Surgical decision-making for breast cancer (BC) brain metastases (BM) remains challenging due to surgery-related mortality risk. This study evaluated survival benefits of craniotomy and investigated causes of early postoperative mortality to refine surgical candidate selection. This retrospective cohort study included 157 BC patients with BM (July 2006-August 2022). Demographics, cancer subtype, performance score, BM number, extracranial metastasis, and treatment details were collected. Survival was analyzed using Kaplan-Meier and Cox regression models. Propensity score matching (PSM) was applied to control for confounding. A subgroup analysis of "short survivors" (survival <6 months) investigated early mortality causes. The mean age was 52.4 years; 36.9% had triple-negative breast cancer and 21.7% were HER2-positive. Craniotomy was performed in 37.4% of patients and was associated with improved overall survival (30-month survival rate: 30%, p < 0.001). Multivariate analysis identified oligometastases (p < 0.001), multiple metastases (p = 0.007), and absence of radiotherapy exposure (p = 0.005) as independent predictors of poor prognosis. Among 16 short-term survivors (<6 months), only three (18.8%) died within 30 days due to surgery-related complications; the remainder died from systemic disease progression. While craniotomy was associated with improved overall survival in the unadjusted analysis, this association was attenuated after PSM, suggesting patient selection may substantially account for the observed difference. Single brain metastasis and radiotherapy exposure were significant independent predictors of survival. Short-term survivor analysis revealed two distinct mortality patterns: the majority died from disease progression, while three patients experienced surgery-related death within 30 days. Rigorous patient selection remains essential in surgical decision-making.