Epilepsy & Behavior 64 (2016) 110–115

Contents lists available at ScienceDirect

Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh

Review

Direct and indirect comparison meta-analysis of levetiracetam versus phenytoin or valproate for convulsive status epilepticus Francesco Brigo a,b,⁎, Nicola Bragazzi c,d, Raffaele Nardone b,e, Eugen Trinka e,f,g a

Department of Neuroscience, Biomedicine and Movement, University of Verona, Italy Department of Neurology, Franz Tappeiner Hospital, Merano, Italy School of Public Health, Department of Health Sciences (DISSAL), University of Genoa, Genoa, Italy d Department of Neuroscience, Rehabilitation, Ophthalmology, Genetics, Maternal and Child Health (DINOGMI), Section of Psychiatry, University of Genoa, Genoa, Italy e Department of Neurology, Christian Doppler Klinik, Paracelsus Medical University, Salzburg, Austria f Centre for Cognitive Neuroscience Salzburg, Austria g Department of Public Health Technology Assessment, UMIT — University for Health Sciences, Medical Informatics and Technology, Hall.i.T., Austria b c

a r t i c l e

i n f o

Article history: Received 6 July 2016 Accepted 18 September 2016 Available online xxxx Keywords: Clinical trials randomized controlled Levetiracetam Meta-analysis Phenytoin Status epilepticus Valproate

a b s t r a c t Aim: The aim of this study was to conduct a meta-analysis of published studies to directly compare intravenous (IV) levetiracetam (LEV) with IV phenytoin (PHT) or IV valproate (VPA) as second-line treatment of status epilepticus (SE), to indirectly compare intravenous IV LEV with IV VPA using common reference-based indirect comparison meta-analysis, and to verify whether results of indirect comparisons are consistent with results of headto-head randomized controlled trials (RCTs) directly comparing IV LEV with IV VPA. Methods: Random-effects Mantel–Haenszel meta-analyses to obtain odds ratios (ORs) for efficacy and safety of LEV versus VPA and LEV or VPA versus PHT were used. Adjusted indirect comparisons between LEV and VPA were used. Results: Two RCTs comparing LEV with PHT (144 episodes of SE) and 3 RCTs comparing VPA with PHT (227 episodes of SE) were included. Direct comparisons showed no difference in clinical seizure cessation, neither between VPA and PHT (OR: 1.07; 95% CI: 0.57 to 2.03) nor between LEV and PHT (OR: 1.18; 95% CI: 0.50 to 2.79). Indirect comparisons showed no difference between LEV and VPA for clinical seizure cessation (OR: 1.16; 95% CI: 0.45 to 2.97). Results of indirect comparisons are consistent with results of a recent RCT directly comparing LEV with VPA. Conclusion: The absence of a statistically significant difference in direct and indirect comparisons is due to the lack of sufficient statistical power to detect a difference. Conducting a RCT that has not enough people to detect a clinically important difference or to estimate an effect with sufficient precision can be regarded a waste of time and resources and may raise several ethical concerns, especially in RCT on SE. © 2016 Elsevier Inc. All rights reserved.

1. Introduction Status epilepticus (SE) is defined as a “condition resulting either from the failure of the mechanisms responsible for seizure termination or from the initiation of mechanisms, which lead to abnormally, prolonged seizures” [1]. If seizure activity persists over time, longterm consequences, including neuronal death, neuronal injury, and alteration of neuronal networks, depending on the type and duration of seizures, may occur [1]. It is therefore crucial to promptly recognize and adequately treat this condition, as SE is associated with high risk of morbidity and mortality [2]. Benzodiazepines represent the first-line treatments for SE. However, in approximately 30–40% of cases, SE in patients fails to respond ⁎ Corresponding author at: Department of Neuroscience, Biomedicine and Movement, University of Verona, Piazzale L.A. Scuro, 10-37134 Verona, Italy. E-mail address: [email protected] (F. Brigo).

http://dx.doi.org/10.1016/j.yebeh.2016.09.030 1525-5050/© 2016 Elsevier Inc. All rights reserved.

