Yield of Outpatient Sleep EEG for Epileptiform Alterations' Detection in Children
Santiago Ontiveros1, Jan Novy, Marine Jequier-Gygax
1*Service de Neurologie, Département des neurosciences cliniques, CHUV and University of Lausanne, Lausanne, Switzerland; †Department of Clinical Neurophysiology, Marqués de Valdecilla University Hospital, Santander, Spain; and ‡Unité de Neurologie et Neuroréhabilitation Pédiatrique, Service de Pédiatrie, Département médico-chirurgical de pédiatrie CHUV and University of Lausanne, Lausanne, Switzerland.
Purpose:
Ambulatory EEGs in children are frequently ordered as sleep studies. However, the yield according to different clinical situations has received little attention to date. The authors aimed to quantify the added value in terms of detection of epileptiform features of an EEG containing sleep, as compared with only wakefulness, according to the referral diagnoses.
Methods:
The authors retrospectively selected consecutive outpatients' EEG recordings of patients between 6 months to 16 years old, performed between January 2014 and February 2015. The authors excluded those lacking at least 10 minutes of waking and/or at least 5 minutes of behavioral sleep. Interictal epileptiform activity (IEA) in wakefulness and sleep was compared among referral suspected diagnoses. Additional yield of sleep was considered if at least one of the following was observed: appearance of interictal epileptiform activity or increase by >50%; interictal epileptiform activity change in localization or morphology, seizure occurrence.
Results:
A total of 425 recordings (mean age 6.9 ± 4.7 years) were analyzed. Of them, 194 (45.6%) presented an additional yield during sleep, which was dependent on the occurrence of IEA during wakefulness: 77/251 (30.7%) in those without versus 117/174 (67.2%) in those with wakefulness IEA (P < 0.001, χ). The yield was markedly lower in studies performed for nonepileptic referral diagnoses (7% vs. 43% to 100%; P < 0.001, Fisher).
Conclusions:
When wakefulness EEG lacks epileptiform features, the yield of sleep EEG in our pediatric population appeared modest, especially in patients without a suspected epileptic syndrome. This information may be used to optimize the request of sleep EEG in children.
