Abstracts

Rationale: Transcranial Magnetic Stimulation (TMS) is an emerging non-invasive technique used in mapping language cortex. In younger children who may have difficulty tolerating cortical stimulation mapping or other non-invasive mapping modalities, TMS can be particularly effective. In this study, we examined the success of TMS in identifying critical language areas prior to surgery in patients with epilepsy or brain tumor.

Methods: In 83 patients (5 – 64 y, 62 ≤ 18 y, 47 males) who underwent TMS language mapping prior to surgery involving either temporal or frontal lobe, we assessed the relationship between excision of TMS-identified language locations and post-operative language function. The patients’ brains were registered to standard MNI space for comparison and identification of language cortices.

Results: TMS identified language areas in at least one hemisphere in all patients. In 12 patients, mapping was performed only in the lesioned hemisphere; TMS successfully determined hemispheric dominance in 61, and dominance was nonconclusive in the remaining 10. Of the 83 patients, 70 had no post-operative language deficits, although 27 of these patients had TMS sites removed. In 26 of those, TMS localized a large number of language areas away from the lesion either in the same hemisphere (n = 16) or in the contra-lesional hemisphere (n = 10) (See Figure 1B for example). Another patient had intact language post-operatively despite the removal of several functional regions and minimal remaining TMS-delineated language areas. Of the 13 patients who had post-operative deficits, 4 patients exhibited transient language deficits following TMS area resection; all points resected were in Brodmann areas 20, 21, and/or 22 (Table 1, see Figure 1A for example). In the other 9 patients who exhibited post-operative deficits, TMS did not survey the resected areas (n = 7), or surgery included underlying white matter tracts (n = 2). The 3 patients with persistent deficits at 3- to 6-month follow up had their left temporal pole resected, an area not surveyed by TMS. Based on these findings, the sensitivity of TMS for localizing critical language areas was found to be 98% and the accuracy was 88%.

Conclusions: TMS successfully determined the dominant hemisphere for language in a majority of patients and localized critical language regions in a predominantly pediatric cohort with epilepsy or brain tumors. Our findings identify Brodmann areas 20, 21, and 22 as being particularly crucial for language function, as well as suggest further temporal pole mapping with TMS may be useful in preventing lasting deficits. These data are the largest to show the utility of TMS in optimizing post-operative language outcome and indicate that TMS is a safe, highly effective language mapping technique.

Funding: Please list any funding that was received in support of this abstract.: n/a.