Craniotomy under local anesthesia and monitored conscious sedation for the resection of tumors involving eloquent cortex

Objective. Resection or even biopsy of an intra-axial mass lesion in close relationship to eloquent cortex carries a major risk of neurological deficit. We have assessed the safety and effectiveness of craniotomy under local anesthesia and monitored conscious sedation for the resection of tumors involving eloquent cortex. Methods. We have performed a retrospective review of a consecutive series of 157 adult patients who underwent craniotomy under local anesthesia by one surgeon (P.M.B.) at Brigham and Women's Hospital in Boston. All patients had tumors in close proximity to eloquent cortex, including speech, motor, primary sensory, or visual cortex. In most cases the lesion was considered inoperable by the referring surgeon. All resection was verified by post-operative imaging approximately one month after surgery and all cases were reviewed by an independent neurosurgeon (A.D.). Results. In 122 cases, brain mapping was performed to identify eloquent cortex and in the remainder neurological monitoring was maintained during the procedure. Radiological gross total resection was achieved in 57% of patients and greater than 80% resection was achieved in 23%. Thus 4 out of 5 of patients had major resection despite the close relationship of tumor to eloquent cortex. In 13%, resection could be done but less than 80% of tumor was removed because of danger of neurological deficit. In 7% of patients, only a biopsy could be done because of infiltration into eloquent cortex that could only be assessed at surgery. In 76 patients with pre-operative neurological deficits, there was complete resolution of these deficits in 33%, improvement in 32%, no change in 28%, and long-term worsening in 8%. Among 81 patients with no pre-operative neurological deficit, 1 patient suffered a major permanent neurological deficit, and 2 developed minor deficits. There was a transient post-operative deficit in one-third of cases, but this had resolved at one month in all but three patients. Monitored conscious sedation was performed without anesthetic complications using midazolam, sufentanyl and fentanyl with or without propofol. Only one case needed to be converted to general anesthesia. Patient satisfaction with the procedure has been good. Operating time and hospital stay were lower than the mean for brain tumor craniotomy at this hospital. Conclusions. Tumor surgery with conscious sedation is a safe technique that allows maximal resection of lesions in close anatomical relationship to eloquent cortex, with a low risk of new neurological deficit. Only 7% of intrinsic cortical tumors were ineligible for partial or complete resection with this technique.