Beyond the cortex, early deficits, late recovery: a systematic review of language outcomes and subcortical mapping after awake craniotomy for diffuse gliomas

Background To evaluate immediate and long-term postoperative language outcomes in patients undergoing awake craniotomy with direct electrical stimulation (DES) for diffuse gliomas in eloquent regions. Methods PRISMA-compliant systematic review (PROSPERO: CRD420261342690) of adults (≥ 18 years) with supratentorial diffuse glioma in the language-dominant hemisphere enrolled in both randomised and non-randomised studies of awake craniotomy with direct electrical stimulation and language mapping. Databases searched included PubMed, MEDLINE, Embase, and Cochrane CENTRAL (from 2010 to March 2026). Study selection, data extraction, and risk of bias assessment were performed independently by multiple reviewers using standardised tools (JBI Critical Appraisal Checklists). Due to methodological heterogeneity, narrative synthesis was performed. Results Ten studies (532 patients enrolled; 55.3% male) published 2011–2025 were included; tumour grade was reported inconsistently, with higher-grade tumours predominating in the larger cohorts and at least 114 patients having WHO grade 4 disease. Early postoperative deficits ranged from 2.2 to 100% depending on assessment timing and measure (immediate post-op: up to 100%; day 7 to week 1: 2.2–22%). Persistent deficits at ≥ 3 months ranged from 0 to 47%. In three studies reporting whole-cohort outcomes with standardised assessment, rates were 0–5.6%. Papagno et al., using 3-month cut-offs, found that 47% of patients with uncinate fasciculus resection had impairment in famous face naming versus 10% with the tract preserved. The general trend was early decline followed by improvement, with most reported deficits resolving by around 3 months. Heterogeneous deficit definitions and inconsistent baseline assessment preclude treating this as a prognostic pattern, and resections involving the uncinate fasciculus were a notable outlier, underscoring the need for systematic subcortical tract mapping. Conclusion Awake craniotomy with DES remains the gold standard for preserving language. Early deficits are common but predominantly transient, and reported deficit rates appear closely associated with the timing of assessment. Persistent deficits assessed at ≥ 3 months were low in most studies, although rates varied considerably with the outcome measure used and with subcortical tract involvement. Subcortical tract involvement warrants systematic mapping. We advocate for the global adoption of standardised perioperative linguistic protocols to improve data comparability. Systematic review registration PROSPERO CRD420261342690