OBJECTIVE: Whole brain radiotherapy (WBRT) provides benefit for patients with brain metastases but may result in neurological toxicity for patients with extended survival times. Stereotactic radiosurgery in combination with WBRT has become an important approach, but the value of WBRT has been questioned. As an alternative to WBRT, we managed patients with stereotactic radiosurgery alone, evaluated patients' outcomes, and assessed prognostic factors for survival and tumor control. METHODS: One hundred seventy-two patients with brain metastases were managed with radiosurgery alone. One hundred twenty-one patients were evaluable with follow-up imaging after radiosurgery. The median patient age was 60.5 years (age range, 16-86 yr). The mean marginal tumor dose and volume were 18.5 Gy (range, 11-22 Gy) and 4.4 ml (range, 0.1-24.9 ml). Eighty percent of patients had solitary tumors. RESULTS: The overall median survival time was 8 months. The median survival time in patients with no evidence of primary tumor disease or stable disease was 13 and 11 months. The local tumor control rate was 87%. At 2 years, the rate of local control, remote brain control, and total intracranial control were 75, 41, and 27%, respectively. In multivariate analysis, advanced primary tumor status (P = 0.0003), older age (P = 0.008), lower Karnofsky Performance Scale score (P = 0.01), and malignant melanoma (P = 0.005) were significant for poorer survival. The median survival time was 28 months for patients younger than 60 years of age, with Karnofsky Performance Scale score of at least 90, and whose primary tumor status showed either no evidence of disease or stable disease. Tumor volume (P = 0.02) alone was significant for local tumor control, whereas no factor affected remote or intracranial tumor control. Eleven patients developed complications, six of which were persistent. Nineteen (16.5%) of 116 patients in whom the cause of death was obtained died as a result of causes related to brain metastasis. CONCLUSION: Brain metastases were controlled well with radiosurgery alone as initial therapy. We advocate that WBRT should not be part of the initial treatment protocol for selected patients with one or two tumors with good control of their primary cancer, better Karnofsky Performance Scale score, and younger age, all of which are predictors of longer survival.