Effect of Treatment of Clinical Seizures vs Electrographic Seizures in Full-Term and Near-Term Neonates

Rod W Hunt ,Kristy Elsayed,Nicky Roberts,Yen Kok,Jan Stewart,Katherine J Lee,Judith Macey,Emma Yeomans,Lisa Mckeown,Karen Simmer,Ian Wright,Stacey Worthington,Rowena Dela Puerta,Connie Wong,Helen Liley ,Samantha Francis-Pester,Rachel Schembri,Monika Olischar,Flora Yuen-Wait Wong ,Paul B Colditz,Jeanie L.y Cheong ,Bevan M Headley ,A.d Shearman ,Robert W Black ,Nadia Badawi,Koert Dewaal,David A Osborn ,Deepika Wagh,Null Author_Id ,Victor Samuel Rajadurai,Peter A Dargaville,Nirmal Kavalloor Visruthan

doi:10.1001/jamanetworkopen.2021.39604

Abstract

Seizures in the neonatal period are associated with increased mortality and morbidity. Bedside amplitude-integrated electroencephalography (aEEG) has facilitated the detection of electrographic seizures; however, whether these seizures should be treated remains uncertain. To determine if the active management of electrographic and clinical seizures in encephalopathic term or near-term neonates improves survival free of severe disability at 2 years of age compared with only treating clinically detected seizures. This randomized clinical trial was conducted in tertiary newborn intensive care units recruited from 2012 to 2016 and followed up until 2 years of age. Participants included neonates with encephalopathy at 35 weeks' gestation or more and younger than 48 hours old. Data analysis was completed in April 2021. Randomization was to an electrographic seizure group (ESG) in which seizures detected on aEEG were treated in addition to clinical seizures or a clinical seizure group (CSG) in which only seizures detected clinically were treated. Primary outcome was death or severe disability at 2 years, defined as scores in any developmental domain more than 2 SD below the Australian mean assessed with Bayley Scales of Neonate and Toddler Development, 3rd ed (BSID-III), or the presence of cerebral palsy, blindness, or deafness. Secondary outcomes included magnetic resonance imaging brain injury score at 5 to 14 days, time to full suck feeds, and individual domain scores on BSID-III at 2 years. Of 212 randomized neonates, the mean (SD) gestational age was 39.2 (1.7) weeks and 122 (58%) were male; 152 (72%) had moderate to severe hypoxic-ischemic encephalopathy (HIE) and 147 (84%) had electrographic seizures. A total of 86 neonates were included in the ESG group and 86 were included in the CSG group. Ten of 86 (9%) neonates in the ESG and 4 of 86 (4%) in the CSG died before the 2-year assessment. The odds of the primary outcome were not significantly different in the ESG group compared with the CSG group (ESG, 38 of 86 [44%] vs CSG, 27 of 86 [31%]; odds ratio [OR], 1.83; 95% CI, 0.96 to 3.49; P = .14). There was also no significant difference in those with HIE (OR, 1.77; 95% CI, 0.84 to 3.73; P = .26). There was evidence that cognitive outcomes were worse in the ESG (mean [SD] scores, ESG: 97.4 [17.7] vs CSG: 103.8 [17.3]; mean difference, -6.5 [95% CI, -1.2 to -11.8]; P = .01). There was little evidence of a difference in secondary outcomes, including time to suck feeds, seizure burden, or brain injury score. Treating electrographic and clinical seizures with currently used anticonvulsants did not significantly reduce the rate of death or disability at 2 years in a heterogeneous group of neonates with seizures. http://anzctr.org.au Identifier: ACTRN12611000327987.

Highlights

Seizures are the most common manifestation of neonatal encephalopathy, and with an estimated incidence of 1 to 5 per 1000 live births in term newborns, they are more common in the neonatal period than during any other time of life.[1-3]
A total of 86 neonates were included in the electrographic seizure group (ESG) group and 86 were included in the clinical seizure group (CSG) group
Ten of 86 (9%) neonates in the ESG and 4 of 86 (4%) in the CSG died before the 2-year assessment

Summary

Introduction

Seizures are the most common manifestation of neonatal encephalopathy, and with an estimated incidence of 1 to 5 per 1000 live births in term newborns, they are more common in the neonatal period than during any other time of life.[1-3]. The neurodevelopmental associations of neonatal seizures have been extensively described.[4]. They include motor and cognitive deficits,[5] behavioral problems, such as attention deficit hyperactivity disorder and autism,[6] and postneonatal epilepsy.[7]. The increasing awareness of the dangers of neonatal seizures has engendered a liberal approach to the use of anticonvulsants in the neonatal intensive care unit (NICU). Most neonatal seizures are subclinical[10]; bedside neuromonitoring with amplitudeintegrated EEG (aEEG) has been widely adopted in NICUs11,12 to achieve recognition of most electrographic seizures[13] and improve diagnostic precision.[14]. While the standard investigation for seizures in the NICU is still conventional EEG (cEEG), aEEG monitoring systems are suited to continuous use and bedside interpretation. In studies comparing cEEG with aEEG, aEEG has been shown to have sensitivity and specificity between 70% to 80% for detecting electrographic seizures.[15]

Methods

Results

Discussion

Conclusion