Speech and Language Errors during Awake Brain Surgery and Postoperative Language Outcome in Glioma Patients: A Systematic Review

Abstract

Awake craniotomy with direct electrical stimulation (DES) is the standard treatment for patients with gliomas in eloquent areas. Even though language is monitored carefully during surgery, many patients suffer from postoperative aphasia, with negative effects on their quality of life. Some perioperative factors are reported to influence postoperative language outcome. However, the influence of different intraoperative speech and language errors on language outcome is not clear. Therefore, we investigate this relation. A systematic search was performed in which 81 studies were included, reporting speech and language errors during awake craniotomy with DES and postoperative language outcomes in adult glioma patients up until 6 July 2020. The frequencies of intraoperative errors and language status were calculated. Binary logistic regressions were performed. Preoperative language deficits were a significant predictor for postoperative acute (OR = 3.42, p < 0.001) and short-term (OR = 1.95, p = 0.007) language deficits. Intraoperative anomia (OR = 2.09, p = 0.015) and intraoperative production errors (e.g., dysarthria or stuttering; OR = 2.06, p = 0.016) were significant predictors for postoperative acute language deficits. Postoperatively, the language deficits that occurred most often were production deficits and spontaneous speech deficits. To conclude, during surgery, intraoperative anomia and production errors should carry particular weight during decision-making concerning the optimal onco-functional balance for a given patient, and spontaneous speech should be monitored. Further prognostic research could facilitate intraoperative decision-making, leading to fewer or less severe postoperative language deficits and improvement of quality of life.

Keywords: awake craniotomy; brain mapping; direct electrical stimulation; glioma; intraoperative language monitoring; language outcome; speech and language errors.

Figure 1

PRISMA flowchart of total records identified through searching of databases.

Figure 2

Language status in the preoperative (T0) and postoperative phases (T1–T4).

Figure 3

Significant preoperative and intraoperative predictors for postoperative language deficits at T1 (n = 589) and T3 (n = 456). T0 = preoperatively and postoperatively, T1 = 1–10 days, T3 = ≥3–8 months, OR = odds ratio and CI = confidence interval. This figure describes the binary logistic regression (T1 and T3) with the predictors of preoperative language status, intraoperative speech and language error categories, where the dependent variable is postoperative language outcome and reference categories are no preoperative language deficits and intraoperative speech arrest. * = p < 0.05. ** = p < 0.001.