Elsevier

Pediatric Neurology

Volume 32, Issue 3, March 2005, Pages 155-161
Pediatric Neurology

Original articles
Electroencephalography in neonatal seizures: Comparison of a reduced and a full 10/20 montage

https://doi.org/10.1016/j.pediatrneurol.2004.09.014Get rights and content

This study compares a reduced electrode montage (9 electrodes) with the full 10/20 electrode montage for the ability to detect and characterize neonatal seizures and background electroencephalographic (EEG) characteristics, utilizing new digital technology allowing “remontage” of previously acquired records. A total of 151 neonatal EEG records were retrospectively and blindly analyzed by two readers. Records were first analyzed for seizure number, topography, duration, and characteristics of EEG background using the reduced montage, before reanalysis with the full montage. One hundred eighty-seven seizures were identified in 31 ictal recordings using the full montage. Using the reduced montage, 166 seizures were identified in 30 records. The sensitivity and specificity of the reduced montage for detecting electrographic seizures was 96.8% and 100% respectively. In only one patient’s record, the single seizure was missed altogether. For grading background abnormalities, the sensitivity and specificity of reduced montage was 87% and 80%. Although there are inherent weaknesses in reduced montages with respect to both underestimation and overestimation of seizure number, a nine-electrode reduced montage can be a sensitive tool for identification of neonatal seizures and assessment of background characteristics of neonatal electroencephalography.

Introduction

Electroencephalography (EEG) is an important clinical tool in the identification of neonatal seizures, estimation of central nervous system maturity, and assessment of prognosis in neonatal encephalopathy [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13]. Accurate diagnosis of neonatal seizures requires skilled clinical staff, and is greatly assisted by the use of EEG, and clinical assessment alone is an insensitive method of quantifying neonatal seizures [14], [15]. In newborns treated with antiepileptic medications for seizures, electroclinical dissociation may occur, with continuing electrographic seizures without clinical manifestation [16], [17]. Continuous EEG monitoring of such infants may therefore be important not only for diagnosis, but also for assessment of the efficacy of intervention [8], [17]. Guidelines published by the American Electroencephalography Society (now the American Clinical Neurophysiology Society) recommend either a full (16 electrodes) montage, or a reduced (9 electrodes) montage, with four additional channels devoted to polygraphic variables (respiration, eye movement, electrocardiogram, electromyography) [18]. The potential advantages of using a reduced montage in these ill infants include: shorter electrode application time reducing the risk of heat loss, decreased handling of vulnerable infants, and increased scalp space for performance of cranial ultrasound and/or intravenous access. However, while it is known that extremely reduced (4 electrodes) montages may fail to identify topographically restricted seizures [19], comparison of a reduced (9 electrodes) montage with a full montage has not been reported. With the advent of digital technology, it has become possible to “remontage” previously recorded EEGs for such a comparison.

In this study, the sensitivity, specificity, and predictive value of a reduced montage (RM, 9 electrodes) for the detection of electrographic seizures and for assessment of background abnormalities are defined, in comparison to a full 10/20 electrode montage (FM, 19 electrodes).

Section snippets

Material

One hundred fifty-one neonatal EEG records, obtained between March 1999 and October 2001, were retrospectively obtained from the master database in the Bio-logic System of the Clinical Neurophysiology Laboratory, Children’s Hospital, Boston. The EEGs had been recorded using a full 10/20 electrode array with additional polygraphic channels. For the purpose of blinded analysis, each EEG file was given a study code number and patient names were deleted from the records before analysis.

EEG montages

A digital

Subject data

A total of 151 EEG recordings were obtained from 139 infants, between 29 and 48 weeks conceptional age. Ages at recording were as follows: ≪30 weeks: 2; 30 to 33 weeks: 3; 34 to 37 weeks: 14; 38 to 40 weeks: 51; 41 to 44 weeks: 72; and 45 to 48 weeks: 9. Among the 151 EEG records, 126 were routine neonatal EEG recordings, 35 were prolonged bedside EEG recordings. Seventy-two EEG files included a video-EEG recording.

Electrographic seizures

Using the FM, a total of 187 seizures were identified in 31 EEGs obtained from

Discussion

Neonatal electrographic seizures are typically focal in nature and topographically restricted [22], [28], [29], [30]. Seizures may be missed when using a limited number of electrodes. Some laboratories use full 10/20 electrode montages (FM) to detect neonatal electrographic seizures more reliably. Handling time and head mobilization maneuvers of babies increase in proportion to the number of electrodes applied. For these reasons, double-distance electrode placement is used in some laboratories.

References (34)

  • K.R. Holden et al.

    Neonatal seizures. I. Correlation of prenatal and perinatal events with outcomes

    Pediatrics

    (1982)
  • E.D. Mellits et al.

    Neonatal seizures. II. A multivariate analysis of factors associated with outcome

    Pediatrics

    (1982)
  • E.M. Mizrahi

    Electroencephalographic-video monitoring in neonates, infants, and children

    J Child Neurol

    (1994)
  • M. Andre et al.

    Neonatal seizuresRecent aspects

    Neuropediatrics

    (1988)
  • A. Legido et al.

    Neurologic outcome after electroencephalographically proven neonatal seizures

    Pediatrics

    (1991)
  • A.M. Bye et al.

    Spatial and temporal characteristics of neonatal seizures

    Epilepsia

    (1995)
  • M.C. McBride et al.

    Electrographic seizures in neonates correlate with poor neurodevelopmental outcome

    Neurology

    (2000)
  • Cited by (75)

    • Sensory assessment: Neurophysiology in neonates and neurodevelopmental outcome

      2020, Handbook of Clinical Neurology
      Citation Excerpt :

      Continuous video EEG (cEEG) is considered the gold standard for diagnosing and prognosing in neonatal seizures. Despite reduced neonatal montage (sometimes 11 electrodes instead of 20 in full montage) due to the small head of neonates, cEEG has a sensitivity and specificity for seizure detection of 96.8% and 100%, respectively (Tekgul et al., 2005). The influence of sedation and other medications on the EEG must be considered during trace analysis.

    • Diagnosis and Management of Acute Seizures in Neonates

      2018, Neurology: Neonatology Questions and Controversies
    • Neonatal Seizures

      2018, Volpe's Neurology of the Newborn
    View all citing articles on Scopus
    View full text