{"id":95847,"date":"2025-02-14T16:20:57","date_gmt":"2025-02-14T16:20:57","guid":{"rendered":"https:\/\/touchneurology.com\/?p=95847"},"modified":"2025-02-14T16:20:57","modified_gmt":"2025-02-14T16:20:57","slug":"phenobarbital-versus-levetiracetam-as-a-first-line-treatment-of-neonatal-seizures-a-systematic-review","status":"publish","type":"post","link":"https:\/\/touchneurology.com\/epilepsy\/journal-articles\/phenobarbital-versus-levetiracetam-as-a-first-line-treatment-of-neonatal-seizures-a-systematic-review\/","title":{"rendered":"Phenobarbital Versus Levetiracetam as a First-line Treatment of Neonatal Seizures: A Systematic Review"},"content":{"rendered":"
Seizures are one of the most frequent neurological disorders in neonates \u2212 the incidence of seizures in infants born at term is 1\u20133 per 1,000 live births, and is even higher in both preterm and very-low-birth-weight infants at 1\u201313 per 1,000 live births.1<\/sup><\/span>\u00a0Seizures may signify serious malfunction of, or damage to, the immature brain and constitute a neurological emergency demanding urgent management.<\/p>\n Although there is no consensus on the ideal treatment of neonatal seizures, phenobarbital (PHB) remains the most popular first-line treatment.2,3<\/sup><\/span>\u00a0However, PHB and phenytoin often fail to control seizures with the first dose.4,5<\/sup><\/span><\/p>\n Moreover, PHB is associated with several harmful side effects, both immediate \u2013 such as sedation, hypotension and respiratory depression \u2013 and long term, with concerns about its potential to adversely affect psychomotor development and neurological outcomes in the developing brain.6,7<\/sup><\/span>\u00a0For these reasons, the search for an alternative first-line therapy for neonatal seizures began in the last two decades.<\/p>\n Levetiracetam (LEV) is a newer anti-seizure medication (ASM) approved for the treatment of focal seizures in Europe since 2000 in infants older than 28 days and is increasingly being used to treat neonatal seizures.8<\/sup><\/span>\u00a0LEV appears to have excellent tolerability in neonates.9<\/sup><\/span>\u00a0It has good efficacy and an excellent safety profile.10,11 <\/sup><\/span>Unlike PHB, it does not cause sedation and has no recorded adverse cardiovascular effects. Although uncommon, a recognized side effect of LEV is increased irritability and tiredness. In this systematic review, we aimed to compare the efficacy and safety of both medications in the treatment of neonatal seizures.<\/p>\n This systematic review was conducted according to the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines.12<\/sup><\/span>\u00a0An agreed protocol was established and followed. The details of the PRISMA checklist are provided in the supplemental table (Table S1<\/span><\/span><\/span><\/span>).<\/p>\n We included all relevant studies, including observational and randomized controlled trials, that reported on the outcomes of interest for this review. The included articles compared the use of PHB versus LEV in the treatment of neonatal seizures. Studies that reported the use of a single agent, either PHB or LEV, were excluded from our analysis. There were no limitations regarding the publication date or language.<\/p>\n We conducted an extensive search of the online published literature in January 2024 for potentially relevant studies. We used the following search terminologies in various combinations to produce our search results ([\u2018neonates\u2019 OR \u2018new-born\u2019 OR \u2018new born\u2019 OR \u2018full term\u2019 OR \u2018full-term\u2019] AND [\u2018seizures\u2019 OR \u2018convulsions\u2019 OR \u2018fits\u2019] AND [\u2018phenobarbital\u2019] AND [\u2018levetiracetam\u2019]). We searched the following online databases: PubMed, Ovid\u00a0MEDLINE<\/span><\/span>, SCOPUS, EMBASE, CINAHL, Google Scholar and the Cochrane\u00a0L<\/span>ibrary. We also searched the grey literature by searching the bibliographies of included studies for potential citations. There were no restrictions applied to the search in terms of date, language or publication types. We limited our search to studies performed on humans only.