The Value of Electroencephalogram in Assessing Children With Speech and Language Impairments.

Purpose We sought to estimate the prevalence of isolated epileptiform activity (IEA) in children with speech and language impairments and discuss the utility of an electroencephalogram (EEG) in assessing these children. Method We conducted a systematic review and searched for eligible studies in 8 databases. All languages were included, and meta-analyses were performed. Results We found 55 prevalence estimates (8 with control group). The odds of having IEA were 6 times greater for children with speech and language impairments than for typically developing children. The overall pooled prevalence of IEA was 27.3%. A wide variation between the prevalence estimates was, to a certain degree, explained by type of impairment (8.1% in speech impairments, 25.8% in language impairments, and 51.5% in language regression). Sleep EEGs detected a significantly higher prevalence than awake EEGs. Although the presence of epilepsy gave a significantly higher prevalence than if epilepsy was not present, 33.5% of children with language impairment but without epilepsy were found to have IEA in sleep EEGs. Conclusions This systematic review shows that IEA is 6 times more prevalent in children with speech and language impairment than in typically developing children. However, the prevalence rates vary to a great extent. Uncovering IEA will, in addition to information from other clinical assessments, provide a more comprehensive understanding of the child's impairments. We argue that, although EEG is of questionable value when assessing children with speech impairments, sleep EEG could be valuable when assessing children with language impairments and, in particular, children who experience language regression.