepina tengan el mismo efecto en el control de las crisis, ya que la falta de diferencia puede deberse al escaso número de personas incluidas.
El estudio que comparó clonazepam con etosuximida no proporcionó resultados sobre el efecto en el control de las crisis. No se encontraron diferencias entre los dos medicamentos en cuanto a la tolerabilidad. Sin embargo, la proporción de personas que abandonaron o se retiraron del estudio debido a los efectos secundarios, la falta de eficacia u otras razones fue mayor en el grupo de clonazepam en comparación con el grupo de etosuximida.
Certeza de la evidencia
Hasta ahora, la evidencia sobre el efecto en el control de las crisis epilépticas y la tolerabilidad del clonazepam utilizado como único antiepiléptico para el tratamiento de la epilepsia es escasa y de certeza muy baja. Por lo tanto, no hay información suficiente en la que basar las decisiones sobre el uso del clonazepam como monoterapia.
La evidencia está actualizada hasta septiembre de 2021.
Authors' conclusions
Summary of findings
| Clonazepam compared to carbamazepine for treating people with newly diagnosed epilepsy | ||||||
| Patient or population: people with newly diagnosed epilepsy | ||||||
| Outcomes | Relative effect | Anticipated absolute effects* (95% CI) | Certainty of the evidence | Comments | ||
|---|---|---|---|---|---|---|
| Without Clonazepam | With Clonazepam | Difference | ||||
| Proportion of participants seizure‐free at 12 months after randomization ‐ not reported | Outcome not reported | n/a | ||||
| Proportion of participants seizure‐free at 24 months after randomization ‐ not reported | Outcome not reported | n/a | ||||
| Proportion of responders (those with at least a 50% reduction in seizure frequency from baseline to end of treatment) | Outcome not reported | n/a | ||||
| Proportion of participants with treatment‐emergent adverse events (TEAEs) during the treatment period | Outcome not reported | n/a | In the only study included, all participants (except one receiving carbamazepine) had at least one side effect (no further details provided). The duration of side effects was generally short, without significant difference between clonazepam and carbamazepine for sedation, headache, dizziness, irritability and other complaints. | |||
| Proportion of participants with treatment‐emergent adverse events (TEAEs) leading to discontinuation during the treatment period (1 RCT; n = 36) | RR 2.61 | Study population | ⊕⊝⊝⊝ | |||
| 15.8% | 41.2% | 25.4% more | ||||
| *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). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 Study conducted in a very small population. No calculation of sample size was made. Very wide confidence intervals, indicative of imprecise results. 2 High risk of attrition bias and reporting bias; unclear risk of selection and performance bias. | ||||||
| Clonazepam compared to ethosuximide for treating people with newly diagnosed epilepsy | ||||||
| Patient or population: people with newly diagnosed epilepsy | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with ethosuximide | Risk with Clonazepam | |||||
| Proportion of participants seizure‐free at 12 months after randomization | Outcome not reported | n/a | ||||
| Proportion of participants seizure‐free at 24 months after randomization | Outcome not reported | n/a | ||||
| Proportion of responders (those with at least a 50% reduction in seizure frequency from baseline to end of treatment) | Outcome not reported | n/a | ||||
| Proportion of participants with treatment‐emergent adverse events (TEAEs) during the treatment period | Outcome not reported | n/a | In the only study included, drowsiness and ataxia were common side effects and seemed dose‐related (no further details were provided). | |||
| Proportion of participants with TEAEs leading to discontinuation during the treatment period | Outcome not reported | n/a | ||||
| *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). | ||||||
| GRADE Working Group grades of evidence | ||||||
Background
Description of the condition
Epilepsy is "a disorder of the brain characterized by an enduring predisposition to generate epileptic seizures" (Fisher 2005). According to the practical definition proposed by the International League Against Epilepsy (ILAE) in 2014, epilepsy can be diagnosed in "any of the following conditions: (1) at least two unprovoked (or reflex) seizures occurring > 24 hours apart; (2) one unprovoked (or reflex) seizure and a probability of further seizures similar to the general recurrence risk (at least 60%) after two unprovoked seizures, occurring over the next 10 years; (3) diagnosis of an epilepsy syndrome" (Fisher 2014).
Epilepsy is one of the most common neurological disorders worldwide, with an age‐adjusted prevalence in developed countries of 4 to 8 per 1000 population (Hauser 1991), and an age‐adjusted incidence of 44 per 100,000 person‐years (Hauser 1993). In low‐ and middle‐income countries its incidence is generally higher, ranging from 100 to 190 per 100,000/year (Sander 1996). It affects people of any age, and in adulthood its incidence and prevalence increase with age (Hauser 1993).
Monotherapy represents the best therapeutic option in people with newly diagnosed epilepsy, and the choice of the initial antiepileptic drug should take into account the person's seizure type and epilepsy syndrome, age, childbearing potential, comorbidities, medical history, tolerability and risk of drug interactions (Perucca 2011). In this review we investigated the efficacy and tolerability of clonazepam used as monotherapy in people with newly diagnosed epilepsy.
