Stereotactic radiosurgery (SRS) offers the precise, local delivery of radiation for the treatment of recurrent gliomas. We examined the comparative characteristics, treatments, and outcome in a population having with low– and high–grade gliomas. Between September 1991 and December 1995, 20 patients (13 males, 7 females) had SRS for low-grade [9 patients: World Health Organization (WHO) grade II] vs. high-grade (11 patients: 9 WHO grade IV and 2 WHO grade III) gliomas. The patients with low-grade gliomas were younger (mean age ± SE, 39.6 ± 5.4 years; range, 11.4–61.0 years) than those with high-grade gliomas (51.3 ± 13.9 years; range, 32.9–78.5 years) (P = 0.09). Tumor locations were similar in the two groups: lobar for 7 of 9 low-grade vs. 9 of 11 high-grade gliomas (P = NS) and diencephalic or cerebellar for the remainder. The initial surgical treatments were biopsy, subtotal resection, and total resection for three, three, and three patients with low-grade gliomas, vs. three, seven, and one patients with high-grade gliomas, respectively (P = NS). Except for three patients with low-grade gliomas, all patients had conventional postoperative fractionated external-beam radiotherapy. The doses were 5583 ± 342 vs. 5345 ± 261 cGy (P = NS) for low- vs. high-grade gliomas, respectively. Intervals from surgery and conventional radiation (if given) to progression and SRS tended to be longer for low-grade gliomas: 37.5 ± 9.5 vs. 30.6 ± 11.1 months (P = NS) for low- vs. high-grade gliomas, respectively. High-grade gliomas were larger. The diameters of the collimators that allowed enclosure of the enhancing tumor volume within the specified treatment isodoses were 22.4 ± 2.0 mm for low-grade vs. 29.8 ± 2.8 mm for high-grade gliomas (P = 0.02, ANOVA). SRS doses and isodose percentiles were similar, however, for the two groups: 1650 ± 191 cGy and 79 ± 4.0% vs. 1932 ± 182 cGy and75 ± 3.5% for low- vs. high-grade gliomas, respectively (P = NS, dose and isodose). All patients with high-grade gliomas were followed until death. The mean survival after SRS was 11.6 ± 1.5 months (42 ± 12 months after surgery). Five of nine patients with low-grade gliomas expired 31.6 ± 6.0 months after SRS (P < 0.001, Kaplan–Meier log rank) (74.0 ± 16.0 months after surgery). The four survivors have been followed for 8, 13, 35, and 38 months after SRS, respectively. Multivariate analysis shows that the category of histologic grade correlates significantly with survival after radiosurgery (P = 0.01). SRS may be an important therapeutic option for patients with recurrent gliomas, regardless of their grade.