P32. Improved reliabilty of intraoperative language mapping through preoperative nTMS and baseline linguistic scores

Abstract

Objective Repetitive navigated transcranial magnetic stimulation (rnTMS) and direct cortical stimulation (DCS) during awake surgery are complementary tools to locate speech functional areas of the brain. However, patients show inter-individual differences in their preoperative linguistic capabilities, which can impact the validity of intraoperative DSC mapping. We developed a standardized approach which utilizes preoperative aphasia scores and rnTMS to screen patients for their capability to perform reliable object naming during awake surgery. Methods We prospectively included 18 patients with language-eloquent brain tumors, scheduled for awake surgery. Aachen Aphasia Score (AAS), Demtect Score (DS) and Berlin Aphasia Score (BAS) were used to evaluate linguistic eloquence of each patient preoperatively. Moreover, all patients underwent 3 runs of naming all objects without rnTMS (baseline run), from which only promptly and correctly named objects were retained for later rnTMS mapping. These selected objects were also used for the intraoperative baseline testing and DCS mapping under level 2 of Ramsay Sedation Score. Results The degree of preoperative language impairment significantly correlated with the incidence of errors during preoperative as well as intraoperative baseline errors (AAT p = 0.01, BAS p = 0.001, DS = 0.001). Patients with significant aphasia (BAS >=2, AAT >=1) and cognitive impairment (DS >=2) committed over 50% of incorrect or delayed namings in the preoperative baseline (p = 0.001). Despite utilizing only the images retained after preoperative baseline naming, these patients still made significantly more mistakes during intraoperative baseline naming than patients with less severe aphasia. Moreover, the dosage of anaesthetic medication significantly correlated with the incidence of errors during intraoperative baseline naming (propofol p = 0.028, remifentanil p = 0.036), despite stopping all systemic analgetic and sedative medication at least 8 min before testing. Conclusion Patients with significant aphasia (BAS >=2, AAT >=2) and cognitive impairment (DS >=2) are not able to perform reliable intraoperative object naming and are therefore not suited for this procedure. In addition, dosages of analgetic and sedative medication during craniotomy should be kept as low as possible to allow for reliable intraoperative cognitive testing.