Objective: We retrospectively studied the efficacy of bevacizumab as salvage therapy for recurrent malignant glioma with a focus on the overall survival (OS). Methods: Patients who received a therapy other than surgery for recurrent malignant glioma were included. Efficacy was evaluated using MRI. Neurological function was evaluated using the Response Assessment in Neuro-Oncology (RANO). The survival rate was calculated using the Kaplan-Meier method. Results: Fifty-one patients with recurrent glioma (31 grade III, 20 grade IV) were included. Among them, 22 subjects (43.1%) received bevacizumab. The median OS was 10.2 months (range, 1 to 27 months). Patients receiving bevacizumab had comparable OS (a median of 9.9 vs. 10.0 months) and similar 6-month survival rate (43% vs. 34%) to those who did not receive bevacizumab. A subgroup analysis failed to notice any significant difference in grade III glioma patients receiving bevacizumab vs. those who did not. The median survival was significantly longer at 8.9 months (range, 4 to 13 months) in grade IV glioma patients receiving bevacizumab than in those who did not (5.6 months, range, 2 to 7 months, P=0.042). The 6-month survival rate was higher (83%) in those who received bevacizumab than in those who did not (47%, P=0.046). No grade 3/4 adverse events were observed in any patient. Conclusions: Bevacizumab, as a rescue therapy, provides a survival benefit for recurrent grade IV glioma. Objective: We retrospectively studied the efficacy of bevacizumab as salvage therapy for recurrent malignant glioma with a focus on the overall survival (OS). Methods: Patients who received a therapy other than surgery for recurrent malignant glioma were included. Efficacy was evaluated using MRI. The survival rate was calculated using the Kaplan-Meier method. Results: Fifty-one patients with recurrent glioma (31 grade III, 20 grade IV) were included. Among The median OS was 10.2 months (range, 1 to 27 months). Patients receiving bevacizumab had comparable OS (a median of 9.9 vs. 10.0 months) and similar 6-month survival rate ( 43% vs. 34%) to those who did not receive bevacizumab. A subgroup analysis failed to notice any significant difference in grade III glioma patients receiving bevacizumab vs. those who did not. The median survival was significances 9 months (range, 4 to 13 months) in grade IV glioma patients receiving bevacizumab than in those who did not (5.6 months, range, 2 to 7 months, P = 0.042). The 6-month survival rate was higher (83%) in those who received bevacizumab than in those who did not (47%, P = 0.046). No grade 3/4 adverse events were observed in any patient. Conclusions: Bevacizumab, as a rescue therapy, provides a survival benefit for recurrent grade IV glioma.