Background: Retrospective studies showing improved survival in patients with metastatic breast cancer (MBC) who undergo surgical treatment of the primary tumor have been criticized for bias in favor of younger, healthier women with lower disease burden. We attempted to identify these biases in our population. Methods: Our institutional cancer registry was queried for patients with MBC from 1994-2010. Demographics, clinical, radiologic and pathologic staging, as well as treatments and outcomes were recorded. Surgical and non-surgical groups were compared for differences in overall survival (OS) and clinicopathologic variables, including comorbidities, using uni- and multivariate analysis. Results: Ninety-one patients with metastatic disease identified within 3 months of initial diagnosis were eligible. 53% (48 pts) had primary breast surgery and 47% (43 pts) did not undergo surgery. Patients in the surgery group were younger on univariate analysis (mean age 53 vs. 62, p<0.01). Neither BMI (mean 30 vs. 29 kg/m²) nor Charlson comorbidity score (mean 6 in both groups) were significantly different, p=NS. Bone metastases were more common in the surgery group (48 vs. 26%) and multiple metastases in the non-surgery group (35 vs. 17%), p<0.05. Patients in the non-surgery group had ≥ 1 visceral metastasis when compared to the surgery group (62 vs. 35%), p<0.05. Higher OS was demonstrated in the surgery group both with Kaplan Meier curves (p<0.05) and univariate analysis (mean 3 vs. 2 yrs, 95% CI 2.6, 3.7), p<0.05. Survival was higher in the surgery group (p<0.01), at 1 year, but this difference did not persist at 3 and 5 years. On multivariate analysis, only difference in age remained significant (p<0.01). Conclusions: Our study supports existing data that women with MBC who have surgical treatment of the primary tumor have an improved survivorship. However, it also suggests a bias towards increased use of surgery in patients who are younger with smaller burden of metastatic disease. We did not find a bias in favor of healthier patients. Further study to determine the mechanism and magnitude of benefit of primary tumor extirpation is still needed.