We retrospectively evaluated and compared the efficacy and the toxicity profile of stereotactic radiosurgery (SRS) and fractionated stereotactic radiotherapy (FSRT) for the treatment of patients with brain metastases (BM). Between 2000 and 2009, 260 patients with 1-3 BM were treated using either SRS (median dose 20 Gy; n = 138) or two different FSRT dose concepts: 7 × 5 Gy (n = 61) or 10 × 4 Gy (n = 61). The median survival for SRS, 7 × 5 Gy and 10 × 4 Gy was 8, 7 and 10 months (p = 0.575), respectively, and the overall survival (OS) was 9 months. Follow-up imaging data were available in 214 of the 260 patients. The 1-year local progression-free survival (LPFS) was 73, 75 and 71 %, respectively (p = 0.191). After a mean follow-up of 28 months (range: 2.1-77 months), the rate of complete remission, partial remission, stable disease and progressive disease were 29, 40, 21 and 10 %, respectively. On multivariate analysis, RPA class I was associated with better OS and regional progression-free survival (both p < 0.001). SRS was associated with a higher toxicity rate (grade I-III) compared to the 7 × 5 Gy and 10 × 4 Gy groups (14 vs. 6 vs. 2 %, respectively; p = 0.01). Although FSRT was used for large lesions and/or lesions near critical structures, the LPFS was comparable to SRS. Importantly, FSRT presented low toxicity and appears to be an effective and safe treatment for BM not amenable to SRS. The 10 × 4 Gy fractionation scheme warrants further investigation due to its efficacy and safe toxicity profile.