• 
• 

Three Part Question

In [a 16-month-old neurologically healthy girl with the first episode of a complex febrile seizure] what is the [probability of abnormalities] after [postictal EEG]?

Clinical Scenario

A previously healthy 16-month-old girl attends with the first episode of a complex febrile seizure (prolonged more than 15 minutes). She is neurologically normal, and examination reveals otitis media as the source of her fever. According to local protocol, her evaluation includes an EEG. One cannot but wonder as to the value of this routine practice.

Search Strategy

Medline. Secondary sources - nil
" (febrile, seizures, complex)" [MeSH] AND " EEG".

Search Outcome

Search results - 68 individual articles found, one relevant

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Maytal et al 2000 USA	33 patients with complex febrile seizures. Mean age:17.8 months. Neurologically normal children.	Retrospective chart review (4).	EEG	100% normal (95% CI 89.6-100).	Thirteen (39%) patients experienced prior febrile seizures.

Comment(s)

A febrile seizure is defined as a seizure accompanied by fever without central nervous system infection. Complex febrile seizures, also called atypical or complicated, were defined as focal, prolonged more than 15 minutes, or repetitive. The purpose of an EEG in the evaluation of complex febrile seizures is to help identify the nature of underlying acute or remote cerebral pathology and predict the risk of future afebrile seizures; no published study demonstrates that early EEG after a first episode of febrile seizure in postictal neurologically normal children will predict the occurrence of afebrile seizures. Studies that investigate the relations of signs or symptoms with EEG abnormalities or between clinical subgroups of complex febrile seizures, for example, focal, prolonged or recurrent, are not found. Only one small and not independently validated descriptive report with 33 patients appears that specifically answers question (2); this study addresses the value of an early postictal sleep EEG to detect the prevalence of abnormalities in neurologically normal children with the first complex febrile seizures, up to one week after the seizure, the study is retrospective and does not determine whether EEGs have been repeated over a follow-up period. 24 patients were qualified as complex cases based on one factor (prolonged in 9, repetitive in 13 and focal in 2). Nine other patients had 2 complex factors (in 6 the seizures were long and repetitive, in 2 focal and repetitive, and in 1 the seizures were long, focal and repetitive), which reduces the actual useful number of patients comparable to our patients in a clinical scenario. The study is uncontrolled and includes only neurologically normal children. Maytal and colleagues make no differences between complex febrile seizure clinical subgroups in this work. An important number of patients experienced prior febrile seizures, which means that not all patients were assembled at a common point in the course of the disease. The rate of abnormalities after an early postictal EEG in these patients is low and similar to the reported rate of abnormalities in children with simple febrile seizures, a fact that could be confirmed on a larger number of patients. EEG should be considered in all children with complex febrile seizures who recur with afebrile convulsions or in those children who recur with febrile seizures and exhibit developmental delays or abnormal neurological signs and symptoms.

Clinical Bottom Line

Abnormal EEG was uncommon in children with complex febrile seizures (95% probability less than 8.6%). Current local practice of EEG in neurologically normal children with complex febrile seizure does not appear to be evidence based. There is some limited evidence to suggest that it may not be useful.

References

1. Maytal J, Steele R. Eviatar L. Novak G. The value of early postictal EEG in children with complex febrile seizures. Epilepsia 2000;41(2):219-21.
2. Hanley J. A and Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. JAMA 1983;249(13):1743-45.