Brain metastases treated with stereotactic radiosurgery may show delayed enlargement on post-treatment imaging that is of ambiguous etiology. Histopathologic interpretation of brain specimens is often challenging due to the presence of significant radiation effects admixed with irradiated residual tumor of indeterminate viability. The purpose of this study was to assess the impact of histologic findings on clinical outcomes following resection of these lesions. Between 2004 and 2010, 690 patients with brain metastases were enrolled in a prospective gamma knife data repository, and lesions requiring excision were identified. Tissue specimens were divided into four groups based on the ratio of treatment related inflammatory changes (TRIC) to tumor cells, and subsequently patient outcomes were assessed. Of 2,583 metastases treated, 36 were excised due to symptomatic enlargement. Only TRIC, without residual evidence of tumor, was seen in 36 % (13/36) of specimens. Resection of these lesions resulted in 100 % local control in follow-up. Of the remaining 23 lesions that contained any viable-appearing tumor within the resected specimen, 8 recurred after resection. Lesions that enlarged in the first 6 months were more likely to contain higher amounts of residual tumor cells. Patients with even <2 % tumors cells on excision had significantly worse local control (75 vs. 100 %, p = 0.024) and survival (HR 0.27, p = 0.029) compared with those patients with exclusively TRIC. In summary, our findings underscore the importance of surgically obtaining tissue in a method that facilitates complete lesional interpretive histology in order to accurately guide ongoing patient management.