<\/p>\n The eligibility of studies produced from the initial search to be considered for this review was assessed by the first author (Hiba\u00a0Bashar) and the second author (Khalid\u00a0Bashar), according to the pre-agreed protocol. The abstracts from the initial search were read in full before a decision was made on whether to exclude an article. In the cases where the first and second authors disagreed about a certain study, the article was examined in full. Any remaining disputes were settled following a consultation with the senior author (Alison\u00a0Walker). After excluding irrelevant studies, the full articles of the remaining studies were obtained and analysed for outcomes of interest to this review.<\/p>\n The main objective of our review was to compare the efficacy of both drugs in controlling neonatal seizures during the acute phase. We also compared PHB and LEV in terms of safety by assessing the rate of complications and adverse reactions. Our secondary objective focused on the effects of PHB and LEV on long-term development, including motor and language development, as well as cognitive function.<\/p>\n The first (Hiba\u00a0Bashar) and second (Khalid\u00a0Bashar) authors independently assessed data extracted from individual studies. All relevant data were extracted on a Microsoft Excel spreadsheet. The first two authors met regularly to discuss progress and settle disagreements about the included data. The senior author was then consulted to resolve any possible disputes or ambiguities related to the collected data and to settle any possible differences between the first two authors.<\/p>\n Demographic data were also recorded, including gestational age, gender, types of seizures, aetiology of seizures wherever possible, drug dose and the duration of follow-up. In addition, whenever possible, data were pooled to perform meta-analysis.<\/p>\n The assessment of the quality of individual studies was performed by the first author (Hiba\u00a0Bashar). As most of the studies included in this review were retrospective observational studies, we decided to use the Downs and Black tool to assess individual articles.13<\/sup><\/span>\u00a0This tool has widely been validated in the literature and consists of 27 different questions assessing the internal and external validity of included studies, along with bias and confounding factors. The original tool uses five questions to assess the inclusion of sample size calculations. For simplicity, we chose to combine those questions regarding sample size calculations into one. Therefore, our modified tool will result in a maximum score of 27 points, with lower scores indicating poor quality studies. The results of the quality assessment are provided in a supplemental table\u00a0<\/span><\/span>(Table S2<\/span>).<\/p>\n Data from seven of the individual studies were pooled into a meta-analysis using Review Manager version 5.2.14<\/sup><\/span>\u00a0We opted to use the random-effects model of DerSimonian and Laird to compare categorical data, which were pooled as odds ratios.15<\/sup><\/span>\u00a0The Cochran\u2019s Q-test was used to assess statistical heterogeneity.16<\/sup><\/span>\u00a0A cut-off of\u00a0<\/em>p<0.05 was used to report statistically significant differences.<\/p>\n The PRISMA flow diagram summarizes the process used in the literature search and the selection of studies for inclusion in this review (Figure 1<\/span><\/span>). First, the search criteria detailed earlier were used to create the primary list of potential studies. This generated 181 articles. Restricting the search to studies performed on humans reduced the number to 138 articles. We did not restrict our search by language; however, all the included articles were published in English. In addition, there were no restrictions in terms of publication date.<\/p>\n Figure 1: <\/span><\/span>Preferred Reporting Items for Systematic Review and Meta-analysis flow<\/p>\n LEV = levetiracetam; PHB = p<\/span>henobarbital.