Description of the intervention
Clonazepam, or 5‐(2‐chlorophenyl)‐7‐nitro‐1,3‐dihydro‐1,4‐benzodiazepin‐2‐one, is a drug with anxiolytic and antiepileptic properties. It is rapidly absorbed from the gastrointestinal tract, with an oral bioavailability that is almost complete (more than 85%) and time to peak levels after a single oral dose of one to four hours (Trinka 2016). Clonazepam is extensively bound to plasma proteins (86%) and has no active metabolite. The drug is highly lipophilic, which enables a rapid cross of the blood‐brain barrier, resulting in a rapid onset of action. Clonazepam is eliminated by nitroreduction and has an elimination half‐life of 17 to 55 hours in healthy adults not taking enzyme‐inducing agents (Trinka 2016).
Clonazepam has a broad‐spectrum efficacy against many seizure types, both generalized (i.e. of generalized onset) and focal (i.e. of focal onset), including absences (i.e. generalized onset non‐motor seizures) (Sato 1977; Trinka 2016; Wheless 2005). It is particularly effective against myoclonic seizures occurring in different epilepsy types, including progressive myoclonic epilepsies (Browne 1976; Livanainen 1982; Obeso 1995‐1996). Clonazepam is mainly used as an add‐on treatment, but has been proven to be effective also when used in monotherapy. In particular, a paediatric study conducted in 60 children with epilepsy other than infantile spasms, showed that clonazepam monotherapy led to clinical seizure cessation in 71% of the participants with generalized seizures and 89% of focal seizures; the drug was well tolerated, with adverse effects (mainly drowsiness and ataxia) reported in 5% of children (Ishikawa 1985).
Common or clinically relevant adverse events related to clonazepam use include behavioural disorders (irritability, aggressive behaviour, hyperactivity), drowsiness, ataxia, sedation, cognitive dysfunction, and drooling (Browne 1978; Trinka 2016). Adverse effects occur in up to 50% of people treated with this drug (Browne 1978; Trinka 2016), and in some cases may improve after dose reduction and over time due to tolerance. Intolerable adverse effects have been reported in approximately 20% of people with epilepsy (Browne 1978; Keränen 1983).
How the intervention might work
Clonazepam belongs to the drug class of benzodiazepines. Benzodiazepines enhance the binding of the gamma‐aminobutyric acid (GABA) to the GABA‐A receptors, increasing channel opening frequency; this leads to increased chloride conductance and neuronal hyperpolarization, and eventually to enhanced inhibitory neurotransmission and antiepileptic action (Trinka 2016).
Why it is important to do this review
Due to its antiepileptic properties, clonazepam could show promise as monotherapy; monotherapy represents the gold standard of treatment in newly‐diagnosed epilepsy and is preferred over add‐on therapy, particularly in this group of people who are not refractory to other treatments.
We planned to comprehensively search for, and critically assess, the scientific literature on the clinical role of oral clonazepam used as monotherapy treatment for people with newly diagnosed epilepsy. In particular, we wanted to evaluate whether high‐quality randomized controlled trials assessing the efficacy and tolerability of this drug are available.
Objectives
To assess the efficacy and tolerability of oral clonazepam used as monotherapy for newly diagnosed epilepsy, when compared with placebo or a different anti‐seizure medication.
Methods
Criteria for considering studies for this review
Types of studies
We included randomized controlled trials (RCTs) or quasi‐RCTs comparing oral clonazepam used as monotherapy treatment (i.e. where participants were randomized to treatment with a single drug throughout the study period) versus placebo or a different anti‐seizure medication (active comparator) in people of any age with newly diagnosed epilepsy, defined according to the clinical practical definition proposed by the ILAE (Fisher 2014). We did not include RCTs assessing the role of clonazepam as treatment of status epilepticus.
We included only RCTs of parallel or cross‐over designs, excluding uncontrolled and non‐randomized trials. We did not exclude trials on the basis of dose, duration of treatment, or length of follow‐up.
Types of participants
We included people with newly diagnosed epilepsy of any type (Scheffer 2016) — defined as two or more unprovoked seizures or a single unprovoked seizure and high risk of seizure recurrence (Fisher 2014) ‐ regardless of sex, age, or ethnicity.
Types of interventions
We included clonazepam monotherapy at any dosage versus placebo or a different anti‐seizure drug (active comparator). We also included dose‐controlled studies (i.e. multiple‐arm trials assessing clonazepam at different dosages).
