Impact of preoperative functional magnetic resonance imaging during awake craniotomy procedures for intraoperative guidance and complication avoidance

Conclusions: The routine use of fMRI was not useful in identifying language sites as performed and, more importantly, practiced tasks failed to prevent neurological deficits following awake craniotomy procedures.

Background: We wanted to study the role of functional MRI (fMRI) in preventing neurological injury in awake craniotomy patients as this has not been previously studied.

Objectives: To examine the role of fMRI as an intraoperative adjunct during awake craniotomy procedures.

Methods: Preoperative fMRI was carried out routinely in 214 patients undergoing awake craniotomy with direct cortical stimulation (DCS).

Results: In 40% of our cases (n = 85) fMRI was utilized for the intraoperative localization of the eloquent cortex. In the other 129 cases significant noise distortion, poor task performance and nonspecific BOLD activation precluded the surgeon from using the fMRI data. Compared with DCS, fMRI had a sensitivity and specificity, respectively, of 91 and 64% in Broca's area, 93 and 18% in Wernicke's area and 100 and 100% in motor areas. A new intraoperative neurological deficit during subcortical dissection was predictive of a worsened deficit following surgery (p < 0.001). The use of fMRI for intraoperative localization was, however, not significant in preventing worsened neurological deficits, both in the immediate postoperative period (p = 1.00) and at the 3-month follow-up (p = 0.42).