This retrospective, single-center study aimed to evaluate the predictors of prolonged awakening in adult patients who underwent the MAC technique for awake craniotomy. This study was permitted by the Institutional Review Board of the Chang Gung Medical Foundation, Taiwan (approval number: 202201692B0). Due to the retrospective nature and untraceable personal information in the study, the requirement for written informed consent was waived.

The inclusion criteria consisted of adult patients who (1) had a pathology-proven glioma (World Health Organization (WHO) grade I to IV) located in or near the cerebral eloquent cortex and (2) underwent awake craniotomy using the MAC technique at Linkou Chang Gung Memorial Hospital. Patients with non-glioma pathologies, incomplete medical records, or incomplete preoperative neuropsychological evaluation were excluded from the study [12]. From July 1, 2020, to January 31, 2023, 117 adult patients were enrolled in the analysis (Fig. 1). Each participant received T1-weighted magnetic resonance imaging (MRI) after contrast enhancement, and T2 fluid-attenuated inversion recovery (FLAIR) imaging for volumetric analysis before and after surgical resection. The extent of resection was determined by T1 contrast-enhancing volume for WHO grade IV gliomas and by T2/FLAIR volume for low grade gliomas (WHO grades II and III) [13]. Furthermore, the extent of resection was classified as gross total resection (GTR) (95–100%), subtotal resection (85–95%), or partial resection (< 85%), according to the classification described by Sawaya et al. [14]. Tumor location was divided into frontal, parietal, temporal, corpus callosum, insular, and hippocampal regions. For tumors that invaded more than one lobe, only the major involved lobes were counted. The histopathological diagnoses were all determined by a senior neuropathologist, and the grading criteria of gliomas were based on the WHO 2021 classification system [15]. The IDH1 mutation status was determined by immunohistochemistry using an anti-R132H-IDH1 antibody. The first craniotomy was defined as the initial awake surgery for newly diagnosed brain tumors, and repeated craniotomy referred to a redo awake surgery for recurrent tumors in or near the eloquent cortex following a prior awake or anesthetic surgery.

The MAC technique with propofol sedation was administered to all patients as the main anesthesia regimen. Initially, moderate sedation was achieved using propofol infusion accompanied by an intermittent bolus of midazolam or fentanyl at the discretion of duty anesthesiologist. Following sedation, invasive hemodynamic monitoring, including arterial line and central venous catheterization, was performed, and a urinary catheter was inserted. A scalp block was administered using either 20 ml of 0.5% levobupivacaine (Chirocaine, Abbott) or 20 ml of 1% xylocaine (Lidocaine, LITA Pharmacy Co.). Intravenous parecoxib 40 mg (Dynastat, Pfizer) was given for multimodal analgesia. Besides, local anesthetics with epinephrine was infiltrated at the headpin sites and the planned incision line. Dexamethasone (5 mg; Methasone; Taiwan Biotech Co.) and ondansetron (8 mg; Supren; Taiwan Biotech Co.) were intravenously administered to prevent nausea and vomiting. Subsequently, the patients were positioned either supine or semilateral, depending on the location of the glioma, with the head fixed in a Mayfield holder. Standard neuro-navigation was applied for craniotomy guidance with Medtronic Stealth S7 system (Medtronic®).

Adequate airway patency was maintained using a suitable nasal airway with the tip above the glottis, supplying oxygen at a flow rate of 6 L/min via a simple mask. Adequate gas exchange was confirmed through arterial blood gas analysis, showing fair oxygenation and normocapnia. Propofol was the sole sedative used from skin incision to dural opening (turning off), administered in an effect-site target-controlled infusion pump (Schnider model). Dexmedetomidine and remifentanil were not utilized in our MAC technique because these drugs were not covered by health insurance or unavailable in our institute during the study period. The effect-site concentration (Ce) of propofol was titrated between 2.0 and 2.2 µg/mL to maintain moderate sedation, assessed using the modified Observer’s Assessment of Alertness/Sedation Scale (OAA/S) score 0–2 or Bispectral index (BIS) 60–80 if used [16, 17]. After discontinuing propofol at dural opening, the patients were prompted every 30 s to awaken by opening their eyes or speaking their names. The amount of propofol infused before cessation was recorded as the total dose (mg). The estimated propofol Ce at the time point of “turning off,” responsive to “opening eyes,” “speaking names,” and “beginning of functional mapping,” were recorded. Since speaking is crucial for functional mapping and indicates subcortical-cortical connectivity [18], we defined the time from propofol cessation to speaking one’s name as “time to awakening.” Because anesthesia depth monitoring is not universally covered by health insurance in Taiwan, the BIS value was not recorded for every participant and was not analyzed in this cohort. After the patients became alert and could perform functional testing, experienced neuropsychologists continuously assessed their awake language and motor functions throughout the whole tumor resection process, which typically lasted 1.5–2.5 h. Following maximal tumor resection, propofol infusion was resumed to maintain sedation until surgery completion. Subsequently, patients were transferred to a neurosurgical intensive care unit (ICU) for postoperative care.

The primary outcome was the contributing factors to delayed awakening after discontinuation of propofol. We defined delayed awakening as a time to speaking one’s name ≥ 20 min for two reasons: the median time to speaking name in our cohort was 19 min, and a previous Dutch cohort also used 20 min as the threshold for prolonged awakening [10]. The secondary outcome was the comparison of different predictors of delayed awakening in patients receiving the first or repeated craniotomy and in patients with IDH1 mutant or wild-type gliomas, as these factors are significant contributors to prolonged awakening in age- and BMI-matched comparisons (Table 1). Enrolled patients’ demographics including age, sex, Karnofsky performance scale (KPS) scores, neuropsychological evaluations, tumor characteristics, anesthetic and surgical profiles, histopathological examinations, and tumor resection rates were compared. Data are expressed as mean ± standard deviation for continuous variables and percentages for qualitative variables (sex and tumor characteristics). The statistical analyses utilized Student’s t-test or χ2 tests for comparison between two-groups and analysis of variance (ANOVA) for comparison between multiple groups. Because age, sex, weight, and height are required parameters in Schnider model of propofol target-controlled infusion pump [19], age and BMI 1:1 propensity score-matched comparisons or adjustments for age, sex, and BMI were performed during comparison to eliminate demographic differences. SAS software (version 9.4; SAS Institute, Inc. NC, USA) was utilized for all statistical analyses, and a two-sided p value < 0.05 was considered statistically significant.