Processed EEG for Monitoring of Anesthetic Depth in Intracranial Tumor Surgery

Background The incidence of primary tumors in the central nervous system in adults in Sweden is approximately 1400 per year. Gliomas constitutes the most common form, approximately 50%. The neurosurgery aims at securing material for pathological-anatomical diagnosis, reduction of symptoms, maximal reduction of tumor mass with minimal functional impact. An important feature of neurosurgical anesthesia, apart from intraoperative stability, is early postoperative recovery of consciousness with minimal residual sedation. This is a key factor to enable early neurological assessment and early discovery of postoperative complications (hematomas).

In total intravenous anesthesia the anesthesia drugs (e.g. propofol and remifentanil) are supplied by continuous intravenous infusion. The pumps used for administration can be programmed manually (in mg/kg/h or mcg/kg/min, manual total intravenous anesthesia, mTIVA) or using a programmed algorithm (target controlled infusion, TCI, where the algorithm calculates the concentration of drug at the effect site). Regardless of programing system, the anesthetist or anesthetic nurse adjusts the programming to the patients´ needs during anesthesia and surgery. The Neurosurgical anesthesia section of Lund University Hospital, Lund, Sweden, uses both mTIVA and TCI and all exposed anesthetic personnel are comfortable with both mTIVA and TCI.

Montoring of anesthetic depth is based on physiological reactions. In neurosurgical anaesthesia access to face and pupils are restricted/absent and thus monitoring further restricted. To add to the monitoring of anesthetic depth several processed simplified EEG systems (pEEG) have been developed. EEG systems have proprietary underlying algorithms and are thus not directly comparable. The recommendations regarding adequate anesthesia dept differs considerably numerically. pEEG -systems are expected to add to the evaluation of anesthetic depth, reduce the risk of awareness and reduce the risk of over- anesthetizing with hypotension and prolonged recovery of consciousness during emergence. pEEG has predominantly been assessed in other subspecialities than neuroanesthesia. In neuroanesthesia interference of the pEEG monitoring system with the surgical field is one factor that has to be taken into account when placing the monitoring system, especially in frontally placed tumors.

Bispectal index (BIS, Medtronic,) is one of several commercially processed EEG systems (pEEG). BIS is a pEEG index ranging from 0-100, where 0 corresponds to isoelectric EEG and 100 to full alertness. Recommended anesthesia depth is 40-60 intraoperatively. The EEG-monitoring in the BIS system is available as a unilateral strip increasing clinical usefulness in this setting.

The aim of this study is to improve anaesthesia and monitoring for intracranial tumor resection by evaluating the effect of pEEG monitoring for propofol/remifentanil anesthesia delivered by manual total intravenous anesthesia (mTIVA) or target controlled infusion (TCI).

Methods Ethical review board approval (Dnr 2024-01935-01, Stockholm, Sweden) is present.

The study is a single centre randomized trial between mTIVA and TCI with concealed pEEG-monitoring.

A couple of days, up to two weeks peroperatively, the patient is assessed at the preoperative neurosurgical out patient clinic. At that time an anesthesiologic preoperative assessment is performed and elegible patients receives information about the study and are able to give informed consent.

Patients will be randomized to mTIVA or TCI with stratification based on presence of hypertension. Randomization (mTIVA or TCI) is performed on the day of surgery. Monitoring, including BIS-monitoring is started prior to induction. All patients receive propofol/remifentanil based general anesthesia with norepinephrine infusion to support blood-pressure. Endotracheal intubation is facilitated with rocuronium. The result of the pEEG-monitoring is concealed, but anesthesia caregiver(s) can obviously not be blinded to mTIVA or TCI. Extubation is expected to take place in the operating room. Postoperative level of sedation is assessed using Karolinska Sleepiness Scale (KSS) during the first two postoperative hours. KSS is an ordinal scale assessing sleepiness during the last five minutes from "extremely alert" to "very sleepy, great effort to keep awake, fighting sleep". Follow up regarding awareness is performed once day 1-3 postoperatively.

Data regarding age, sex, length, weight, comorbidities, tumor location, ASA- classification and preoperative medication will be collected. Peroperative vital parameters and pEEG -results and data from infusion pumps are primarily collected electronically. The manually kept anesthetic notes are copied and stored as back-up. Times for start and end of anesthesia end surgery and time of recovery of consiousness are collected. All individual data are de-identified and coded after collection.

Inclusion and exclusion criteria Inclusion Adult (>18 years old) Elective supra or infratentorial tumor resection via craniotomy Cognitive function allowing informed consent.

Exclusion Intracranial tumor with operation via biopsy Need for intraoperative neurophysiological monitoring Tumor resection via awake surgery Tumor localization not allowing placement of BIS electrodes due to interference with surgery Morbid obesitas

Outcomes Primary endpoint is per cent of time spent within recommended BIS- levels during anesthesia (from preparation until end of surgery).

Secondary endpoints are: mean pEEG, time from end of surgery to extubation, peroperative propofol/remifentanil consumption, postoperative degree of sleepiness measured with Karolinska Sleepiness Scale (KSS) and awareness assessment.

Technical aspects of the pEEG- monitoring will be assessed with special reference to interaction with sterile field and performance in different types of surgical positions.

An assessment of the "smart pilot view" (Dräger, Draeger Medical, Mississauga, ON, Canada) as an indicator of time to recovery of consuiousness may be added.

Randomization Randomization will take place on the day of surgery by opening the pre-prepared sealed randomization envelope. Randomization between mTIVA and TCI (1:1) is stratified for hypertension.

Statistical methods and power calculation Since no similar studies were identified, a difference between the groups (mTIVA and TCI) of 5% with an SD of 10% was used. With a power of 80% and alpha 0,05 and 1:1 allocation 126 patients, 63 in each group should be randomized. These figures are based on the estimation that a difference of <5% would not be clinically significant.

The main analysis will be done according to intention to treat. Depending on distribution parametric or non-parametric analysis methods will be used.

For the secondary endpoints differences in mean pEEG, drug use and relevant times will be analysed. Depending on distribution parametric or non-parametric analysis methods will be used. As primary KSS variable the KSS on arrival in the postoperative unit will be used, a deviation of two or more units will be considererd a significant change. KSS will be used as an ordinal scale. A per protocol analysis may be performed as a secondary analysis. The "smart pilot view" will be analyzed comparing observed with calculated time of regaining consciousness using a relevant test depending on distribution. Awareness assessment will be descriptive.