to benzodiazepines, requiring an intravenous (IV) administration of antiepileptic drugs (AEDs). The most common second-line treatments for SE include phenytoin (PHT), phenobarbital, valproate (VPA), levetiracetam (LEV), and lacosamide [3]. To date, the evidence supporting the use of LEV in SE is mostly limited to retrospective case series [4,5]. Only two recent randomized controlled trials (RCTs) have directly compared IV LEV with either IV VPA [6] or IV PHT [6,7] showing no differences with the comparator(s) in terms of clinical seizure cessation. The data supporting the use of VPA as second-line treatment for SE are more robust, as this drug has been studied in six randomized controlled trials [8–13] showing good efficacy and tolerability [14,15]. The relative efficacy of VPA, LEV, and the other second-line treatments for SE (lacosamide, phenytoin, and phenobarbital) has been assessed in a systematic review with meta-analysis [5]. Efficacy of LEV (68.5%; 95% CI: 56.2–78.7%) and VPA (75.7%; 95% CI: 63.7–84.8%) were found to be comparable with that of phenobarbital (73.6%; 95%

F. Brigo et al. / Epilepsy & Behavior 64 (2016) 110–115

CI: 58.3–84.8%) and somewhat higher than that of PHT (50.2%; 95% CI: 34.2–66.1%), suggesting that LEV and VPA may represent valid alternatives to phenobarbital and PHT as second-line treatments of SE. In this study, we aimed 1. to perform a meta-analysis of IV LEV compared with IV VPA or IV PHT as second-line treatment of convulsive SE (generalized or focal) in patients of any age, 2. to indirectly estimate the efficacy of IV LEV and IV VPA through indirect comparison metaanalyses using IV PHT as common comparator, and 3. to assess whether results of indirect comparisons are consistent with results of a recent head-to-head RCT directly comparing IV LEV with IV VPA [6]. 2. Methods 2.1. Criteria for considering studies for this review Randomized controlled trials comparing IV LEV or IV VPA against IV PHT and RCTs comparing IV LEV versus IV VPA used as second-line treatment for convulsive SE (generalized or focal) were included in the meta-analysis. Patients from any age group diagnosed with convulsive SE persisting despite first-line AEDs (benzodiazepines) were included. Status epilepticus was defined as convulsive seizures lasting N 5 min [1,16]. We included all RCTs, blinded or not blinded, and excluded uncontrolled and nonrandomized trials. Trials were not excluded on the basis of dose, duration of treatment, or length of follow-up. The following electronic databases and data sources were systematically searched: 1. MEDLINE (January 1966–27th of October 2015), accessed through PubMed; 2. Cochrane Central Register of Controlled Trials (CENTRAL) (Issue 12, The Cochrane Library, December 2014) (accessed 27th of October 2015); 3. EMBASE (accessed 27th October 2015); 4. ClinicalTrials.gov (available at: https://clinicaltrials.gov/; accessed 27th of October 2015); and 5. handsearching of the references quoted in the identified trials. Search strategy adopted for all databases mentioned above is reported in Appendix.

111

All resulting titles and abstracts were evaluated, and any relevant article was considered. There were no language restrictions. 2.2. Study selection Retrieved articles were independently assessed for inclusion by two review authors (FB, RN); any disagreement was resolved through discussion. 2.3. Methodological quality assessment Trials were scrutinized, and the methodological quality of all included studies was evaluated. The randomized trials were judged on the risk of bias as outlined in the Cochrane Handbook for Systematic Reviews of Interventions version 5.1.0 [updated March 2011] [17]. 2.4. Data extraction The following trial data were extracted: main study author and age of publication; definition of SE applied in the study; type of participants (children, adults, or both); total number, age, and sex of participants for each treatment group; SE type; intervention details (dose, route of administration); definition of successful treatment adopted in each trial; and proportion of seizures controlled after drug administration in each treatment group. 2.5. Types of outcome measures We chose dichotomous primary outcomes to have hard outcome measures of treatment efficacy. We evaluated efficacy as the number of patients with clinical seizure cessation within 15 min after the start of drug administration. 3. Statistical analysis For each outcome, an intention-to-treat primary analysis was made to include all patients in the treatment group to which they were allocated, irrespective of the treatment they actually received.

Fig. 1. Direct and indirect comparisons between randomized controlled trials. Conventional meta-analyses of RCTs focus on direct, pair-wise comparisons between two treatments (treatment A versus treatment B). If direct head-to-head comparisons between two treatments are not available, definite data on treatment effect cannot be estimated. In this situation, it is possible to estimate the indirect effect of treatment A versus treatment B using evidence from trials comparing treatment A with treatment C and trials comparing treatment B with treatment C. The key assumption for this indirect comparison based on a common comparator (treatment C) is that of exchangeability of the treatment effect across all included trials.

