Awake Craniotomy in the MRI Suite - a review of local practice

ABSTRACT

Introduction: Modern awake craniotomy (AC) has been performed since the 1980s, initially for epilepsy surgery but expanding to surgery for intracranial tumours (1). Intra-operative magnetic resonance imaging (ioMRI) was first utilised in 1994 in Boston (2), to overcome the issue of intra-operative brain shift during craniotomy, and permit the surgical team to check the extent of resection before closing. The techniques have been more recently combined, aiming to remove as much tumour from eloquent areas as possible. The interventional MRI (iMRI) suite at the National Hospital for Neurology and Neurosurgery (NHNN) consists of a 1.5 Tesla MRI scanner with an MR-conditional anaesthetic machine and operating table just outside the 5 Gauss line. This can be rotated to connect to the MRI table, and the patient is transferred into the bore of the scanner. There have been very few studies looking at iMRI and awake craniotomies, however there has been a suggestion that the addition of the MRI scan to awake craniotomy may reduce the requirement for redo surgery (3), and that awake craniotomies in iMRI may reduce the incidence of neurological impairment compared to surgery under general anaesthesia in iMRI. As the number of iMRI theatre suites increases across the UK, increasingly AC is being performed in this environment. In our study, we looked at these patients and their various pathologies, undergoing awake tumour resections in our iMRI suite, and their clinical management. Method(s) The theatre log book in MRI was reviewed for all awake cases, a longer time window was selected due to the impact of covid. Records reviewed to exclude procedures other than awake tumour resections with intraoperative MRI scanning. Identified total of 43 cases, a number grossly affected by covid interruptions. Post operative notes and discharge letters were reviewed to ascertain Clavien-Dindo scoring for postop complications. Result(s) 43 cases, with an average patient age of 36 years (spanning 19 y to 72 y), gender ratio MF=165. Mode ASA 2 (1-3), mean weight 78 kg (55-114 kg) and mean BMI 25.6 kg/m2 (20.2-35.6). * Most had a single ioMRI except three cases which had 2 scans, and 40% of cases had further resection after the ioMRI. * 44% noted complete resection on the post-operative MRI * Anaesthetic technique varied but asleep-awake-asleep/sedation comprised 88% of cases, with iGel used in 74% and classical LMA in 23%, and propofol/remifentanil used in 81%. * All patients had urinary catheters and arterial lines, no patients had central venous catheters. * Anaesthetic time (WHO sign-in to WHO time-out) ranged from 5 hours to 13 hours10 minutes with an average of 8 hours 54 minutes. * Postoperative destination was overnight recovery in 76%, HDU in 14%, and the remainder direct to the ward, where length of stay mean was 10.5 days (though mode was 4 d). * Clavien-Dindo score on discharge was 0 in 40%, 1 in 50%, 2 in 4.6% and 3b and 4 in 2.3%. * 44% were discharged with no new neurological deficit. Conclusion(s) We interpret the outcomes here as very positive, with a high proportion of patients leaving hospital with low Clavien-Dindo scores or with no new deficits identified post-operatively. It is clear that awake craniotomy is safely performed in the iMRI suite. As is often the case in anaesthesia, whilst we saw some absolute consistencies (such as 100% rate of urinary catheters and arterial lines), we saw here that the anaesthetic approaches were as varied as the anaesthetists themselves. Anaesthetists should be prepared for prolonged surgical time to ensure satisfactory surgical resection.