Levetiracetam Proves Favorable to Phenobarbital for Infants with NSE

February 13, 2018

In a head-to-head comparison, levetiracetam was shown to increase the rate of infants free of monotherapy failure by 28% more than phenobarbital in pediatric patients with nonsyndromic epilepsy.

Anne T. Berg, PhD

Levetiracetam (Keppra, UCB Pharma) has been shown to significantly reduce both the treatment regimen and seizure rates for infants with nonsyndromic epilepsy (NSE), in comparison with phenobarbital.1

According to new findings from a study conducted by Anne T. Berg, PhD, Stanley Manne Children’s Research Institute at Ann & Robert H. Lurie Children’s Hospital of Chicago, and colleagues, if 100 infants who were treated with phenobarbital received levetiracetam instead, 44% would be free from monotherapy failure, compared to the standard rate of 16% observed in the study.

Berg told MD Mag, a sister publication of NeurologyLive, that although levetiracetam is already far and away the preferred initial therapy for children suffering from NSE, this is substantial analysis showing that it is superior to the next-prescribed therapy in managing short-term seizures.

The cohort study assessed the oral anticonvulsant therapy in comparison with phenobarbital in 243 infants (mean age, 4.7 months; interquartile range [IQR], 3.5 to 8.2 months) with NSE, though 88 were excluded for a variety of reasons. Of the 155 included for analysis (74 boys; 81 girls), 117 received levetiracetam and 38 received phenobarbital for a period of 6 months. The median target dose for levetiracetam was 25 mg/kg/d and for phenobarbital was 5 mg/kg/d.

The infants were administered treatment after onset of first afebrile seizure when aged between 1 month and 1 year. Infants that were treated with levetiracetam were, on average, 2.2 months older at the time of seizure onset in comparison to infants on the comparator barbiturate (mean, 5.2 months [IQR, 3.5 to 8.2] vs. 3.0 months [IQR, 2.0 to 4.4]; P <.001). Additionally, those on levetiracetam tended to be treated later after their first seizure, but no other clinically significant differences were observed.

Ultimately, 40.2% (n = 47) of those treated with levetiracetam were free from monotherapy failure, compared to 15.8% (n = 6) of those administered phenobarbital (odds ratio [OR], 3.6; 95% CI, 1.5 to 10; number needed to treat [NNT], 4.1 [95% CI, 2.0 to 17]). According to Berg and colleagues’ final estimates, levetiracetam remained superior to phenobarbital when adjusting for covariates, observable selection bias, correlation of outcomes within each center, and multiple imputation for missing data (OR, 4.2; 95% CI, 1.1 to 16; NNT, 3.5 [95% CI, 1.7 to 60]).

“There are several things at stake,” Berg said. “The seizures, especially in a developing brain, seem to have an enduring negative impact on reaching certain milestones that leads to mental cognition impairment. Getting those seizures under control early is very much so worth it.”

When analyzing for sensitivity, the phenobarbital group showed a higher proportion of early failures—18.4% (n = 7) of those on phenobarbital experienced early failure compared to 6.8% (n = 8) of those on levetiracetam (P = .07). Additionally, the investigators noted that monotherapy failed due to a perceived lack of efficacy in 81.3% of cases (n = 26 of 32) in the phenobarbital group and 85.7% (n = 60 of 70) of cases in the levetiracetam group (P = .80). When excluding early failures (OR, 4.8; 95% CI, 1.3 to 18) and when excluding individuals who failed monotherapy for reasons other than efficacy (OR, 3.6; 95% CI, 1.2 to 11)—adverse events (AEs), a planned wean, or unknown reasons—the analyses showed that levetiracetam was superior to phenobarbital.

Due to the research not being a randomized clinical trial, the investigators called for a follow-up prospective clinical trial. Berg noted that she hoped a long-term analysis of the 2 therapies would be conducted, saying that while short-term seizures are a decent gauge for infants with NSE, both therapies have behavioral and developmental effects linked to them.

Berg added that phenobarbital’s reputation among physicians and patients is aided by its history as an anticonvulsant, as it was originally brought to market in the early 1900s. Although, it has been linked to psychiatric effects after discontinuation.

“I’d like to see children followed systematically with seizure logs by both physicians and parents — not just for seizures, but for behavior because both drugs have behavioral effects,” Berg said. She noted that the US Pediatric Epilepsy Research Consortium, consisting of 17 centers, is seeking to emulate the extensive research found in other cooperative groups such as the Children’s Oncology Group, hinting that it could be part of a larger trial.


1. Grinspan ZM, Shellhass RA, Coryell J, et. al. Comparative effectiveness of levetiracetam vs phenobarbital for infantile epilepsy. JAMA Pediatr. 2018;172(4):352-360.


: 10.1001/jamapediatrics.2017.5211.