Types of outcome measures
Primary outcomes
-
Proportion of participants seizure‐free at one, three, six, 12, and 24 months after randomization
-
Proportion of responders (defined as those who experienced at least a 50% reduction in seizure frequency from baseline to end of treatment)
-
Proportion of participants with treatment‐emergent adverse events (TEAEs)
-
During the treatment period
-
Leading to discontinuation during the treatment period
-
We assessed seizure freedom and proportion of responders both for all seizure types and for specific seizure type (e.g. myoclonic, absence, tonic, atonic, tonic‐clonic, and focal seizures). For each outcome, we planned to perform an intention‐to‐treat (ITT) primary analysis to include all participants in the treatment group to which they were allocated, irrespective of treatment actually received. We planned to use ITT data in the analysis for all randomly assigned participants recorded during the entire treatment period, including both titration and evaluation phases.
Secondary outcomes
-
Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons (we used this as a measure of global effectiveness)
-
Improvement in quality of life, as assessed by validated and reliable rating scales
Search methods for identification of studies
Electronic searches
Searches were run for the original review on 24 July 2018. For the latest update we searched the following databases on 14 September 2021.
-
Cochrane Register of Studies (CRS Web), using the search strategy shown in Appendix 1.
-
MEDLINE (Ovid) (1946 to 13 September 2021), using the search strategy shown in Appendix 2.
CRS Web includes randomized or quasi‐randomized, controlled trials from PubMed, Embase, ClinicalTrials.gov, the World Health Organization International Clinical Trials Registry Platform (ICTRP), the Cochrane Central Register of Controlled Trials (CENTRAL), and the Specialized Registers of Cochrane Review Groups, including Epilepsy. In MEDLINE (Ovid) the coverage end date always lags a few days behind the search date.
Searching other resources
We also carried out the following.
-
Handsearched the references quoted in the identified trials.
-
Contacted pharmaceutical companies (Roche) to identify unpublished trials or data missing from articles.
-
Contacted authors and known experts to identify any additional data.
-
Handsearched relevant conference proceedings (American Academy of Neurology (AAN), ILAE).
We did not impose any language restrictions on our searches, and we attempted to obtain translations of retrieved articles where necessary.
Data collection and analysis
Although we planned to perform several data analyses (reported in detail in the review protocol), we did not use most of them due to the small number of studies and data.
Selection of studies
Two review authors (FB, SL) independently screened all titles and abstracts to assess the eligibility of publications identified by the searches. We excluded publications that did not meet the criteria at this stage. After screening, we assessed the full text of potentially eligible citations for inclusion. We listed studies that initially appeared to meet the inclusion criteria, but we later excluded, in the Characteristics of excluded studies table. We collated multiple reports of the same study so that each study, rather than each report, was the unit of interest in the review. We also provided any information we could obtain about ongoing studies. We recorded the selection process in sufficient detail to complete a PRISMA flow diagram (Liberati 2009). We resolved any disagreement through discussion.
Data extraction and management
Two review authors (FB and SL) independently extracted data from trial reports onto standardized forms, and cross‐checked them for accuracy. We resolved any disagreement regarding data extraction by discussion and consensus between the review authors. We extracted the following trial data.
-
Main study author and age of publication.
-
Total number and demographics of participants for each group (age, sex, weight, height, body mass index, number of seizures in the past three months, number of seizures in the past 12 months, epilepsy duration, type of seizures).
-
Intervention details (study design; inclusion and exclusion criteria; description of study phases with details on starting and target dose, titration, and length of each phase; primary and secondary end points).
-
Trial methods (method of generating random list; method of concealing randomization; blinding methods).
-
Definitions of ITT/full analysis, safety, and per‐protocol population adopted in each study.
-
Proportion of participants achieving seizure freedom at one, three, six, 12 and 18 months after randomization in each group.
-
Proportion of responders (those who experienced at least a 50% reduction in seizure frequency from baseline to end of treatment).
-
Proportion of participants with TEAEs during the treatment period in each group.
-
Proportion of participants with TEAEs leading to discontinuation during the treatment period in each group.
-
Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons.
-
Improvement in quality of life.
Assessment of risk of bias in included studies
We scrutinized trials and evaluated the methodological quality of all included studies. Two review authors (FB and SCI) assessed the risk of bias of each trial according to the approaches described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement regarding data extraction by discussion and consensus between the review authors.
For each study, we evaluated the following characteristics and assigned a judgement of low, high or unclear risk of bias.
-
Random sequence generation (selection bias).
-
Allocation concealment (selection bias).
-
Blinding of participants and personnel (performance bias).
-
Blinding of outcome assessment (detection bias).
-
Incomplete outcome data addressed.
-
Selective reporting (reporting bias).
-
Other bias, including outcome reporting bias.
Measures of treatment effect
We used statistical methods in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011), to measure treatment effect. We planned to use mean differences (MDs) with 95% confidence intervals (CIs) for continuous data (improvement in quality of life, as assessed by validated and reliable rating scales) where we found data provided as means and standard deviations (SDs). Alternatively, where possible, we planned to calculate these data by using conventional statistical formulae. In the event of different quality‐of‐life scales, we planned to combine the different scales and present results as standardized mean differences (SMDs). No continuous data were found in the present review, and hence the planned method was not used.
