Interobserver reliability of visual interpretation of electroencephalograms in children with newly diagnosed seizures
Hans Stroink1, Robbert-Jan Schimsheimer, Al W de Weerd
1Department of Neurology, St Elisabeth-TweeSteden Hospital, Tilburg, The Netherlands. H.Stroink@Elisabeth.nl
Insights
Visual interpretation of electroencephalograms (EEG) in children is highly reliable for detecting epileptiform abnormalities. This diagnostic tool shows substantial reliability for other EEG abnormalities in pediatric epilepsy evaluations.
Area of Science:
- Neurology
- Clinical Neurophysiology
Background:
- Electroencephalograms (EEG) are crucial for diagnosing epilepsy.
- Assessing the reliability of visual EEG interpretation is vital for its clinical utility in pediatric populations.
Purpose of the Study:
- To evaluate the interrater reliability of visual interpretation of EEGs in children with newly diagnosed, unprovoked seizures.
- To determine the reliability for specific EEG findings, including background patterns and epileptiform or non-epileptiform abnormalities.
Main Methods:
- Prospective collection of 50 standard EEGs and 61 EEGs after partial sleep deprivation from 93 children.
- Independent classification of EEG data by two clinical neurophysiologists.
- Calculation of interrater agreement using kappa statistics for various EEG features.
Main Results:
- Almost perfect agreement (kappa=0.83) for epileptiform discharges.
- Substantial agreement for overall interpretation (kappa=0.66) and abnormal background patterns (kappa=0.73).
- Moderate agreement for focal non-epileptiform abnormalities (kappa=0.54).
Conclusions:
- Visual EEG interpretation demonstrates high reliability for identifying epileptiform abnormalities in children.
- Reliability for other EEG abnormalities ranges from moderate to substantial, supporting its role in pediatric epilepsy diagnosis.
Abstract:
The reliability of visual interpretation of electroencephalograms (EEG) is of great importance in assessing the value of this diagnostic tool. We prospectively obtained 50 standard EEGs and 61 EEGs after partial sleep deprivation from 93 children (56 males, 37 females) with a mean age of 6 years 10 months (SE 5 mo; range 4 mo-15 y 7 mo) with one or more newly diagnosed, unprovoked seizures. Two clinical neurophysiologists independently classified the background pattern and the presence of epileptiform discharges or focal non-epileptiform abnormalities of each EEG. The agreement was substantial for the interpretation of the EEG as normal or abnormal (kappa 0.66), almost perfect for the presence of epileptiform discharges (kappa 0.83), substantial for the occurrence of an abnormal background pattern (kappa 0.73), and moderate for the presence of focal non-epileptiform discharges (kappa 0.54). In conclusion, the reliability of the visual interpretation of EEGs in children is almost perfect as regards the presence of epileptiform abnormalities, and moderate to substantial for the presence of other abnormalities.
