Speech delay in children: a functional MR imaging study.

Abstract

PURPOSE: To determine if children with speech delay who have been sedated have patterns of activation to passive language paradigms that are different than those of children with normal speech. MATERIALS AND METHODS: Seventeen children with speech delay (age range, 2-7 years; mean, 4.0 years) and 35 age-matched children with normal speech (age range, 2-8 years; mean, 4.2 years) were evaluated. The subjects in the control group were selected from patients referred for conventional magnetic resonance (MR) imaging. All children had absence of auditory impairment or mental retardation, and MR findings indicated that brain structure was normal. Sedation was achieved with pentobarbital (3-5 mg/kg) or chloral hydrate (75 mg/kg). Functional MR imaging was performed with a single-shot echo-planar blood oxygen-level-dependent technique and a passive block paradigm, in which the child listened to his or her mother's prerecorded voice. Statistical postprocessing of functional MR images was performed with the t test and cluster detection methods. Comparison between groups was performed depending on the type of data with a nonparametrical Mann-Whitney test, parametrical t test, or Fisher exact test. RESULTS: Five (83%) of the six children older than 3 years with speech delay had lateralized activation of functional MR imaging signal in the right hemisphere. Ten (71%) of 14 age-matched patients with normal speech had activation in the left hemisphere when exposed to the same passive listening tasks. When these groups were compared, this difference was statistically significant. (P =.036). No statistically significant lateralization was seen across all age groups in children with activation. CONCLUSION: Children older than 3 years with speech delay have activation in the right hemisphere more frequently than children older than 3 years with normal speech, who often have the expected finding of activation in the left hemisphere.