We analyzed dichotomous data by calculating risk ratios (RRs) for each trial with the uncertainty in each trial being expressed using 95% CIs. For individual adverse effects we used 99% CIs to allow for multiple testing. The outcomes that we analyzed as dichotomous data are: (1) proportion of participants seizure free at one, three, six, 12 and 18 months after randomization; (2) proportion of responders (those who experienced at least a 50% reduction in seizure frequency from baseline to end of treatment); (3) proportion of participants with TEAEs during the treatment period; (4) proportion of participants with TEAEs leading to discontinuation during the treatment period; and (5) proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons.
Both for dichotomous and continuous data, we planned to calculate a pooled treatment effect across trials; however, due to lack of information and small number of studies, we did not do this.
Unit of analysis issues
For any unit of analysis issues, we planned to deal with them using the guidance in the Cochrane Handbook (Higgins 2011). We planned to analyze multiple‐armed trials as follows: in the event of dose‐controlled studies with only one control group (i.e. two or more different doses of clonazepam versus control) we planned to "include each pair‐wise comparison separately, but with shared intervention groups divided out approximately evenly among the comparisons" (Higgins 2011), in order to overcome a unit of analysis error. We planned to analyze randomized cross‐over studies in meta‐analyses using the results from paired analyses (Elbourne 2002). However, due to lack of information and small number of studies, we did not use these planned methods.
Dealing with missing data
For individual missing data, such as information on dropouts or losses to follow‐up, we planned to carry out an ITT analysis, using as the denominator the total number of people who underwent randomization. However, due to lack of information and the small number of studies we did not use this planned method.
Assessment of heterogeneity
We did not implement our planned methods to assess heterogeneity, due to the small number of studies included. Details on assessment of heterogeneity are provided below.
We planned to visually inspect the forest plots to investigate the possibility of statistical heterogeneity. We planned to evaluate statistical heterogeneity using the I2 statistic, which provides an estimate of the percentage of variability due to heterogeneity rather than a sampling error (Higgins 2003). We planned to interpret the level of heterogeneity using I2 according to the Cochrane Handbook, as follows: 0% to 40%: might not be important; 30% to 60%: moderate heterogeneity; 50% to 90%: substantial heterogeneity; 75% to 100%: considerable heterogeneity (Higgins 2011).
We planned to combine trial outcomes to obtain a summary estimate of effect (and the corresponding CIs) using a fixed‐effect model, unless there was considerable heterogeneity (that is, an I2 value greater than 75%). If there was substantial heterogeneity we planned to explore the factors contributing to heterogeneity. If there was substantial heterogeneity that could not readily be explained, we planned to use a random‐effects model. We planned to supplement homogeneity among trial results using a standard Chi2 test, and we rejected the hypothesis of homogeneity if the P value was less than 0.10. We planned to assess possible sources of heterogeneity (for example, clinical, methodological or statistical heterogeneity) by using sensitivity analysis as described below.
Assessment of reporting biases
We planned to use a funnel plot to detect reporting biases if sufficient studies (more than 10) were available. We planned to analyze possible sources of funnel plot asymmetry (e.g. publication bias, language bias, citation bias, poor methodological quality, true heterogeneity, etc.) according to the trials. However, due to lack of information and the small number of studies we did not use this planned method.
Data synthesis
Provided we thought it clinically appropriate, and no considerable clinical and methodological heterogeneity was found, we planned to synthesize the trial results in a meta‐analysis. We planned to use the Mantel‐Haenszel method for analyzing dichotomous data, and inverse variance for continuous data. However, due to lack of information and the small number of studies we did not synthesize the study results in a meta‐analysis.
Subgroup analysis and investigation of heterogeneity
We planned to perform subgroup analysis to separately assess RCTs conducted in paediatric and adult populations, and to assess different doses and lengths of treatment. However, due to lack of information and the small number of studies we did not use this planned method.
Sensitivity analysis
In the case of residual unexplained heterogeneity, we planned to evaluate the robustness of the results of the meta‐analysis by comparing fixed‐effect and random‐effects model estimates, removing trials with low methodological quality (i.e. studies with inadequate allocation concealment or lack of blinded outcome assessor). If the conclusions we observed were unchanged, then we would have considered the evidence to be robust. We did not use this method, due to the small number of studies included. We also planned to use the worst‐case and best‐case scenarios for taking into account missing data, however again we were unable to implement these methods.
Summary of findings and assessment of the certainty of the evidence
The summary of findings table for each comparison includes information on the overall certainty of the evidence from the trials and information of importance for healthcare decision‐making. We used the GRADE approach to determine the certainty of evidence for each outcome, on the basis of an evaluation of five criteria for RCTs (risk of bias, inconsistency, indirectness, imprecision, publication bias) (Schunemann 2011). If we had serious concerns regarding one of the five criteria, we downgraded the evidence from 'high quality' by one level; if we had very serious concerns we downgraded the evidence by two levels. We used these assessments to guide our conclusions and recommendations.
