Secondary Logo

Neurology Today At the Meetings: CNS Annual Meeting

Tuesday, October 23, 2018

Study Identifies Risk Factors for Incomplete Neonatal Response to Anti-Seizure Drugs


CHICAGO—Neonates with acute symptomatic seizures who didn't receive therapeutic hypothermia, had a higher seizure burden, and had an abnormal initial interictal electroencephalogram (EEG) background were all more likely to have an incomplete response to anti-seizure medications, according to findings presented here at the Child Neurology Society annual meeting.

The analysis was based on data from a prospective, observational study of a neonatal seizure registry, including nine sites from the US. The neonates were enrolled from 2013 to 2018 and underwent continuous video EEG monitoring, said Hannah C. Glass, MD, director of neonatal critical care services at the University of California, San Francisco.

"Novel treatments are badly needed," Dr. Glass said, noting that conducting trials in this population is challenging. The need for rapid enrollment and consent, randomization, identifying which patients to enroll and the etiologies of seizures, and the fact that neonates may have already received anti-seizure medications prior to enrollment make the trials difficult to conduct, she said.

The current study included neonates younger than 44 weeks at the time of their seizure onset. Their seizures were caused by different conditions, including hypoxic-ischemic encephalopathy (HIE), ischemic stroke, or intracranial hemorrhage. The study enrolled only neonates who got a loading dose of anti-seizure medication and for whom there was adequate documentation of the response.

When EEG seizures were documented more than 30 minutes after the initial loading dose, it was considered an incomplete response. A high seizure burden was defined as seven or more recorded seizures including status epilepticus. And an abnormal initial interictal EEG background was considered an abnormal reading but not burst suppression.

Most of the neonates were term babies, about half had seizures due to HIE, and 95 percent received phenobarbital as their initial loading medication.

Sixty-seven percent of the 534 children in the study had an incomplete response, regardless of the medication, sex, or gestational age of the neonate, Dr. Glass reported.

Researchers found that risk factors for an incomplete response to medication were a lack of use of therapeutic hypothermia in term babies with HIE (p<.0005) and a high seizure burden (p<.0005). They called abnormal initial interictal EEG background a "slightly higher" risk for an incomplete response (p=.06) compared to a normal background.

"We feel that to plan efficient trials, investigators should consider including neonates with the most acute symptomatic seizure etiologies and consider stratifying or excluding lower-risk participants — for example, those who received therapeutic hypothermia, those with low initial pre-treatment seizure burden, and those with a normal initial EEG background."

Jacqueline A. French, MD, FAAN, professor of neurology who specializes in epilepsy at NYU Langone Medical Center, said the findings underscore the need for improved therapies in these children, but said there were caveats in assessing drug efficacy in the data.

"It is possible that a portion of children would have remitted spontaneously, although likely all drugs were partially effective," she said.

The choice of a therapy could be more complicated than the study suggests, she said.  Adverse events, which were not discussed, could also contribute to the therapy chosen, she noted.

"It is reassuring to know that the therapies worked across all etiologies and with similar efficacy," she said. "The fact that the majority of neonates did not respond to any of the drugs is a call to action for better therapies in this age group."


Glass HC, Shellhaas RA, Tsuchida TN, et al. Seizures in preterm neonates: A multicenter observational cohort study. Pediatr Neurol 2017; 72:19-24.