Abstracts

Rationale: The aim of this study is to describe the clinical and EEG characteristics of preterm neonates undergoing continuous electroencephalography (cEEG) in the neonatal intensive care unit (NICU).Methods: Retrospective observational study of preterm neonates born at less than 37 weeks gestational age who underwent a clinically indicated cEEG (defined as greater than 3 hrs of uninterrupted recording) in the NICU at Boston Children’s Hospital from January 2011 through December 2013. The patients were further stratified by degree of prematurity. Moderate to late preterm was defined as 32 weeks to less than 37 weeks, very preterm defined as 28 to less than 32 weeks and extremely preterm defined as less than 28 weeks.Results: Fifty-two preterm infants were studied (46% male) with a median [29(p25)-36.14(p75)] gestational age of 35 weeks (Table 1). There was an acute intracranial symptomatic structural lesion in 30 patients (58%), a symptomatic non-structural abnormality in 3 patients (6%), systemic etiology in 18 patients (35%) and unknown etiology in 1 patient (2%). Main indications for monitoring were to characterize clinical events (N=39) followed by concern for non-convulsive seizures (N=19). Of the 39 patients in whom cEEG was used to characterize events, 9/39 (23%) had seizures. Electrographic seizures (ES) were detected in 12/52 (23%) of patients, and 4/12 (33%) were electrographic only. ES were then stratified by degree of prematurity. Of the moderate to late preterm children 7/34 (20%), 4/9 (44%) in the very preterm neonates, and 1/9 (11%) in the extremely preterm neonates had ES. One patient met the criteria for electrographic status epilepticus (ESE), defined as seizure lasting greater than 30 minutes or totaling greater than 50% of a one hour epoch. Thirty three percent were EEG only and 58% both clinical and subclinical features. The typical seizure duration was 1 to 5 minutes (66.7%). The median time to first seizure was 0.5 hours [0.24(p25)-4(p75)] with seizures captured in all 12 patients by 7.3 hours (Figure 1). The median total duration of monitoring was 19 hrs [16.6(p25)-35.4(p75)]. In-hospital mortality was 25% (N=13) for the entire group. Of the patients that died, 38 %( N=5) had electrographic seizures.Conclusions: Patients born preterm are at high risk for seizures. The majority of these seizures would go undetected without the use of cEEG given the high percentage of electrographic only seizures. In-hospital mortality is high in premature infants selected for monitoring. Next steps include further characterization of effect of seizure burden and therapeutic intervention on outcome in this cohort.