<\/em><\/p>\n<\/div>\n We screened the titles and abstracts of all 138 eligible articles and reduced the number of potential studies to 17. After reading the full articles, we found that only six satisfied our agreed protocol.17\u201322<\/sup><\/span>\u00a0These six articles were then analysed in detail, and the findings are discussed below.<\/p>\n The total number of patients included in all studies was 420; of these, 200 were primarily treated with PHB, while 220 were treated with LEV as the first-line treatment. There were 186 (44.2%) female patients and 234 (55.8%) male patients. The characteristics of individual studies are summarized in\u00a0Table 1<\/em><\/span>.17\u201323<\/sup><\/span><\/p>\n Table 1: <\/span>Characteristics of studies17\u201323<\/sup><\/span><\/p>\n Studies<\/p>\n<\/td>\n Design<\/p>\n<\/td>\n Inclusion criteria<\/p>\n<\/td>\n Exclusion criteria<\/p>\n<\/td>\n Overview of cases and controls<\/p>\n<\/td>\n Number and characteristics of cases (LEV)<\/p>\n<\/td>\n Number and characteristics of controls (PHB)<\/p>\n<\/td>\n Comment<\/p>\n<\/td>\n<\/tr>\n<\/thead>\n Rao et al. (2018)<\/span>17<\/sup><\/span><\/span><\/p>\n<\/td>\n A retrospective cohort study<\/p>\n<\/td>\n 44 patients who had VEEG-confirmed seizures; inclusion criteria were (1<\/span>) greater than 36 weeks of gestational age, (2<\/span>) less than 6 h of age, and (3<\/span>) underwent therapeutic hypothermia with continuous VEEG monitoring<\/p>\n<\/td>\n Clinical seizures without electrographic correlation<\/p>\n<\/td>\n 78 newborn infants with HIE; 44 patients exhibited VEEG-confirmed seizures, of whom 34 became seizure-free after treatment; the remaining 10 patients died<\/p>\n<\/td>\n Number of cases: 20. Gestational age: 39.0 (37.5\u201340.4) weeks. Birth weight: 3,123 (2,837\u20133,461<\/span>) g. M\/F<\/span>=14\/6 patients. APGAR score: 1 min; 1.0 (0.0\u20131.0) min. APGAR score: 5 min; 3.0 (2.5\u20134.0) min. APGAR score: 10 min; 5.0 (3.0\u20136.0) min<\/p>\n<\/td>\n Number of cases: 24. Gestational age: 38.7 (38.0\u201339.7) weeks. Birth weight: 3,376 (3,081\u20133,565<\/span>) g. M\/F<\/span>=15\/9 patients. APGAR score, 1 min; 1.0 (0.0\u20132.0). APGAR score, 5 min; 3.0 (1.5\u20134.0). APGAR score, 10 min; 5.0 (2.0\u20136.0). The PHB group exhibited higher injury severity, with both higher ad hoc severity score (<\/em>p=0.047) and higher proportion with short-term mortality (<\/em>p=0.01)<\/p>\n<\/td>\n Patients received PHB only (n=13), LEV only (n=18), LEV followed by PHB within 24 h of first LEV administration (n=2), or PHB followed by LEV within 24 h (n=10) after VEEG confirmation of seizures. The type of cooling method was not associated with demographic characteristics, seizure burden, injury severity, or the use of LEV versus PHB (all\u00a0<\/em>p>0.05). Accordingly, the selective head cooling and whole-body cooling subgroups were combined in all analyses<\/p>\n<\/td>\n<\/tr>\n Falsaperla et al. (2019)<\/span>18<\/sup><\/span><\/span><\/p>\n<\/td>\n A randomized, one-blind prospective study<\/p>\n<\/td>\n 30 neonates were included. Inclusion criteria were (1<\/span>) term neonates and (2<\/span>) seizures manifesting within the first 28 days of life. Seizures confirmed by EEG<\/p>\n<\/td>\n 1. Requiring therapeutic hypothermia. 2. Seizures secondary to transient metabolic disorders. 3. Neonates with a positive history of maternal drug ingestion. 4. Those who received more than one anti-convulsant medication. 