We used GRADEpro GDT software (GRADEpro GDT 2015), and imported data from Review Manager 5 to create summary of findings tables for each comparison included in the review for the following primary outcomes (Review Manager 2020).
-
Proportion of participants seizure‐free at 12 and 24 months after randomization.
-
Proportion of responders (those who experienced at least a 50% reduction in seizure frequency from baseline to end of treatment).
-
Proportion of participants with treatment‐emergent adverse events (TEAEs) during the treatment period.
-
Proportion of participants with TEAEs leading to discontinuation during the treatment period.
Results
Description of studies
We included two randomized controlled trials in the review. One study (Mikkelsen 1981) compared clonazepam to carbamazepine as monotherapy for participants with newly diagnosed psychomotor epilepsy (a condition corresponding to what is now termed mesial temporal lobe epilepsy). One study (Sato 1977) compared clonazepam to ethosuximide as monotherapy for children with absence seizures.
The results of Mikkelsen 1981 were published as a full‐length article, whereas those of Sato 1977 were provided as abstract.
Results of the search
The search strategy mentioned above returned 270 records (183 through CRS Web, 87 through MEDLINE), resulting in 6 articles, the full‐text of which were assessed for eligibility, after 47 duplicates were removed and 216 items were excluded after title and abstract screening. Of these, we further excluded three records (see Excluded studies). Despite our efforts, we were unable to retrieve the abstract and full‐length report of one study (Miyasaka 1977), which was hence listed among studies awaiting classification (Studies awaiting classification). See Figure 1.
Study flow diagram
Included studies
The first study was an RCT conducted in 36 previously untreated patients with newly diagnosed psychomotor epilepsy (a condition corresponding to what is now termed mesial temporal lobe epilepsy); it compared clonazepam to carbamazepine (Mikkelsen 1981). Patients with progressive brain diseases, presenile dementia, liver and kidney disease and pregnant women were excluded. The study randomly allocated 17 participants to clonazepam and 19 to carbamazepine. After two weeks of titration (further details were not provided), participants received either 6 mg clonazepam or 900 mg carbamazepine, each divided into three daily doses (maintenance dose). In participants younger than 18 years and with a body weight lower than 60 kg, carbamazepine was given at the dose of 15 mg/kg. Dose adjustments were possible “only in agreement with the department”. Occurrence of psychomotor seizures (corresponding to focal seizures with impaired awareness) or grand mal seizures (corresponding to generalized tonic‐clonic seizures) was recorded daily by the participants. The occurrence of headache, dizziness, impaired memory, and irritability was assessed after two weeks, and at one, two, four and six months. The mean age of the 36 included participants (14 males, 22 females) was 22.3 years (range: 6 to 72 years), and the median body weight was 60.5 kg (range: 23.3 kg to 97.5 kg). The median duration of epilepsy was 1.7 years (range: 0 to 10 years), and the median number of seizures per month at baseline was 3.5 (range: 0.5 to 20) in the carbamazepine group and 3.8 (range: 0.4 to 30) in the clonazepam group. At baseline, 7 out of 17 participants in the clonazepam group and 2 out of 19 in the carbamazepine group had experienced grand mal seizures.
The second included study was an RCT which compared clonazepam to ethosuximide for the treatment of children with absence seizures (Sato 1977). The study included 79 children (25 males, 54 females) with absence seizures, aged 5 to 16 years. The study was described as “double‐blind” (no further details were provided), and it lasted 17 weeks. In the study, 43 participants were assigned to clonazepam and 36 to ethosuximide; 30 of the participants in the clonazepam group, and 33 of those in the ethosuximide group, completed the study.
Excluded studies
In total we excluded three studies. One article was excluded as it assessed clonazepam as an add‐on treatment (Dahlin 2000); one was excluded because it did not assess the clinical efficacy of clonazepam, but only its effect on epileptiform discharges (Mitsudome 1997); and the last study was excluded because it was not an RCT (Yamatogi 1997). Two expert neuropaediatrics and one expert adult neurologist (Acknowledgements) were contacted by mail (on 26 September 2018 and 8 December 2018) for information on ongoing trials; they kindly replied that they were not aware of any ongoing trials.
Studies awaiting classification
Despite our efforts, we were unable to retrieve the abstract and full‐length report of one study published in 1977 (Miyasaka 1977). According to the title, this study, which is currently listed in Studies awaiting classification, assessed the efficacy of clonazepam compared with nitrazepam in a double‐blind cross‐over trial conducted in patients with psychomotor epilepsy (a condition corresponding to what is now termed mesial temporal lobe epilepsy). Should we continue to be unable to contact the study authors in future updates of this review, we will add this as an excluded study.
Risk of bias in included studies
See: Figure 2.
