PDF icon PDF herunterladen
Neuropediatrics 2016; 47(05): 281-282
DOI: 10.1055/s-0036-1586224
Editorial Comment
Georg Thieme Verlag KG Stuttgart · New York

Of Phenobarbital for Neonatal Seizures: Response Rate and Refractoriness Predictors[*]

Georgia Ramantani
1   Division of Child Neurology, University Children's Hospital, Steinwiesstrasse 75, 8032 Zurich, Switzerland
2   Swiss Epilepsy Centre, Zurich, Switzerland
› Institutsangaben
Weitere Informationen

Publikationsverlauf

Publikationsdatum:
01. August 2016 (online)

Auch verfügbar aufeRef
   Lizenzen und Reprints
Preview

Neonatal seizures constitute the most common and grave manifestation of neurologic dysfunction in the first weeks of life, reflecting a wide range of pre-, peri- or postnatal disorders of the central nervous system. The rapid developments in neonatal care in the past decades have facilitated a decrease in mortality, particularly in premature neonates, and at the same time, altered the scope of insults to which the immature brain is exposed. This has steered a shift from mortality to morbidity for neonates with seizures, since surviving neonates have an increased risk for postneonatal epilepsy and long-term neurodevelopmental disability.[1] [2] [3] [4] Neonatal seizures are acute symptomatic events, commonly triggered by hypoxic-ischemic encephalopathy (HIE), stroke, or infection, and are only rarely associated with epilepsy per se. However, there is a growing body of evidence supporting that seizures themselves are aggravating brain damage, particularly in HIE,[5] [6] and therefore require treatment. The detection of seizures in neonates based on clinical presentation alone is hampered by the variability of their clinical expression, together with the high rate of electrographic-only seizures. The widespread use of continuous electroencephalography (EEG) or amplitude-integrated (a) EEG in the past years has facilitated a decrease in the latency to diagnosis and initiate treatment.[7]

Despite the urgent need for an effective as well as age-specific treatment for neonates, no clearly defined and evidence-based recommendations are currently available, as emphasized in the World Health Organization Guidelines.[8] The avalanche of antiepileptic drug (AED) development for older children and adults witnessed in the past decades had no (or little) impact on the prevalent management strategies for this most vulnerable age group.[9] Recent surveys indicate that phenobarbital remains the AED of choice among both neonatologists and child neurologists,[10] [11] [12] although seizure control is reportedly achieved in less than half of the cases, with an even lower efficacy regarding electrographic seizures.[13] [14] [15] Phenobarbital, chosen mainly due to extensive vast clinical experience as well as its inclusion in randomized controlled studies, has been shown to increase electroclinical dissociation,[14] thus providing false reassurance, unless EEG monitoring is used. Most importantly, there is evidence that phenobarbital itself may cause neurodevelopmental impairment and increase neuronal apoptosis.[16] In the meantime, several newer AEDs have been applied to the treatment of neonatal seizures,[17] though data from randomized controlled trials is still missing. Booth and Evans state in their Cochrane review regarding the treatment of neonatal seizures that “there is little evidence from randomized controlled trials to support the use of any of the anticonvulsants currently used in the neonatal period”.[18]

In this issue of Neuropediatrics, Spagnoli et al[19] bring this old drug, phenobarbital, back to the spotlight. In a retrospectively studied cohort of 91 neonates presenting with seizures to the neonatal intensive care unit (NICU), the authors have quantified the response rate to phenobarbital and identified predictors of lack of efficacy. Interestingly, phenobarbital response rates of 63% were higher than in past studies, most probably due to the exclusion of neonates with status epilepticus, previously highlighted as a predictor of refractoriness by the very same study group.[3] In this study, abnormal EEG background activity and the presence of electrographic-only seizures in neonates have been confirmed as adverse outcome predictors.

Poor response to AEDs has been identified as a predictor of adverse outcomes, particularly of postneonatal epilepsy in several studies.[2] [15] [20] The role of abnormal background EEG activity as a robust indicator of outcome following neonatal seizures has been recognized before, in contrast to the presence of isolated sharp waves.[21] In particular, the presence of an abnormal background EEG activity has been associated with the lack of response to phenobarbital,[13] [14] whereas responders presented with normal or mildly abnormal EEG background.[22] On the contrary, the identification of electrographic-only seizures in neonates as an adverse outcome predictor is particularly alarming, as it exposes a potential treatment gap and its dire consequences. At this time of ongoing debate regarding the treatment of electrographic-only seizures,[11] [23] this finding, together with the indication of adverse neurological outcomes in neonates with electrographic-only seizures[24] underlines the necessity for continuous EEG surveillance in the NICU to timely detect neonatal seizures and initiate AED administration. Although randomized controlled trials and large longitudinal cohort studies are a prerequisite for true progress in neonatal seizure management, optimizing seizure identification by the implementation of continuous EEG monitoring in itself may have the potential to improve outcomes.

* This article is an editorial on “Phenobarbital for neonatal seizures: Response rate and predictors of refractoriness” by Spagnoli et al (Neuropediatrics 2016;47(5):318–326).