OBJECTIVES

: There is recurrence at some point following initial resection or radiosurgery in about 50% of patients previously treated for brain metastases (BMs); yet, little is known about the optimal management of recurrent BMs. To better understand the management of recurrent brain metastases, we carried out a systematic review and meta-analysis to describe outcomes of craniotomy for recurrent BMs after resection or stereotactic radiosurgery, with particular attention to recurrence in the previous resection cavity.

METHODS

: A literature search in the Pubmed, Embase and Cochrane databases was conducted on November 1st, 2018. The resulting literature and relevant references were screened, and data were extracted on study, patient, disease and treatment characteristics. Studies were grouped by initial local treatment strategy to compare resection after resection (RAR) vs. resection after stereotactic radiosurgery (RAS). Pooled analysis using random effects model was performed in R version 3.4.1 using the meta package.

RESULTS

: Of 4429 references screened, 13 retrospective analyses (605 patients) were included in this study (7 studies reported on RAR, 5 on RAS, while one study reported a stratified analysis on both treatments). Stratifying by initial treatment (RAR vs RAS), the pooled incidence comparing RAR to RAS were; gross total resection (GTR) (69% vs 74% p- interaction <0.01, complication (11% vs 16%, p-interaction <0.04), local failure (37% vs 28%, p- interaction =0.11), intracranial failure (54% vs 56%, p- interaction=0.03), six-months survival (83% vs 67%, p- interaction <0.01), and one-year survival (50% vs 36%, p-interaction <0.01), respectively.

CONCLUSIONS

: Resection of recurrent BMs following initial resection or radiosurgery was associated with relatively favorable outcomes that were comparable irrespective of initial treatment, offering compelling evidence for consideration of a more aggressive treatment of recurrent BMs.