112

F. Brigo et al. / Epilepsy & Behavior 64 (2016) 110–115

Analyses were conducted using RevMan 5 (conventional metaanalysis for each AED), Excel and R 2.15.1 (common reference-based indirect comparison meta-analysis).

For the indirect comparisons, an OR N 1 indicates that the outcome is more likely after IV LEV than after IV VPA administration. A P value of 0.05 was considered to be statistically significant. Heterogeneity was not assessed for indirect comparisons.

3.1. Conventional meta-analysis per antiepileptic drug 4. Results Conventional meta-analyses of comparisons between IV VPA or IV LEV and IV PHT and between IV LEV and IV VPA were performed using random effects, inverse variance, and weighted meta-analysis [18]. The outcome of interest (clinical seizure cessation after AED administration) was analyzed by calculating odds ratios (ORs) with 95% confidence intervals (CIs) and weighted treatment effect across trials. Mantel–Haenszel method was used to estimate the OR statistic and to combine ORs [17]. Results from separate studies were pooled together using the random effects model for the quantitative pooling [18], as adjusted indirect comparison using the fixed effect model tends to underestimate standard errors of pooled estimates [19,20].

The search strategy described above yielded 2671 results (1958 MEDLINE, 201 CENTRAL, 500 EMBASE, and 12 ClinicalTrials.gov) (Fig. 2). One study initially included was eventually excluded, as it assessed the efficacy of IV VPA used as first-line (and not as second-line) treatment for SE [8]. Hence, four studies were included (with a total of 321 episodes of SE), three comparing IV VPA with IV PHT [6,9,11] and two comparing IV LEV with IV PHT [7] or with IV

3.1.1. Assessment of heterogeneity Visual inspection of the forest plots was used to investigate the possibility of statistical heterogeneity. Homogeneity among trial results was evaluated using a standard Chi-squared test rejecting the hypothesis of homogeneity with P value lower than 0.10. Statistical heterogeneity was assessed by means of the I-squared (I2) statistic, which provides an estimate of the percentage of variability due to heterogeneity rather than a sampling error [21]. The I2 statistic for heterogeneity was interpreted according to Higgins and Green [17]. 3.2. Common reference-based indirect comparisons by combining meta-analyses of AEDs Common reference-based indirect comparisons were performed using the method suggested by Bucher [22] and adopted in previous reviews [23–26]: the indirect comparison of IV LEV and IV VPA was adjusted by the results of their direct comparisons with IV PHT (common intervention) (Fig. 1). 3.2.1. Suitability of indirect comparisons The suitability of indirect comparisons was investigated considering whether studies were suitably similar by adopting the framework for assessing exchangeability assumption proposed by ICWG [20]. 3.2.2. Comparison method The comparison between each AED and other AEDs was performed using the ORs derived from the conventional meta-analyses described above. Comparison of each binary outcome measure was performed using the log of OR and its variance derived from the conventional metaanalyses [22]. Since the logs of the OR of each meta-analysis are asymptotically normally distributed and statistically independent, the estimate of the treatment effect (i.e., IV LEV versus IV VPA) was calculated by the difference (diff) between the logs of the 2 ORs: Diff ¼ ln OR IV LEV − ln OR IV VPA : The 95% confidence interval of this estimated effect was derived from the standard error of the difference: ðð ln OR IV LEV − ln OR IV VPA Þ  ð1:96  SE ðdiff ÞÞÞ where SE (diff) = (variance (ln OR IV LEV) + variance (ln OR IV VPA))1/2. Back transformation was then performed to give the OR and its 95% CIs for the indirect comparisons.

Fig. 2. Study flow diagram. From: Moher D, Liberati A, Tetzlaff J, Altman DG, and The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6(6): e1000097.

