Abstract
Contemporary Profile of Seizures in Neonates: A Prospective Cohort Study
Glass HC, Shellhaas RA, Wusthoff CJ, Chang T, Abend NS, Chu CJ, Cilio MR, Glidden DV, Bonifacio SL, Massey S, Tsuchida TN, Silverstein FS, Soul JS; Neonatal Seizure Registry Study Group. J Pediatr 2016;174:98–103.
OBJECTIVE: To determine the contemporary etiology, burden, and short-term outcomes of seizures in neonates monitored with continuous video-electroencephalogram (cEEG). STUDY DESIGN: We prospectively collected data from 426 consecutive neonates (56% male, 88% term) ≤44 weeks’ postmenstrual age with clinically suspected seizures and/or electrographic seizures. Subjects were assessed between January 2013 and April 2015 at 7 US tertiary care pediatric centers following the guidelines of the American Clinical Neurophysiology Society for cEEG for at-risk neonates. Seizure etiology, burden, management, and outcome were determined by chart review by the use of a case report form designed at study onset. RESULTS: The most common seizure etiologies were hypoxic-ischemic encephalopathy (38%), ischemic stroke (18%), and intracranial hemorrhage (11%). Seizure burden was high, with 59% having ≥7 electrographic seizures and 16% having status epilepticus; 52% received ≥2 antiseizure medications. During the neonatal admission, 17% died; 49% of survivors had abnormal neurologic examination at hospital discharge. In an adjusted analysis, high seizure burden was a significant risk factor for mortality, length of hospital stay, and abnormal neurological examination at discharge. CONCLUSIONS: In this large contemporary profile of consecutively enrolled newborns with seizures treated at centers that use cEEG per the guidelines of the American Clinical Neurophysiology Society, about one-half had high seizure burden, received ≥2 antiseizure medications, and/or died or had abnormal examination at discharge. Greater seizure burden was associated with increased morbidity and mortality. These findings underscore the importance of accurate determination of neonatal seizure frequency and etiology and a potential for improved outcome if seizure burden is reduced.
Wietstock SO, Bonifacio SL, Sullivan JE, Nash KB, Glass HC. J Child Neurol 2016;3:328–332.
The objective of this study was to determine the diagnostic yield of continuous video electroencephalographic (EEG) monitoring in critically ill neonates in the setting of a novel, university-based Neonatal Neurocritical Care Service. Patient demographic characteristics, indication for seizure monitoring, and presence of electrographic seizures were obtained by chart review. Among 595 patients cared for by the Neonatal Neurocritical Care Service, 400 (67%) received continuous video EEG. The median duration of continuous video EEG monitoring was 49 (interquartile range = 22–87) hours. Electrographic seizures were captured in 105 of 400 (26% of monitored patients) and of those, 25 of 105 (24%) had no clinical correlate. In addition, 52 of 400 subjects (13%) were monitored due to paroxysmal events concerning for seizures, but never had electrographic seizures. Continuous video EEG monitoring helped confirm or rule out ongoing seizures in more than one-third of the cases. This finding helps to support the use of continuous video EEG in critically ill neonates.
Commentary
Neonates are a high-risk population for seizures. The teaching has been that until proven otherwise, most neonatal seizures are due to an acute symptomatic neurologic insult, which implies a time-sensitive acute intervention. The problem, however, is that neonatal seizures are difficult to diagnose by clinical bedside observation. Treatments purely guided by clinical observations often result in overtreatment of nonparoxysmal events or undertreatment of unappreciated subclinical EEG seizures.1,2 Continuous video EEG (cVEEG) has, therefore, become a promising diagnostic tool in newborn intensive care units (NICUs), and a large body of literature has emerged and helped refine the pragmatic use of this costly and personnel-intensive test.3,4 Common indications for cVEEG in the NICU are paroxysmal events suspicious for seizures, unexplained alteration in sleep-awake cycle or behavior state (encephalopathy), and the presence of high-risk diagnoses/interventions—hypoxic ischemic encephalopathy, intracranial hemorrhage, acute toxic metabolic cause, hypothermia, extracorporeal membrane oxygenation, and certain congenital heart diseases. Recently, guidelines for use of cVEEG in the NICU were also published.5 These two NICU studies provide further confirmation by showing a high diagnostic yield of cVEEG in seizure detection in neonates at risk, demonstrating the utility of cVEEG in guiding acute treatment, and providing further insights into the etiologies and demographic characteristics of neonatal seizures. In addition, the study of Glass et al. attempted to examine the relationship of seizures to short-term morbidity (neurologic exam at discharge).
The report by Wietstock et al. was a single-center NICU study. Of 595 NICU patients seen by the neurologist in the NICU during 2008 to 2012, cVEEG was ordered in 400 (67%). The median duration of cVEEG was 49 hours (range, 22–87). The cVEEG cohort had a high number of full-term neonates (66%), and a majority (66%) were transferred from an outside facility for tertiary care; hence, the variable timing (unknown) of initiating cVEEG was based on the clinical and logistic circumstances. Of the monitored neonates, 26% (105/400) had EEG seizures, and of those with EEG seizures, 24% (25/105) had no clinical correlate. In addition, 13% (52/400) had paroxysmal seizure-like events, but cVEEG did not confirm a corresponding ictal EEG pattern. Phenobarbital was the most commonly used antiepileptic drug, and 23% (93/400) of neonates received phenobarbital on clinical suspicion of seizures before cVEEG was initiated. The most common diagnoses associated with seizures in this study were hypoxic-ischemic encephalopathy (HIE), ischemic stroke, intracranial hemorrhage, and infection; the highest frequency of subclinical EEG seizures was seen in the HIE group. Neonatal epilepsy as a chronic condition due to brain malformations and genetic etiologies was infrequent.
