Studies looking at the benefit of surgery at first relapse (second surgery) for recurrent glioblastoma were confounded by including patients with varying grades of glioma, performance status and extent of resection. This case-controlled study aims to remove these confounders to assess the survival impact of second surgery in recurrent glioblastoma. Retrospective data on patients with glioblastoma recurrence at two tertiary Australian hospitals from July 2009 to April 2015 was reviewed. Patients who had surgery at recurrence were matched with those who did not undergo surgery at recurrence, based on the extent of their initial resection and age. Overall survival (OS1 assessed from initial diagnosis and OS2 from the date of recurrence) as well as functional outcomes after resection were analysed. There were 120 patients (60 in each institution); median age at diagnosis was 56 years. Median OS1 was 14 months (95% CI 11.5-15.7) versus 22 months (95% CI 18-25) in patients who did not undergo second surgery and those with surgery at recurrence. OS2 was improved by second surgery (4.7 vs 9.6, HR 0.52, 95% CI 0.38-0.72, P < 0.001), and by chemotherapy, given at recurrence, (HR 0.47, 95% CI 0.24-0.92, P = 0.03). After second surgery, 80% did not require rehabilitation and 61% were independently mobile. Second surgery for recurrent glioblastoma was associated with a survival advantage. Chemotherapy independent of surgery, also improved survival. Functional outcomes were encouraging. More research is required in the era of improved surgical techniques and new antineoplastic therapies.