5. Those neonates in whom LEV was used as second-line therapy were excluded<\/p>\n<\/td>\n 30 term neonates with seizures admitted to the neonatal intensive care unit of S. Bambino Hospital, University Hospital \u2018Policlinico Vittorio Emanuele\u2019, Catania, Italy, from February 2016 to February 2018 were randomized to receive PHB or LEV<\/p>\n<\/td>\n Number of cases: 15. Gestational age: 38.13 \u00b1 1.24. Sex (F\/M): 4\/11. Prenatal anomalies 40%. APGAR score: 1 min; 7.66 \u00b1 1.29. APGAR score: 5 min; 9.13 \u00b1 1.12<\/p>\n<\/td>\n Number of cases: 15. Gestational age: 38.33 \u00b1 1.04. Sex (F\/M): 8\/7. Prenatal anomalies 40%. APGAR score: 1 min; 8.66 \u00b1 0.89. APGAR score: 5 min; 9.03 \u00b1 0.84<\/p>\n<\/td>\n The underlying aetiologies for seizure onset included hypoxic-ischaemic encephalopathy not requiring therapeutical hypothermia, stroke and central nervous system infections<\/p>\n<\/td>\n<\/tr>\n Arican et al. (2020)<\/span>19<\/sup><\/span><\/span><\/p>\n<\/td>\n A retrospective non-randomized study<\/p>\n<\/td>\n Term infants who were treated using PHB or LEV monotherapy as the first-line treatment for neonatal seizures and followed up in a paediatric neurology outpatient clinic. Seizures confirmed by EEG<\/p>\n<\/td>\n Infants were excluded from the study group if they had received a second anti-epileptic drug in the neonatal intensive care unit or during the outpatient follow-up<\/p>\n<\/td>\n The study group consisted of 62 infants who received monotherapy with PHB (n=22) and LEV (n=40); the mean duration of monotherapy was 8 \u00b1 6 months. The mean time of the follow-up period in the paediatric neurology outpatient clinic was 19 \u00b1 7 months (range: 9\u201342 months)<\/p>\n<\/td>\n Number of cases: 40. Age (months; means \u00b1 SD) 18 \u00b1 7. Sex (F\/M) 21\/19<\/p>\n<\/td>\n Number of cases: 22. Age (months; means \u00b1 SD) 20 \u00b1 8. Sex (F\/M) 13\/9<\/p>\n<\/td>\n Neurodevelopmental assessments were carried out using the BSID-III by one of the co-authors of this research\u00a0Mete Atasever<\/span>. Neurodevelopmental outcomes were assessed by motor, cognitive and language performance in the BSID-III<\/p>\n<\/td>\n<\/tr>\n Liu et al. (2020)<\/span>20<\/sup><\/span><\/span><\/p>\n<\/td>\n A retrospective cohort study<\/p>\n<\/td>\n Inclusion criteria: (1) gestational age \u226535 weeks and birth weight >1,800 g; (2) occurrence of the first seizure in the first 28 days of life (in term infants) or by 44 weeks corrected gestational age (in preterm infants); (3) written consent from the parent(s)\/guardian(s). Seizures were confirmed by EEG<\/p>\n<\/td>\n Exclusion criteria: 1. PHB or LEV treatment duration of less than 3 days. 2. Seizures due to hypoglycaemia, hypocalcaemia, hypomagnesaemia or other electrolyte disorders<\/p>\n<\/td>\n 125 newborns hospitalized for neonatal seizures between\u00a0October 2014<\/span>\u00a0and October 2018 at the Children\u2019s Hospital of Chongqing Medical University<\/p>\n<\/td>\n Number of cases: 59. Gestational age (weeks), mean (SD): 39.1 (1.1). There were no significant differences in the age of onset, seizure frequency, sex, gestational age, parturition or body weight between the PHB and LEV groups before admission (<\/em>p>0.05 for all)<\/p>\n<\/td>\n Number of cases: 66. Gestational age (weeks), mean (SD): 39.0 (1.0)<\/p>\n<\/td>\n There were 28 patients who completed short-term treatment and 38 patients who completed long-term treatment in the PHB group; there were 12 patients who completed short-term treatment and 47 patients who completed long-term treatment in the LEV group<\/p>\n<\/td>\n<\/tr>\n Sharpe et al. (2020)<\/span>21<\/sup><\/span><\/span><\/p>\n<\/td>\n A multicentre, randomized, blinded, controlled study<\/p>\n<\/td>\n Patients were term infants of a corrected gestational age between 36 and 44 weeks (2 weeks of age) with a weight of at least 2.2 kg. VEEG-confirmed seizures<\/p>\n<\/td>\n Patients were excluded if they had received any previous anti-convulsants (with the exception of short-acting benzodiazepines administered for sedation 24 h before enrolment), if the serum creatinine level was 1.6 mg\/dL, or if seizures were due to correctable metabolic abnormalities. Patients in whom death was imminent were excluded; patients in whom EEG monitoring could not be commenced before the need to treat definite clinical seizures were not recruited<\/p>\n<\/td>\n Randomly assigned (n=106)<\/p>\n<\/td>\n Number of cases: 64. Sex (F\/M): 31\/33. Gestational age: 39.3 (1.3). 