The study by Glass et al. recruited neonates from seven tertiary care NICU centers that participated in a Neonatal Seizure Registry. The study cohort consisted of 426 consecutive neonates at the seven NICUs who were suspected to have clinical seizures and/or electrographic seizures during 2013 to 2015. Recruiting centers followed the American Clinical Neurophysiology Society guidelines to order cVEEG.5 While predesigned case report forms and predetermined definitions were prospectively applied to enter the data, all the key data elements, such as seizure burden, management, and outcome, were determined by chart review. Like the study by Wietstock et al., the cohort was again skewed toward the majority being full-term neonates (88%) and 48% being transferred from an outside facility >24 hours after birth.
Compared with the study by Wietstock et al., Glass et al. found a higher frequency of neonates (82%) with seizures using cVEEG. The remaining 18% had clinical events that either resolved before cVEEG was begun or were found to be nonepileptic. Of the cohort, 62% had at least one EEG seizure without clinical correlate, and in 16% only subclinical EEG seizures were seen. The higher frequency of seizures in the study by Glass et al. could have resulted from more selective and higher-risk indications used to order cVEEG. The center in the study by Wietstock et al. was one of the seven centers in the study by Glass et al., which had a slightly different cohort over a longer period of time. Glass and colleagues also found the three most common etiologies similar to Wietstock and colleagues.
The seizure burden in the study by Glass et al. was high (>six seizures) in 59% of the neonates, and 16% had status epilepticus with no difference in the seizure burden between the term versus preterm neonates across the three most common etiologies. Term neonates with HIE treated with hypothermia showed no difference in seizure burden compared with those not treated by hypothermia. Interestingly, cVEEG was ordered much sooner after birth in term neonates (median, 50 hours; range, 15 hours to 5 days) compared with preterm neonates (median, 11 days; range, 27 hours to 33 days).
There was a significant difference in age at the time of first seizure between term neonates (median, 27; range, 11–80 hours) and preterm neonates (median, 14 days; range, 3–33 days). The age differences at the time of the first seizure between term and preterm babies may be due to earlier initiation of cVEEG in the term babies, but it may also reflect differences in the ictogenic mechanisms between the term and preterm brain. Phenobarbital was again reported to be the first-line antiepileptic drug used in 94% of the neonates; 64% of the neonates were refractory to the first antiepileptic drug and required two or more drugs.
Frequency of refractory seizures did not differ among the three most common acute etiologies of HIE, stroke, and hemorrhage, but it was significantly higher in neonates with inborn errors of metabolism and benign familial neonatal convulsions. Frequency of refractory seizures was quite similar to published studies in the literature.6 Mortality rate was 17% before discharge, similar to the mortality rate of 14% in the study by Wietstock et al. Interestingly, the seizure etiology strongly associated with death was brain malformations, not the three most common acute symptomatic causes. This remains an unexplained finding in clinical practice, as most children with brain malformations survive the neonatal period, and acute symptomatic etiologies are more common causes of neonatal death in clinical experience. Mortality was also strongly associated with higher seizure burden, and the highest mortality, 26%, was found in the neonates with status epilepticus. Higher seizure burden could be reflective of more severe degree of etiology and a more complicated clinical course with multiorgan/system failure, which was not addressed in the registry. In the adjusted analysis, the authors reported etiology, higher seizure burden and preterm birth to be significant risk factors for death.
Glass and colleagues also reported discharge exam as a measure of short-term outcome in the survivors; 49% had an abnormal examination of consciousness, tone, or deep tendon reflexes. Again, the rate of abnormal exam was the highest in neonates with brain malformations followed by those with epileptic encephalopathies and HIE. Term and preterm neonates had similar frequencies of abnormal exam. Discharge exam as a measure of short-term outcome in neonates is unlikely to be reliable as a variety of factors, including medications and involvement of other organ systems, can influence a single exam. A more reliable short-term marker could be serial exams as well as standardized developmental screening tools at 3 and 6 months after discharge adjusted for gestational age.
In summary, both studies reaffirm the role of cVEEG in providing accurate seizure assessment in neonates who are at risk. Because most neonatal seizures are due to an acute symptomatic cause, early diagnosis and treatment, as well as specific therapies to reverse and/or stabilize the etiology, make a good argument for the use of cVEEG. In addition, cVEEG could be a reliable tool to measure the effects of various antiepileptic treatments in successfully treating neonatal seizures, the ideal treatment of which is currently unclear. Prospectively designed clinical studies are needed to measure short-term and long-term outcomes of neonatal seizures. Ultimately, what remains to be seen is whether the added technical and personnel cost of cVEEG and consequent acute interventions will provide improved long-term cognitive and behavioral development and will mitigate neurologic disabilities and diseases in the survivors.