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Neither of the included studies described the process of random sequence generation or allocation concealment, and so we rated both as being at unclear risk of bias. The studies did not provide enough information to judge whether the baseline characteristics were balanced between treatment groups.
Blinding
Although Mikkelsen 1981 was described as “double‐blind”, it is not clear whether and how this was achieved. The study authors report only that “administration of drugs and clinical examination were performed by two persons to ensure double‐blind conditions” (Mikkelsen 1981). It is unclear whether the two interventions (carbamazepine and clonazepam) were identical. Sato 1977 did not specify blinding.
Incomplete outcome data
Both studies reported the number but not the reasons for participant exclusions or missing data in both intervention arms. We therefore rated them as having a high risk of bias for this domain.
Selective reporting
No protocol was available for either study. We considered that the more clinically relevant outcomes in Mikkelsen 1981 were reported (low risk of reporting bias). No data on efficacy were provided in Sato 1977, so we decided there is a high risk of reporting bias.
Other potential sources of bias
No other bias was detected.
Effects of interventions
See: Summary of findings 1 Clonazepam compared to carbamazepine for treating people with newly diagnosed epilepsy; Summary of findings 2 Clonazepam compared to ethosuximide for treating people with newly diagnosed epilepsy
Two comparisons were available: 1. clonazepam versus carbamazepine, and 2. clonazepam versus ethosuximide.
1. Clonazepam versus carbamazepine
Data on this comparison were provided in Mikkelsen 1981.
Primary outcomes
1.1. Proportion of participants seizure‐free at one month after randomization
The duration of treatment differed considerably among study participants. Furthermore, only 17 out of 19 participants in the carbamazepine group and 11 out of 17 in the clonazepam group were treated for longer than 30 days (one month). Among these, eight in the carbamazepine group and seven in the clonazepam group were seizure‐free. No statistical difference was found between the groups (RR 1.97, 95% CI 0.99 to 3.94; 1 study, 30 participants); Analysis 1.1. We downgraded the certainty of the evidence from 'high' to 'very low' for serious risk of attrition bias and imprecision of effect estimates.
1.2. Proportion of participants seizure‐free at three months after randomization
The duration of treatment differed considerably among study participants. Furthermore, only 15 out of 19 participants in the carbamazepine group and 11 out of 17 in the clonazepam group were treated for longer than 90 days (three months). Among these, eight in the carbamazepine group and seven in the clonazepam group were seizure‐free. No statistical difference was found between the groups (RR 1.19, 95% CI 0.62 to 2.29; 1 study; 26 participants); Analysis 1.1. We downgraded the certainty of the evidence from 'high' to 'very low' for serious risk of attrition bias and for imprecision of effect estimates.
1.3. Proportion of participants seizure‐free at six months after randomization
The duration of treatment differed considerably among study participants. Furthermore, only 6 out of 19 participants in the carbamazepine group and 3 out of 17 in the clonazepam group were treated for longer than 180 days (six months). Among these, four in the carbamazepine group and one in the clonazepam group were seizure‐free. No statistical difference was found between the groups (RR 0.50, 95% CI 0.09 to 2.73; 1 study; 9 participants); Analysis 1.1. We downgraded the certainty of the evidence from 'high' to 'very low' for serious risk of attrition bias and for imprecision of effect estimates.
1.4. Proportion of participants seizure‐free at 12 months after randomization
Data on this outcome were not provided.
1.5. Proportion of participants seizure‐free at 24 months after randomization
Data on this outcome were not provided.
2. Proportion of responders (those with at least a 50% reduction in seizure frequency from baseline to end of treatment)
Data on this outcome were not provided.
3. Proportion of participants with treatment‐emergent adverse events (TEAEs) during the treatment period or leading to discontinuation during the treatment period
The study reported that except for one participant receiving carbamazepine, all participants had at least one side effect during the study period. The duration of side effects was generally short, and no significant difference was found between clonazepam and carbamazepine in terms of sedation, headache, dizziness, irritability and other complaints. Three out of 19 participants in the carbamazepine group discontinued the treatment because of mental disturbances (one participant) or exanthema (two participants). In the clonazepam group, 7 out of 17 participants discontinued the drug due to mental disturbances (2), ataxia and mental disturbances (2), dizziness (1) or impotence and sedation (1). No statistical difference was found between the groups (RR 2.61, 95% CI 0.80 to 8.52; 1 study; 36 participants); Analysis 1.2. We downgraded the certainty of the evidence from 'high' to 'very low' for serious risk of attrition bias and for imprecision of effect estimates.
Secondary outcomes
1. Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons
Seven participants allocated to clonazepam were withdrawn during the study, all of them for adverse effects. In the carbamazepine group, five participants were withdrawn (three for adverse effects, one for seizure worsening, and one because the treatment was found to be "superfluous"). No statistical difference was found between the groups (RR 1.56, 95% CI 0.61 to 4.02; 1 study; 36 participants); Analysis 1.3.