F. Brigo et al. / Epilepsy & Behavior 64 (2016) 110–115

113

Table 1 Characteristics of included studies. Comparisons

Study

N

Patients

SE type

Dosages

VPA versus PHT

Agarwal et al. [9]

VPA: 50 PHT: 50

62 Adults/38 children (b18 yrs) 67 male/33 female

VPA: Initial bolus 20 mg/kg at 40 mg/min PHT: Initial bolus 20 mg/kg at up to 50 mg/min

VPA versus PHT

Gilad et al. [11]

VPA versus PHT VPA versus LEV

Mundlamuri et al. [6]

Adults (N18 years) 19 male/7 female Adults (N15 years) 88 male/62 female

LEV versus PHT

Chakravarthi et al. [7]

VPA: 18 PHT: 9 VPA: 50 PHT: 50 LEV: 50 LEV: 22 PHT: 22

Primary generalized seizures: 14 JME: 2 GTC SE

Adults (N15 years) 27 male/17 female

GTC SE

GTC SE Focal SE

VPA: Initial bolus 30 mg/kg in 50 ml of saline over 20 min PHT: Initial bolus of 18 mg/kg in 100 ml of saline over 20 min VPA: Initial bolus 30 mg/kg in 100 ml of saline over 15 min PHT: Initial bolus of 20 mg/kg in 100 ml at 50 mg/min LEV: Initial bolus 25 mg/kg in 100 ml of saline over 15 min LEV: Initial bolus 20 mg/kg at 100 mg/min PHT: Initial bolus of 20 mg/kg at 50 mg/min

GTC: generalized tonic–clonic; JME: juvenile myoclonic epilepsy; LEV: levetiracetam; N: number of patients included; PHT: phenytoin; SE: status epilepticus; VPA: valproate.

VPA and IV PHT [6]. The study by Mundlamuri et al. [6] compared the efficacy of three different AEDs (VPA, PHT, LEV). Characteristics of included trials are reported in Table 1; risk of bias for each included study is reported in Table 2. 4.1. Conventional meta-analysis per AED 4.1.1. IV VPA versus IV PHT There were three studies with 227 episodes of SE. No significant statistical heterogeneity among trials was detected. There was no statistically significant difference in clinical seizure cessation after drug administration between the IV VPA and the IV PHT groups (OR: 1.07; 95% CI: 0.57 to 2.03) (Fig. 3A). 4.1.2. IV LEV versus IV PHT There were two RCTs with 144 episodes of SE. We did not detect any significant statistical heterogeneity in these RCTs. There was no statistically significant difference in clinical seizure cessation after drug administration between the IV LEV and the IV PHT groups (OR: 1.18; 95% CI: 0.50 to 2.79) (Fig. 3B). 4.1.3. IV LEV versus IV VPA Only one RCT provided data on this comparison [6]. A meta-analysis was hence not feasible. After drug administration, SE was controlled

in 78% (39 out of 50) patients receiving LEV and in 68% (34 out of 50) patients treated with VPA. The difference was not statistically significant (P N 0.05). 4.2. Common reference-based indirect comparisons by combining meta-analyses of AEDs (IV LEV versus IV VPA, common comparator PHT) No statistically significant difference was found between IV LEV and IV VPA in clinical seizure cessation after drug administration (OR: 1.16; 95% CI: 0.45 to 2.97). 5. Discussion Direct comparisons meta-analyses showed no difference in clinical seizure cessation, neither between IV VPA and IV PHT nor between IV LEV and IV PHT. The indirect-comparison meta-analysis using data generated from individual comparisons versus IV PHT showed no difference in efficacy between IV LEV and IV VPA used as second-line AEDs for convulsive SE. Results of indirect comparisons are consistent with the results of the only RCT directly comparing IV LEV with IV VPA published so far [6] and, overall, suggest that both drugs share a similar efficacy profile. However, such a conclusion should be considered with caution, as the absence of evidence of a statistically significant difference in seizure

Table 2 Risk of bias in included studies. Study

Type of bias

Authors' judgment

Support for judgment

Agarwal et al. [9]

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 Selective reporting (reporting bias)

Unclear risk Unclear risk Low risk Low risk Unclear risk High risk

Gilad et al. [11]

Random sequence generation (selection bias)

High risk

Allocation concealment (selection bias) Blinding of participants and personnel (performance bias) Blinding of outcome assessment (detection bias) Incomplete outcome data Selective reporting (reporting 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 Selective reporting (reporting bias) Random sequence generation (selection bias)

High risk Low risk Low risk Low risk Unclear risk Low risk Unclear risk Low risk Low risk Low risk Unclear risk High risk

Allocation concealment (selection bias) Blinding of participants and personnel (performance bias) Blinding of outcome assessment (detection bias) Incomplete outcome data Selective reporting (reporting bias)

