Objective: Surgical resection of mass lesions in Broca's area is controversial. To demonstrate that pathology may influence the localization of functional areas and language performance, we reviewed our experience of awake craniotomies in Broca's area.
Methods: Sixteen consecutive patients who underwent awake craniotomy and direct brain mapping for resective surgery in Broca's area were analyzed. Six patients had well-circumscribed lesions, whereas 10 patients had infiltrative gliomas. A short version of the Boston Diagnostic Aphasia Examination test was used for language assessment.
Results: Inferior frontal language sites were found in all but 4 patients. In patients with cavernomas or well-circumscribed tumors, 9 of 9 (100%) of the positive sites were located in the classic Broca's area (BA 44/45). By contrast, in those patients with gliomas, only 5 of 20 (25%) of the positive sites were located in BA 44/45. Patients with infiltrative gliomas demonstrated more deficits in the pre and postoperative periods than those with well-circumscribed mass lesions. All patients returned to their baseline abilities within 6 months.
Conclusion: Intraoperative language maps generated in cases with well-circumscribed lesions are different from those generated in cases with infiltrative gliomas. This supports the view that interindividual language variability and displacement of critical structures by mass effect should first be considered for circumscribed lesions, whereas reshaping should largely be attributed to brain plasticity in gliomas. Surgery in Broca's area can be safely conducted using awake craniotomy and brain mapping.