2. Improvement in quality of life
Data on this outcome were not provided.
2. Clonazepam versus ethosuximide
Data on this comparison were provided in Sato 1977.
Primary outcomes
1.1. Proportion of participants seizure‐free at one month after randomization
Data on this outcome were not provided.
1.2. Proportion of participants seizure‐free at three months after randomization
Data on this outcome were not provided.
1.3. Proportion of participants seizure‐free at six months after randomization
Data on this outcome were not provided.
1.4. Proportion of participants seizure‐free at 12 months after randomization
Data on this outcome were not provided.
1.5. Proportion of participants seizure‐free at 24 months after randomization
Data on this outcome were not provided.
2. Proportion of responders (those with at least a 50% reduction in seizure frequency from baseline to end of treatment)
Data on this outcome were not provided.
3. Proportion of participants with treatment‐emergent adverse events (TEAEs) during the treatment period or leading to discontinuation during the treatment period
Data on this outcome were not provided. The study authors reported only that drowsiness and ataxia were common side effects and seemed dose‐related. Withdrawal from clonazepam therapy was most commonly related to behavioural changes (i.e. hyperactivity, short attention span, and uninhibited and aggressive behaviour) and occasionally to increased seizure frequency.
Secondary outcomes
1. Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons
Thirteen participants out of 43 in the clonazepam group did not complete the study (reasons were not provided); in the ethosuximide group, three out of 36 participants did not complete the study (reasons were not provided). The proportion of dropouts/withdrawal was higher in participants receiving clonazepam compared to participants receiving ethosuximide (RR 3.63, 95% CI 1.12 to 11.74; 1 study; 79 participants); Analysis 2.1. We downgraded the certainty of the evidence from 'high' to 'very low' for serious risk of attrition and reporting bias, and for imprecision of effect estimates.
2. Improvement in quality of life
Data on this outcome were not provided.
Discussion
Since the last version of this review we did not find any new studies.
Summary of main results
Data from RCTs available in the scientific literature on clonazepam monotherapy for epilepsy are scarce and of low quality. In this systematic review of the literature we did not find significant differences for most outcomes; furthermore, there were too little data to draw any conclusions on the role of clonazepam monotherapy for epilepsy. Clonazepam is usually considered as a therapeutic option to be used as adjunctive treatment in epilepsy, particularly for treating myoclonic seizures (Trinka 2016). Furthermore, as for any other benzodiazepine, the chronic use of clonazepam can be associated with tolerance, leading to reduced efficacy over time (Trinka 2016). This may partially explain the paucity of RCTs available in the literature so far.
Overall completeness and applicability of evidence
We identified only two RCTs on clonazepam used as monotherapy for treating epilepsy (Mikkelsen 1981; Sato 1977). Both studies were conducted several years ago. The studies assessed the efficacy and tolerability of this drug in two different epileptic syndromes, mesial temporal lobe epilepsy and absence seizures, and used different comparators (carbamazepine and ethosuximide, respectively). The results of one study were published as a full‐length article (Mikkelsen 1981), whereas those of the other were provided as abstract (Sato 1977).
Quality of the evidence
Based on the available data and the details on methodology provided, we judged both studies as having unclear or high risk of bias for most domains. The length of follow‐up and the total number of participants were insufficient to draw definite conclusions on the efficacy of this drug used as monotherapy.
No statistical differences were found between clonazepam and carbamazepine in terms of the proportion of seizure‐free participants (Mikkelsen 1981). However, these results could be due to there being inadequate statistical power to detect a statistically significant and clinically relevant difference between the tested drugs, considering the low number of participants included and the wide confidence intervals which are indicative of imprecision of the results and limited data available.
In the GRADE assessment, the certainty of the evidence for the individual outcomes on seizure‐freedom (at one, three, and six month(s) after randomization) was downgraded by one level to reflect the fact that this study was subject to a high risk of attrition bias; the evidence was further downgraded for imprecision of the results. Overall, we judged the certainty of the evidence to be very low, which means that we have very little confidence in the effect estimate, and the true effect is likely to be substantially different from the estimate of effect.
Similarly, when assessing the certainty of the evidence for TEAEs leading to discontinuation, we downgraded the GRADE assessment by two levels to reflect the high risk of attrition bias and imprecision of the results. Therefore, we judged the certainty evidence for this outcome to be very low (i.e. the data do not provide a reliable indication of the likely effect; and the true effect is likely to be substantially different from the estimate of effect).
No data on efficacy in the comparison between clonazepam and ethosuximide were provided (Sato 1977). No significant differences in tolerability were found between clonazepam and carbamazepine or ethosuximide, although again the negative results could be due to low statistical power, linked to imprecision of the results and limited data available. However, the proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons was higher in the clonazepam group compared to the ethosuximide group (Sato 1977). It should be noted, however, that in this study data on tolerability were not reported systematically. Accordingly, we downgraded our GRADE assessment of the certainty of the evidence for this outcome by two levels to reflect the high risk of attrition and reporting bias, and the imprecision of the results. Hence, we judged the certainty of evidence for this outcome to be very low (i.e. the data do not provide a reliable indication of the likely effect; and the true effect is likely to be substantially different from the estimate of effect).