Unclear risk Low risk Low risk Low risk Low risk

Insufficient information Insufficient information Outcome is not likely to be influenced by lack of blinding Outcome is not likely to be influenced by lack of blinding Data on neurological outcome at discharge not reported Data on neurological outcome at discharge not reported; not prespecified how to evaluate neurological outcome at discharge Pts randomized chronologically according to the appearance in the emergency room with either VPA or PHT Explicitly unconcealed procedure Outcome is not likely to be influenced by lack of blinding Outcome is not likely to be influenced by lack of blinding No missing outcome data Insufficient information to permit judgment of ‘low’ or ‘high’ risk Computer generated randomization Insufficient information Outcome is not likely to be influenced by lack of blinding Outcome is not likely to be influenced by lack of blinding No missing outcome data Insufficient information to permit judgment of ‘low’ or ‘high’ risk Patients randomized chronologically according to the appearance in the emergency room with either LEV or PHT Insufficient information Outcome is not likely to be influenced by lack of blinding Outcome is not likely to be influenced by lack of blinding No missing outcome data Insufficient information to permit judgment of ‘low’ or ‘high’ risk

Mundlamuri et al. [6]

Chakravarthi et al. [7]

114

F. Brigo et al. / Epilepsy & Behavior 64 (2016) 110–115

Fig. 3. Conventional meta-analyses; outcome: clinical seizure cessation after drug administration A. studies comparing IV VPA versus IV PHT; B. studies comparing IV LEV versus IV PHT. CI: confidence interval; IV: intravenous; LEV: levetiracetam; M-H: Mantel–Haenszel meta-analysis; PHT: phenytoin; VPA: valproate.

control between IV LEV and IV VPA (indirect comparisons) or between IV VPA or IV LEV and IV PHT (direct comparisons) is not synonymous with evidence of no difference. In fact, the included studies and their sample sizes may have been underpowered to detect such a difference. Consequently, it is likely that the statistical analyses performed lacked sufficient statistical power to detect a difference, given the small number of patients included in the RCTs and in the meta-analyses. Of note, none of the four RCTs included in this systematic review explicitly calculated a sample size to detect predetermined critical differences between the drugs being studied. Both RCTs comparing IV LEV with IV PHT were clearly underpowered. With a sample size of 22 patients per group, the study by Chakravarthi et al. [7] had a power of only 26.8% to detect a critical difference in the response between the two drugs set at 20% with a type I error (α) of 0.05 and type II error (β) of 0.1, considering the efficacy of PHT as 50% based on the reported literature [5]. With a sample size slightly higher (50 patients per group), the RCT by Mundlamuri et al. [6] had a power of 53.3% under the same statistical assumptions. Similarly, RCTs comparing IV VPA with IV PHT were also underpowered; keeping type I error (α) 0.05 and type II error (β) 0.1, assuming an efficacy of PHT as 50% [5] and setting the critical difference in the response between the study drugs as 20%, the sample size required to have a statistical power of 90% would have been 248 (124 patients per group). After pooling together the results of the three studies comparing IV VPA with IV PHT, the resulting cumulative statistical power to detect a critical difference of 20% between the two AEDs was 87.3%. If the critical difference was set at a lower value (10%), the resulting cumulative statistical power of the meta-analysis comparing IV VPA with IV PHT was reduced to only 32.7%, with a statistical type II error (β, the probability of incorrectly concluding no statistical difference between the two AEDs) of 67.3%. As this example clearly demonstrates, meta-analysis including results from different RCTs may increase the statistical power. A metaanalysis is the statistical combination of results from two or more separate studies (pair-wise comparisons of interventions), allowing an increase in statistical power [27] and an improvement in precision, sometimes permitting us to answer questions not posed by individual studies and to settle controversies arising from conflicting claims. By their very nature, meta-analyses address broader questions than

