Abstracts

Rationale: Clinical observations may have limited value to estimate the seizure frequency in patients with epilepsy. The objective of this study was to evaluate the accuracy of parental report regarding seizure identification in children with epilepsy.Methods: This was a prospective study conducted at the Columbia Comprehensive Epilepsy Center. Inclusion criteria were age > 1 month-old and < 18 years-old and diagnosis of epilepsy. Family or caregivers were instructed to pay close attention to the child during video-EEG monitoring in order to identify all seizures. Every time a clinical seizure was identified the parent activated the alarm button. The accuracy of parental reporting was evaluated based on the video-EEG monitoring findings. Results: From a group of 139 consecutive patients evaluated, 78 (31 girls, 47 boys) presented at least one event during video-EEG monitoring. Ages ranged from 8 months to 18 years-old (mean=8.3±5.6). A total of 1274 events were recorded. Based on parental report, 472 seizures were correctly identified, 623 seizures were missed, and 149 non-epileptic spells were identified as seizures (sensitivity = 43.1%, positive predictive value = 76%). All seizures were correctly identified by only 18/78 (23%) of the families. In 31/78 (40%) patients, seizures were only partially identified, and 10/78 (12.5%) patients did not identify any seizures. In 19 /78 (24.5%) patients, all events identified as seizures were non epileptic spells. Chronological age, age of seizure onset, type of epileptic syndrome, seizure type (focal versus generalized) and presence of developmental delay were not predictors of accurate seizure identification.Conclusions: Parents and children correctly identify only a minority of clinical seizures based on results from video-EEG monitoring. This finding has important implications for epidemiological and outcomes research in pediatric epilepsy. More accurate methods for seizure detection are necessary to improve both clinical care and research.