Potential biases in the review process
There were no potential biases during the review process.
Agreements and disagreements with other studies or reviews
In the literature there is currently no other systematic review assessing the efficacy and tolerability of clonazepam used as monotherapy for epilepsy.
Study flow diagram
Risk of bias summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1: Clonazepam versus carbamazepine, Outcome 1: Proportion of participants seizure‐free at 1, 3 and 6 months after randomization
Comparison 1: Clonazepam versus carbamazepine, Outcome 2: Proportion of participants with treatment‐emergent adverse events (TEAEs)
Comparison 1: Clonazepam versus carbamazepine, Outcome 3: Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons
Comparison 2: Clonazepam versus ethosuximide, Outcome 1: Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons
| Clonazepam compared to carbamazepine for treating people with newly diagnosed epilepsy | ||||||
| Patient or population: people with newly diagnosed epilepsy | ||||||
| Outcomes | Relative effect | Anticipated absolute effects* (95% CI) | Certainty of the evidence | Comments | ||
|---|---|---|---|---|---|---|
| Without Clonazepam | With Clonazepam | Difference | ||||
| Proportion of participants seizure‐free at 12 months after randomization ‐ not reported | Outcome not reported | n/a | ||||
| Proportion of participants seizure‐free at 24 months after randomization ‐ not reported | Outcome not reported | n/a | ||||
| Proportion of responders (those with at least a 50% reduction in seizure frequency from baseline to end of treatment) | Outcome not reported | n/a | ||||
| Proportion of participants with treatment‐emergent adverse events (TEAEs) during the treatment period | Outcome not reported | n/a | In the only study included, all participants (except one receiving carbamazepine) had at least one side effect (no further details provided). The duration of side effects was generally short, without significant difference between clonazepam and carbamazepine for sedation, headache, dizziness, irritability and other complaints. | |||
| Proportion of participants with treatment‐emergent adverse events (TEAEs) leading to discontinuation during the treatment period (1 RCT; n = 36) | RR 2.61 | Study population | ⊕⊝⊝⊝ | |||
| 15.8% | 41.2% | 25.4% more | ||||
| *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). | ||||||
| GRADE Working Group grades of evidence | ||||||
| 1 Study conducted in a very small population. No calculation of sample size was made. Very wide confidence intervals, indicative of imprecise results. 2 High risk of attrition bias and reporting bias; unclear risk of selection and performance bias. | ||||||
| Clonazepam compared to ethosuximide for treating people with newly diagnosed epilepsy | ||||||
| Patient or population: people with newly diagnosed epilepsy | ||||||
| Outcomes | Anticipated absolute effects* (95% CI) | Relative effect | № of participants | Certainty of the evidence | Comments | |
|---|---|---|---|---|---|---|
| Risk with ethosuximide | Risk with Clonazepam | |||||
| Proportion of participants seizure‐free at 12 months after randomization | Outcome not reported | n/a | ||||
| Proportion of participants seizure‐free at 24 months after randomization | Outcome not reported | n/a | ||||
| Proportion of responders (those with at least a 50% reduction in seizure frequency from baseline to end of treatment) | Outcome not reported | n/a | ||||
| Proportion of participants with treatment‐emergent adverse events (TEAEs) during the treatment period | Outcome not reported | n/a | In the only study included, drowsiness and ataxia were common side effects and seemed dose‐related (no further details were provided). | |||
| Proportion of participants with TEAEs leading to discontinuation during the treatment period | Outcome not reported | n/a | ||||
| *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). | ||||||
| GRADE Working Group grades of evidence | ||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1.1 Proportion of participants seizure‐free at 1, 3 and 6 months after randomization Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1.1 Seizure freedom at 1 month after randomization | 1 | 30 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.97 [0.99, 3.94] |
| 1.1.2 Seizure freedom at 3 months after randomization | 1 | 26 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.19 [0.62, 2.29] |
| 1.1.3 Seizure freedom at 6 months after randomization | 1 | 9 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.50 [0.09, 2.73] |
| 1.2 Proportion of participants with treatment‐emergent adverse events (TEAEs) Show forest plot | 1 | Risk Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.2.1 Leading to discontinuation during the treatment period | 1 | 36 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.61 [0.80, 8.52] |
| 1.3 Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons Show forest plot | 1 | 36 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.56 [0.61, 4.02] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 2.1 Proportion of dropouts/withdrawals due to side effects, lack of efficacy or other reasons Show forest plot | 1 | 79 | Risk Ratio (M‐H, Fixed, 95% CI) | 3.63 [1.12, 11.74] |