individual studies [14]. However, a rigorous methodology should be followed when deciding to pool together results for different studies to avoid the risk of combining different kinds of studies (“apples and oranges”) in the same meta-analysis. Furthermore, results of metaanalyses should always be read with caution as the overall quality of their results inevitably depends on the quality of primary studies being included. Ideally, each RCT should be adequately powered to detect a critical difference between the drugs being studies. However, several difficulties may arise when planning clinical trials in an emergency setting. In this scenario, appropriately conducted meta-analyses — even if they cannot be considered substitute to high-quality primary trials — may prove to be useful in answering questions which cannot be answered by single studies because of an inadequate sample size. Recruiting a sufficient number of patients in RCTs conducted in an emergency setting may be challenging, especially when the patient is unconscious and hence unable to provide the consent to join a clinical trial [28]. However, taking consent from next of kin (an approach adopted in RCTs including patients with head injury and stroke) may overcome this difficulty and allow researchers to conduct informative studies [28]. Unfortunately, the lack of a sample size determination when planning RCTs is the rule in clinical research on SE [14,26]. This raises several ethical concerns regarding the utility of planning RCTs in SE with a sample size based on feasibility rather than on determination of statistical power. Conducting a RCT that has not enough people to be able to detect a clinically important difference or to estimate an effect with sufficient precision represents a waste of time and resources and may raise several ethical concerns, especially when assessing the efficacy of a drug in a condition with high morbidity and mortality such as SE. More efforts should be made to avoid the proliferation of noninformative RCTs based on underpowered sample sizes. The results of our study suggest clinical equipoise of IV LEV, IV VPA, and IV PHT, lending further support for the ESET-Trial, which is an adequately powered RCT to answer the question of what is the best choice of IV drug in benzodiazepine resistant SE [29,30]. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.yebeh.2016.09.030.

F. Brigo et al. / Epilepsy & Behavior 64 (2016) 110–115

Conflict of interest There was no funding related to the preparation of this article. Dr. Francesco Brigo received speakers΄ honoraria from Eisai, PeerVoice, and Sigma-Tau; has acted as a paid consultant to Eisai; has received travel support from UCB, Eisai, and ITALFARMACO. Prof. Eugen Trinka has acted as a paid consultant to Bial, Biogen Idec, Eisai, Ever Neuropharma, Medtronics, Takeda, Upsher-Smith, and UCB; has received speakers' honoraria from Bial, Boehringer, Eisai, GL Lannacher, and UCB Pharma; and has received research funding from Biogen Idec, Merck, Novartis, Red Bull, UCB Pharma, the European Union, FWF (Österreichischer Fond zur Wissenschaftsförderung), and Bundesministerium für Wissenschaft und Forschung. Dr. Raffaele Nardone and Dr. Nicola Luigi Bragazzi report no disclosures. References [1] Trinka E, Cock H, Hesdorffer D, et al. A definition and classification of status epilepticus — report of the ILAE Task Force on Classification of Status Epilepticus. Epilepsia 2015;56:1515–23. [2] DeLorenzo RJ, Pellock JM, Towne AR, Boggs JG. Epidemiology of status epilepticus. J Clin Neurophysiol 1995;12:316–25. [3] Trinka E, Höfler J, Leitinger M, Brigo F. Pharmacotherapy for status epilepticus. Drugs 2015;75:1499–521. [4] Trinka E, Dobesberger J. New treatment options in status epilepticus: a critical review on intravenous levetiracetam. Ther Adv Neurol Disord 2009;2:79–91. [5] Yasiry Z, Shorvon SD. The relative effectiveness of five antiepileptic drugs in treatment of benzodiazepine-resistant convulsive status epilepticus: a meta-analysis of published studies. Seizure 2014;23:167–74. [6] Mundlamuri RC, Sinha S, Subbakrishna DK, et al. Management of generalised convulsive status epilepticus (SE): a prospective randomised controlled study of combined treatment with intravenous lorazepam with either phenytoin, sodium valproate or levetiracetam—pilot study. Epilepsy Res 2015;114:52–8. [7] Chakravarthi S, Goyal MK, Modi M, Bhalla A, Singh P. Levetiracetam versus phenytoin in management of status epilepticus. J Clin Neurosci 2015;22:959–63. [8] Misra UK, Kalita J, Patel R. Sodium valproate vs phenytoin in status epilepticus: a pilot study. Neurology 2006;67:340–2. [9] Agarwal P, Kumar N, Chandra R, Gupta G, Antony AR, Garg N. Randomized study of intravenous valproate and phenytoin in status epilepticus. Seizure 2007;16:527–32. [10] Mehta V, Singhi P, Singhi S. Intravenous sodium valproate versus diazepam infusion for the control of refractory status epilepticus in children: a randomized controlled trial. J Child Neurol 2007;22:1191–7. [11] Gilad R, Izkovitz N, Dabby R, et al. Treatment of status epilepticus and acute repetitive seizures with i.v. valproic acid vs phenytoin. Acta Neurol Scand 2008;118:296–300. [12] Chen L, Feng P, Wang J, et al. Intravenous sodium valproate in mainland China for the treatment of diazepam refractory convulsive status epilepticus. J Clin Neurosci 2009;16:524–6.

