gms | German Medical Science

Objective: Mapping language capacity with its subfunctions during direct cortical stimulation (DCS) in the stetting of awake craniotomy might be challenging without using more than one language paradigm. Applying several language tasks leads to increasing surgery times and to patient’s exhaustion. Additionally, language tasks must satisfy the demands of DCS by not exceeding the assessment time of 4 seconds in a single language trial to avoid DCS-induced seizures. We aimed to (1) design a single language paradigm which pictures the complete language system and (2) simultaneously to restrain the assessment time of single trials up to 4 seconds.

Methods: The intraoperative language task was designed as a single stimulus trial comprising a combination of the classical picture naming task using the DO 80 figures and a semantically associated verb. During DCS periods patients are supposed to generate grammatically and semantically correct sentences according to the language stimulus in a time interval not exceeding 4 seconds. To assess the intraoperative feasibility of the language task we also included 30 healthy subjects in a pilot study to measure subjects’ speed of performing a single language task trial and to evaluate the language paradigm according to its language sensitivity by using the paradigm as a task for functional magnetic imaging (fMRI). Intraoperatively, 21 patients without presurgical language deficits with brain tumors in language associated brain areas were included in the study. All patients received neuropsychological testing including language tests before and after surgery.

Results: The pilot study showed in healthy subjects that by applying the novel paradigm during fMRI there is activation in a strongly language associated left accentuated network of inferior frontal regions, the supramarginal and angular gyrus as well as the superior and middle temporal gyrus (p<.05, FEW) and, that it was feasible to conduct a single language trial of the novel paradigm in a time frame of 4s (performance speed 2.53 s; SD=.32s). Intraoperatively, all tumor patients showed DCS-associated language errors while conducting the novel language task. Postoperatively, mild or no neuropsychological impairments appeared compared to presurgical assessment (p<.05).

Conclusion: We present a novel language paradigm which intraoperatively safely pictures and monitors the whole language system and consequently can minimize postoperative language deficits.