5 min APGAR score, mean (SD): 6.52 (3.01). Cord pH, mean (SD): 7.07 (0.2)<\/p>\n<\/td>\n Number of cases: 42. Sex (F\/M): 24\/18. Gestational age: 39.1 (1.3). 5 min APGAR score, mean (SD): 6.47 (2.4). Cord pH, mean (SD): 7.15 (0.17)<\/p>\n<\/td>\n NEOLEV2 (ClinicalTrials.gov identifier:\u00a0NCT01720667<\/span>) was an investigator-initiated, FDA-funded study23<\/sup><\/span><\/p>\n<\/td>\n<\/tr>\n Thibault et al. (2020)<\/span>22<\/sup><\/span><\/span><\/p>\n<\/td>\n A retrospective single-centre study<\/p>\n<\/td>\n Neonates (\u226430 days of age, corrected gestational age \u226444 weeks) with electrographically confirmed seizures following cardiac surgery from 15 June 2012 to 31 December 2018; following the implementation of routine 48 h cEEG in all neonates following cardiac surgery with CPB<\/p>\n<\/td>\n Authors did not provide exclusion criteria<\/p>\n<\/td>\n Neonates with electrographically confirmed seizures managed with anti-seizure medication after cardiac surgery from 15 June 2012 to 31 December 2018<\/p>\n<\/td>\n Number of cases: 22. Sex (F\/M): 12\/10. Gestational age: 38 (37\u201339<\/span>) weeks. Premature, n (%): 8 (37). Age at surgery, 4 (2\u20136<\/span>) days. CPB time, 68 (45\u201389<\/span>) min. Status epilepticus, n (%): 3 (14)<\/p>\n<\/td>\n Number of cases: 31. Sex (F\/M): 11\/20. Gestational age: 38 (37\u201339<\/span>) weeks. Premature, n (%): 11 (35<\/span>). Age at surgery, 5 (3\u20138<\/span>) days. CPB time: 46 (39\u201362<\/span>) min. Status epilepticus, n (%): 10 (32)<\/p>\n<\/td>\n Because of their anti-seizure properties, data were also collected regarding the concomitant administration of midazolam and ketamine for sedation or analgesia; no difference in concomitant midazolam; more neonates in LEV group received concomitant ketamine<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/div>\n APGAR<\/span>\u00a0=\u00a0Appearance, Pulse, Grimace, Activity and Respiration<\/span>;<\/span>BSID<\/span>\u00a0=\u00a0Bayley Scales of Infant Development<\/span>;<\/span>cEEG<\/span>\u00a0=\u00a0continuous electroencephalographic<\/span>;<\/span>CPB<\/span>\u00a0=\u00a0cardiopulmonary bypass<\/span>;<\/span>EEG<\/span>\u00a0=\u00a0electroencephalogram<\/span>;<\/span>F<\/span>\u00a0=\u00a0female<\/span>;<\/span>FDA<\/span>\u00a0=\u00a0US Food and Drug Administration<\/span>;<\/span>g<\/span>\u00a0=\u00a0grams<\/span>;<\/span>h<\/span>\u00a0=\u00a0hours<\/span>;<\/span>HIE<\/span>\u00a0=\u00a0hypoxic-ischaemic encephalopathy<\/span>;<\/span>LEV<\/span>\u00a0=\u00a0levetiracetam<\/span>;<\/span>M<\/span>\u00a0=\u00a0male<\/span>;<\/span>min<\/span>\u00a0=\u00a0minutes<\/span>;<\/span>PHB<\/span>\u00a0=\u00a0phenobarbital<\/span>;<\/span>SD<\/span>\u00a0=\u00a0standard deviation<\/span>;<\/span>VEEG<\/span>\u00a0=\u00a0video-electroencephalogram<\/span>.<\/span><\/em><\/p>\n<\/div>\n<\/div>\n<\/div>\n Rao\u00a0et al<\/span>.<\/em>\u00a0assessed 44 patients with confirmed seizures on video-electroencephalogram (VEEG).17<\/sup><\/span>\u00a0These patients were full-term neonates with mild-to-severe hypoxic-ischaemic encephalopathy (HIE) who underwent therapeutic hypothermia. The type of cooling used, whether whole body or selective head cooling, was not associated with any difference in terms of treatment effects. They found that the use of LEV was associated with a shorter duration to cessation of seizures compared with PHB (hazard ratio [HR]=2.58,\u00a0<\/em>p=0.007). This remained the case even after excluding patients who crossed over during the study. The severity of HIE was found to be an independent predictor of prolonged treatment duration (HR=0.16, p=0.001).<\/p>\n Liu et al. analysed 125 neonates retrospectively (PHB=66 and LEV=59).20<\/sup><\/span>\u00a0They set two time points for their analysis: 3 days (defined as short term) and 16 weeks (defined as long term). They found no significant difference between the two drugs in the short-term analysis (<\/em>p>0.05). However, the neurodevelopment of those treated with LEV, as assessed by Gesell scores, was significantly better than those treated with PHB (<\/em>p=0.026).<\/p>\n Sharpe\u00a0et al<\/span>.