115

[13] Malamiri RA, Ghaempanah M, Khosroshahi N, Nikkhah A, Bavarian B, Ashrafi MR. Efficacy and safety of intravenous sodium valproate versus phenobarbital in controlling convulsive status epilepticus and acute prolonged convulsive seizures in children: a randomised trial. Eur J Paediatr Neurol 2012;16:536–41. [14] Brigo F, Storti M, Del Felice A, Fiaschi A, Bongiovanni LG. IV valproate in generalized convulsive status epilepticus: a systematic review. Eur J Neurol 2012;19:1180–91. [15] Trinka E, Höfler J, Zerbs A, Brigo F. Efficacy and safety of intravenous valproate for status epilepticus: a systematic review. CNS Drugs 2014;28:623–39. [16] Lowenstein DH, Bleck T, Macdonald RL. It's time to revise the definition of status epilepticus. Epilepsia 1999;40:120–2. [17] Higgins JPT, Green S, editors. Handbook for systematic reviews of interventions version 5.1.0 (updated March 2011)The Cochrane collaboration; 2011. [18] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials 1986;7: 177–88. [19] Glenny AM, Altman DG, Song F, et al. International Stroke Trial Collaborative Group. Health Technol Assess 2005;9:1– [iii–iv]. [20] ICWG. Report of the indirect comparisons working group to the pharmaceutical benefits advisory committee: assessing indirect comparisons. Canberra: Australian Government Department of Health and Aging; 2009[Available from: http://www. pbs.gov.au/industry/useful-resources/PBAC_feedback_files/ICWG%20Report%20 FINAL2.pdf (accessed 27th October 2015)]. [21] Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in metaanalyses. BMJ 2003;327:557–60. [22] Bucher HC, Guyatt GH, Griffith LE, Walter SD. The results of direct and indirect treatment comparisons in meta-analysis or randomized controlled trials. J Clin Epidemiol 1997;50:683–91. [23] Otoul C, Arrigo C, van Rijckevorsel K, French JA. Meta-analysis and indirect comparisons of levetiracetam with other second-generation antiepileptic drugs in partial epilepsy. Clin Neuropharmacol 2005;28:72–8. [24] Brigo F, Igwe SC, Nardone R, Tezzon F, Bongiovanni LG, Trinka E. A common reference-based indirect comparison meta-analysis of intravenous valproate versus intravenous phenobarbitone for convulsive status epilepticus. Epileptic Disord 2013;15:314–23. [25] Zaccara G, Giovannelli F, Maratea D, Fadda V, Verrotti A. Neurological adverse events of new generation sodium blocker antiepileptic drugs. Meta-analysis of randomized, double-blinded studies with eslicarbazepine acetate, lacosamide and oxcarbazepine. Seizure 2013;22:528–36. [26] Brigo F, Nardone R, Tezzon F, Trinka E. A common reference-based indirect comparison meta-analysis of buccal versus intranasal midazolam for early status epilepticus. CNS Drugs 2015;29:741–57. [27] Cohn LD, Becker BJ. How meta-analysis increases statistical power. Psychol Methods 2003;8:243–53. [28] Prasad M, Krishnan PR, Sequeira R, Al-Roomi K. Anticonvulsant therapy for status epilepticus. Cochrane Database Syst Rev Sep 2014 10;9, CD003723. http://dx.doi.org/10. 1002/14651858.CD003723.pub3. [29] Cock HR, ESETT Group. Established status epilepticus treatment trial (ESETT). Epilepsia 2011;52(Suppl. 8):50–2. [30] Bleck T, Cock H, Chamberlain J, et al. The established status epilepticus trial 2013. Epilepsia 2013;54(Suppl. 6):89–92.

Direct and indirect comparison meta-analysis of levetiracetam versus phenytoin or valproate for convulsive status epilepticus.

The aim of this study was to conduct a meta-analysis of published studies to directly compare intravenous (IV) levetiracetam (LEV) with IV phenytoin (...
840KB Sizes 0 Downloads 11 Views