The newborn drug development initiative workshop: Summary proceedings from the neurology group on neonatal seizures.

Background: The neonatal period eclipses all other epochs of the human life span for the highest incidence of seizures. Neonatal seizures are most commonly incited by serious acute illnesses such as hypoxic-ischemic encephalopathy, birth trauma, or infection.

Objective: The goal of this article was to summarize some of the Neurology Group's discussion on neonatal seizures and their treatment with phenobarbital (PB).

Methods: Information for this article was gathered from a workshop held March 29 and 30, 2004, in Baltimore, Maryland, as part of the Newborn Drug Development Initiative. A group of national experts was invited to form the Neurology Group to review certain aspects of neonatal seizures. Literature reviews were conducted using MEDLINE searches for original research studies, commentaries, and review articles between 1980 and 2004 using the key words neonatal seizures, treatment, and phenobarbital.

Results: It has been empirically established that infants who experience seizures face substantially higher mortality and morbidity rates than those who do not. Basic research indicates that neonatal seizures themselves are not innocuous and actively contribute to adverse neurodevelopmental outcomes. Current worldwide clinical practice most often includes empiric treatment with PB for definite or suspected seizures in the newborn. Unfortunately, this common practice has never been proven by even a single, rigorous, randomized controlled trial. The Neurology Group identified the treatment of neonatal seizures with PB as an important topic in the practice of neonatal neurology for further investigation. Three possible frameworks for ethically acceptable, clinical treatment trials were explored. From these, a suitable scenario was selected-an electroencephalographer-blinded study of PB versus placebo in a homogeneous group of newborns who are at high risk of developing early subclinical electroencephalographic neonatal seizures (ENSs). Prospective video-electroencephalogram monitoring performed immediately after an insult (such as major cardiac surgery for a serious congenital heart defect) would establish the presence and number of subclinical ENSs. For a brief period of time, neonates with subclinical ENSs would be randomized to PB (dosed to match the PB-binding characteristics of the individual) or placebo. Clear criteria for escape from the study to active treatment are defined.

Conclusion: The investigation proposed here could refute or confirm the contemporary practice of PB administration as the first-line treatment of neonatal seizures.