<\/em>\u00a0performed a multicentre, randomized, double-blinded study to assess both the safety and efficacy of LEV against PHB as a first-line treatment for neonatal seizures in term infants.21<\/sup><\/span>\u00a0All patients were randomly assigned and continuously monitored by VEEG. They reported that patients treated with PHB had a better rate of freedom from seizures within 24\u00a0<\/span>h (<\/em>p=0.001). However, increasing the LEV dose from 40 to 60 mg\/kg resulted in a 7.5% increased efficacy of LEV in controlling seizures at 24 h.<\/p>\n Thibault\u00a0et al<\/span>.<\/em>\u00a0looked retrospectively at full-term neonates undergoing corrective cardiac surgery within 30 days of life.22<\/sup><\/span>\u00a0Seizure activity was confirmed on electroencephalogram. In their cohort, 31 neonates received PHB, while 22 received LEV as first-line therapy. Both drugs had similar efficacy in controlling seizures (PHB=13\/31 versus LEV=12\/22;\u00a0<\/em>p=1.0). Furthermore, although LEV required less time (1.5\u00a0<\/span>h) to control seizures compared with PHB (5.2\u00a0<\/span>h), the difference was not statistically significant (<\/em>p=0.35).<\/p>\n Two of the included studies reported data on the cessation of seizures within 24\u00a0<\/span>h in a way that allowed for a meta-analysis calculation.21,22<\/sup><\/span>\u00a0The two studies had 136 patients, with 75 neonates receiving LEV as first-line medication compared with 61 neonates in the PHB group. In the LEV group, 27 out of 75 neonates responded to treatment compared with 42 of 61 in the PHB group. This difference was not statistically significant in the pooled analysis (pooled odds ratio=0.29; 95% confidence interval [CI]: 0.03\u20132.45;\u00a0<\/em>p=0.26;\u00a0Figure 2<\/span><\/span>). There was significant heterogeneity detected (Cochran\u2019s Q=7.65; degree of freedom [df]=1;\u00a0<\/em>p=0.006;\u00a0I<\/em>2<\/sup>=87%).<\/p>\n Figure 2: <\/span>Cessation of seizures<\/p>\n CI = confidence interval; M-H =\u00a0Mantel\u2013Haenszel<\/span>.<\/em><\/p>\n<\/div>\n Three of the included articles reported on outcomes related to neurodevelopment.18\u201320<\/sup><\/span>\u00a0Falsaperla\u00a0et al<\/span>.<\/em>\u00a0performed a randomized study (one-blinded) to investigate the use of PHB and LEV as first-line medications in the treatment of seizures in a group of term neonates within 28 days of birth.18<\/sup><\/span>\u00a0They used the Hammersmith Neonatal Neurological Examination (HNNE) in their assessment. They measured baseline scores at recruitment and compared them with scores measured 1 month following treatment. The overall scores improved significantly after 1 month of treatment when compared with scores at baseline in the LEV group (<\/em>p=0.001). The same was true for individual scores used to measure tone and postures (<\/em>p=0.05), reflexes (<\/em>p=0.01), and orientation and behaviour (<\/em>p=0.02). There was no significant improvement in the HNNE score in the PHB group in this study.18<\/sup><\/span><\/p>\n Arican\u00a0et al<\/span>.<\/em>\u00a0studied term infants treated for seizures with either LEV (n=40) or PHB (n=22) as single agents.19<\/sup><\/span>\u00a0They treated their patients for a mean duration of 8 months and used the Bayley Scales of Infant Development (BSID-III) to assess their development. The difference between LEV and PHB groups was not statistically significant in any of the BSID-III score components (cognitive, motor and language).19<\/sup><\/span><\/p>\n Liu\u00a0et al<\/span>.<\/em>\u00a0used the Gesell score at 16 weeks to investigate the effects of both LEV (n=47) and PHB (n=38) on neurodevelopment.Methods<\/h1>\n
Eligibility criteria<\/h2>\n
Search strategy<\/h2>\n
Data\u00a0c<\/span>ollection<\/h2>\n
Quality assessment for risk of bias<\/h2>\n
Data analysis<\/h2>\n
Results<\/span><\/h1>\n
Study selection<\/h2>\n
<\/p>\n
Participants<\/h2>\n
\n\n
\n \n\n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n Cessation of seizures<\/h2>\n
<\/p>\n
Neurological assessment<\/h2>\n