Purpose: Transcranial magnetic stimulation (TMS) has recently emerged as a noninvasive alternative to the intracarotid sodium amytal (Wada) procedure for establishing hemispheric dominance (HD) for language. The accuracy of HD determined by TMS was examined by comparing against the HD derived by magnetoencephalography (MEG), a prominent clinical technique with excellent concordance with the Wada procedure. Methods: Sixty-seven patients (54 patients ≤18 years) underwent language mapping with TMS and MEG as part of clinical epilepsy and tumor presurgical assessment. Language was mapped in MEG during an auditory word recognition paradigm, and a laterality index was calculated using the number of dipoles and their spatial extent in the two hemispheres. Transcranial magnetic stimulation language mapping was performed as patients performed a naming task, and TMS-induced speech disruptions were recorded during 5-Hz TMS applied to anterior and posterior language cortices. Transcranial magnetic stimulation laterality index was estimated using the number and type of speech disruption in the language regions of each hemisphere. Results: Transcranial magnetic stimulation and MEG estimates of HD were concordant in 42 (63%) patients, resulting in a sensitivity of 74% and a specificity of 72%. The overall accuracy of TMS was 73%, equivalent to an odds ratio of 7.35. Conclusions: In this first large-scale comparative study in a clinical population, we demonstrate that TMS is a safe and reliable noninvasive tool in determining HD for language. Improving the accuracy of TMS by optimizing TMS parameters and improving task choice will further facilitate the use of TMS to characterize language function, especially in pediatrics.

BACKGROUND AND IMPORTANCE Presurgical mapping of eloquent cortex in young patients undergoing neurosurgery is critical for timely intervention, surgical planning, and minimizing postoperative deficits. However, invasive direct cortical stimulation has limited success in young children and noninvasive modalities, such as magnetoencephalography and functional MRI, require sedation, often precluding localization of critical language cortices. Transcranial magnetic stimulation (TMS), a noninvasive brain stimulation technique, is well suited to evaluate language areas in young children because it does not require the patient to remain still during mapping. CLINICAL PRESENTATION A 4-yr and 11-mo-old female patient diagnosed with epilepsia partialis continua of the right arm and face and right-sided weakness was evaluated at our institution. MRI findings and clinical examination led to the diagnosis of Rasmussen encephalitis involving left frontal lobe and insula. Language cortices were successfully identified in both hemispheres using TMS. The TMS findings aided in discussing with the family the risks of postsurgical deficits of left functional hemispherectomy, the definitive treatment for Rasmussen encephalitis. Postoperatively, the patient had intact speech and was seizure free. CONCLUSION We illustrate the feasibility and utility of TMS as a noninvasive functional mapping tool in this young child. The preoperative demonstration of bilateral language organization indicated a greater likelihood of preserved language functions postsurgery. We demonstrate that TMS is a safe and noninvasive tool to map language cortices in young children with serious epilepsy syndromes.