Two-minute Semantic Test Predicts Intraoperative Engagement in Non-aphasic Patients

Introduction: Successful engagement during awake language mapping is essential for achieving optimal surgical outcomes in patients with complex brain tumors. This study investigated the predictive value of pre-surgical neuropsychological variables in differentiating between optimal and poor engagers during intraoperative mapping (IOM) in awake craniotomy.

Methods: A total of 35 patients (80% right-handed), aged 18 to 81 years (M=52.26, SD=17.32) with complex brain tumors underwent presurgical neuropsychological evaluation, with 49% subsequently undergoing awake craniotomy with language mapping. A linear regression with mean replacement was performed with language-based neuropsychological variables as predictors and performance on baseline object naming as the dependent variable. We then selected a subset of non-aphasic patients with left hemisphere brain tumor who were deemed good candidates for awake craniotomy, but poorly engaged in IOM and matched them as closely as possible to optimal engagers based on age, sex, location of tumor, tumor grade, and handedness. Ninety percent of patients underwent preoperative magnetoencephalography [MEG] and confirmed left language dominance in all cases. A series of independent sample t-tests were performed to determine if neuropsychological variables differed between these groups.

Results: Results from the linear regression (F=2.92, p=.04) indicate that a two-minute test of semantic fluency (t=2.43, p=.02) is the sole significant predictor among neuropsychological test variables of baseline object naming performance accounting for 28% of the variance. Similarly, the only significant difference in the poor and optimal engagers during awake craniotomy on neuropsychological testing was semantic fluency (p <.001). Baseline object naming performance was statistically the same (p=.23-.71) between the poor and optimal engagers.

Conclusion : A two-minute semantic language screening test reliably predicts the feasibility of using object naming perioperatively and behavioral engagement in IOM among non-aphasic patients. This paradigm requires no test stimuli or materials rendering it an efficient, valuable tool for judging behavioral engagement intraoperatively.