No abstract

Introduction: Intraoperative seizures (IOSs) during awake craniotomy (AC) are associated with significant morbidity. The reported incidence of IOS is between 3% and 30%. The aim of this study was to identify risk factors for IOS during AC for elective resection or biopsy of a space-occupying brain lesion. Methods: In this retrospective study, we reviewed the records of all awake craniotomies performed by a single neurosurgeon at a single university hospital between July 2006 and December 2018. IOS was defined as a clinically apparent seizure that occurred in the operating room and was documented in the medical records. Explanatory variables were chosen based on previously published literature on risk factors for IOS. Results: Five hundred and sixty-two patients had a total of 607 AC procedures during the study period; 581 cases with complete anesthesia records were included in analysis. Twenty-nine (5.0%) IOS events were reported during 29 (5%) awake craniotomies. Most seizures (27/29; 93%) were focal in nature and did not limit planned intraoperative stimulation mapping. Variables associated with IOS at a univariate P-value <0.1 (frontal location of tumor, preoperative radiotherapy, preoperative use of antiepileptic drugs, intraoperative use of dexmedetomidine, and intraoperative stimulation mapping) were included in a multivariable logistic regression. Frontal location of tumor (adjusted odds ratio: 5.68, 95% confidence interval: 2.11-15.30) and intraoperative dexmedetomidine use (adjusted odds ratio: 2.724, 95% confidence interval: 1.24-6.00) were independently associated with IOS in the multivariable analysis. Conclusions: This study identified a low incidence (5%) of IOS during AC. The association between dexmedetomidine and IOS should be further studied in randomized trials as this is a modifiable risk factor.

An awake craniotomy is a common neurosurgical procedure for excising brain tumor(s) located near or in eloquent areas. The use of benzodiazepine (BZD) for sedation in some patients with neuropathological conditions (e.g., stroke, brain tumors) has been previously linked with re-appearance of neurological deficits including limb incoordination, ataxia, and motor weakness, resulting in complications for the patient along with procedural challenges. Whether or not these findings can be extrapolated to patients undergoing brain tumor resection is largely unknown. The current work primarily sought to compare neurological outcome(s) in the immediate postoperative period between BZD-free and BZD-based sedation techniques in patients undergoing awake craniotomy. Using a database composed of awake craniotomies conducted within a single center and by a single surgeon, patients were retrospectively classified based on midazolam administration into BZD-free sedation (n=125) and BZD-based sedation (n=416) groups. Patients from each group were matched based on age, sex, tumor location, tumor grade, preoperative neurological deficits, non-operative BZD use, and Karnofsky Performance Scale scores, resulting in 108 patients within each group. Postoperative neurological deficits were recorded. Logistic regression analyses were conducted comparing postoperative neurological deficits between the matched groups. Postoperative neurological deficits were more prevalent within the BZD-based sedation group compared to the BZD-free sedation group (adjusted odds ratio (aOR)=1.903, 95% CI=1.018-3.560, p=0.044). In addition, subgroup analysis of the matched cohort showed a relationship between preoperative neurological symptoms and postoperative neurological deficits in the BZD-based sedation group (aOR=3.756, 95% CI=1.390-10.147, p=0.009). Our findings support the notion that the increased incidence of postoperative neurological deficits with BZD sedation may in part be related to the unmasking of preoperative neurological deficits. Further studies are required to confirm this phenomenon.