Clinical Utility of Transcranial Magnetic Stimulation (TMS) in the Presurgical Evaluation of Motor, Speech, and Language Functions in Young Children With Refractory Epilepsy or Brain Tumor: Preliminary Evidence

Abstract

Accurate presurgical mapping of motor, speech, and language cortices, while crucial for neurosurgical planning and minimizing post-operative functional deficits, is challenging in young children with neurological disease. In such children, both invasive (cortical stimulation mapping) and non-invasive functional mapping imaging methods (MEG, fMRI) have limited success, often leading to delayed surgery or adverse post-surgical outcomes. We therefore examined the clinical utility of transcranial magnetic stimulation (TMS) in young children who require functional mapping. In a retrospective chart review of TMS studies performed on children with refractory epilepsy or a brain tumor, at our institution, we identified 47 mapping sessions in 36 children 3 years of age or younger, in whom upper and lower extremity motor mapping was attempted; and 13 children 5-6 years old in whom language mapping, using a naming paradigm, was attempted. The primary hand motor cortex was identified in at least one hemisphere in 33 of 36 patients, and in both hemispheres in 27 children. In 17 children, primary leg motor cortex was also successfully identified. The language cortices in temporal regions were successfully mapped in 11 of 13 patients, and in six of them language cortices in frontal regions were also mapped, with most children (n = 5) showing right hemisphere dominance for expressive language. Ten children had a seizure that was consistent with their clinical semiology during or immediately following TMS, none of which required intervention or impeded completion of mapping. Using TMS, both normal motor, speech, and language developmental patterns and apparent disease induced reorganization were demonstrated in this young cohort. The successful localization of motor, speech, and language cortices in young children improved the understanding of the risk-benefit ratio prior to surgery and facilitated surgical planning aimed at preserving motor, speech, and language functions. Post-operatively, motor function was preserved or improved in nine out of 11 children who underwent surgery, as was language function in all seven children who had surgery for lesions near eloquent cortices. We provide feasibility data that TMS is a safe, reliable, and effective tool to map eloquent cortices in young children.

Keywords: brain tumor; children; epilepsy; language mapping; motor mapping; presurgical; speech mapping; transcranial magnetic stimulation.

Figure 1

Setup for TMS motor (A) and language (B) mapping studies in young children. (A): During TMS motor mapping in infants and toddlers, the child is seated on the parent's lap. (B): During TMS language mapping color naming task was used in some preschool children.

Figure 2

Examples of TMS-evoked MEPs and CSP in children under 3 years of age. (A): A 2.4-year-old male with right frontal lobe cortical dysplasia status post resection. MEP in the left APB muscle having an amplitude of 121 μV and a latency of 16 ms was evoked when TMS was applied to the right precentral gyrus. (B) An 18-month-old female with history of left hemisphere perinatal stroke involving left temporal lobe and subsequent infantile spasms demonstrating MEP evoked in the left brachioradialis muscle with an amplitude of 567 μV and a latency of 14 ms following TMS applied to the right precentral gyrus. (C) 1.7-year-old female with TSC- 2 demonstrating the MEP elicited in the right tibialis anterior muscle having an amplitude of 1.01 mV and a latency of 30 ms when TMS was applied to the left medial frontal lobe. (D) An 18-month-old female with history of left hemisphere perinatal stroke involving left temporal lobe and subsequent infantile spasms demonstrating CSP evoked in the right APB muscle following TMS applied to the left precentral gyrus. The left hemisphere is on the left side of the image.

Figure 3

TMS motor mapping demonstrating normal motor development in children under 3 years of age. (A): A 3-month-old male with dysplasia along the inferior frontal sulcus involving the inferior aspect of the right precentral gyrus, inferior gyrus, frontal gyrus, and middle frontal gyrus and history of infantile spasms. The motor cortices were localized along the precentral gyrus with MEPs elicited in contralateral and ipsilateral hand muscles, representing a normal developmental variant. (B): An 18-month-old female with history of left hemisphere perinatal stroke involving left temporal lobe and subsequent infantile spasms demonstrating a normal motor map. (C): A 2.4-year-old female with history of left frontal lobe focal cortical dysplasia, type IIb. Motor representation was localized posterior to the dysplasia.

Figure 4

TMS motor mapping demonstrating cortical reorganization in children under 3 years of age. (A): TMS motor mapping in a 1.7-year-old male with history of refractory seizures involving left-sided tonic flexion with a cortical dysplasia in the medial frontal side of the right frontal lobe on the pre- and post-central gyrus. TMS localized left hand and forearm representation to the precentral gyrus. Additionally, MEPs in the hand and forearm muscles were elicited while stimulating the area of cortical dysplasia. No MEPs were elicited in the left lower extremity even at 100% of stimulator output. The child underwent surgical resection of the lesion. At 9 months follow up, he was seizure-free with intact left-hand function and mild left leg monoparesis. (B): A 2-year-old male with history of left hemisphere perinatal stroke and right hemiparesis presenting with refractory epilepsy. TMS motor mapping demonstrated no motor representation for right upper extremity in the left hemisphere. Instead, both left and right upper extremities were represented around the precentral gyrus in the right hemisphere. The left hemisphere is on the left side of the image.

Figure 5

Validation of TMS motor mapping by CSM. Presurgical TMS-derived motor mapping in a 2.3-year-old female with tuberous sclerosis complex type 2. (A): TMS localized motor cortex in the right hemisphere in the precentral gyrus in the vicinity of a tuber. (B): The child underwent subdural grid placement, and the epileptogenic focus was localized to be anterior to the motor cortex. (C): The child underwent right anterior frontal lobectomy including the epileptogenic focus. Post-operatively, the child moves all extremities equally with normal bulk and strength and uses either hand to reach for an object.

Figure 6

Examples of language mapping with TMS. Speech errors in the form of speech arrest, semantic errors, and performance errors were elicited in both hemispheres. (A): Language mapping with TMS in a 5.6-year-old female with refractory cryptogenic focal epilepsy and asymptomatic cervical and thoracic syringohydromyelia. Her brain MRI was normal. TMS language mapping was completed using a color naming task and showed bilateral dominance for expressive language. (B): Language mapping with TMS in a 5.6-year-old male with right parietal cortical dysplasia that was in the inferior parietal lobule, predominantly superior to the marginal gyrus. TMS language mapping was completed using an object naming task and indicated a right hemisphere dominance for expressive language.

Figure 7

Clinical utility of presurgical TMS-derived language mapping in preschool children with brain tumors. (A): Left hemisphere temporal lobe language mapping in a 5.3-year-old female with recurrent pilomyxoid astrocytoma. She underwent a left temporal microsurgical subtotal (70%) tumor resection. Post-operatively, she had no speech deficits. (B): TMS language map from a 5-year-old male with recurrent left sylvian anaplastic ependymoma. Critical language areas were found around the margin of the tumor. The tumor was resected in full without any postoperative language deficits. (C) Right hemisphere temporal lobe language mapping in a 5.9-year-old female with a lesion in the right temporal lobe. Critical language areas were identified in bilateral temporal lobes. She underwent a resection of the right anterior temporal lobe, right amygdala, and hippocampus. The pathology classified the specimen as grade I ganglioglioma and focal cortical dysplasia type IIIb. Post-operatively, she had no speech deficits.