1 3 J Neurooncol DOI 10.1007/s11060-016-2330-7 CLINICAL STUDY Survival outcomes following repeat surgery for recurrent glioblastoma: a single-center retrospective analysis Paolo Perrini 1  · Carlo Gambacciani 1  · Alessandro Weiss 1  · Francesco Pasqualetti 2  · Durim Delishaj 2  · Fabiola Paiar 2  · Riccardo Morganti 3  · Riccardo Vannozzi 1  · Lodovico Lutzemberger 1   Received: 21 April 2016 / Accepted: 8 November 2016 © Springer Science+Business Media New York 2016 hazards model was created demonstrating that preoperative performance status at recurrence (HR 0.418, p = 0.035) and the extent of repeat resection (HR 0.513, p = 0.043) were independent predictors of survival. Gross-total resection at repeat craniotomy is associated with longer overall survival and should be performed whenever possible in patients with recurrent glioblastoma and in good performance status. Keywords Extent of resection · Glioblastoma · Recurrent glioma · Surgery · Survival Introduction Glioblastoma is the most common primary brain tumor in adults and its prognosis remains dismal due to the high pro- pensity for tumor recurrence [1, 2]. The recurrence pattern after surgical resection and administration of temozolomide concurrent with radiation therapy is mainly local occur- ring within 2 cm of the tumor bed in approximately 80% of cases [3]. Over the recent years, treatment strategies have become more aggressive, with several studies reporting the clinical results of repeat surgery and salvage chemotherapy for recurrent high-grade gliomas. Recent investigations emphasized the importance of extent of resection (EOR) at the time of recurrence as an important predictor of over- all survival [4, 5]. However, only few studies analyzed the efect of EOR considering both the initial and repeat resec- tion [4]. In addition, the efective role of several variables, which may afect the prognosis of patients with recurrent glioblastoma, including age, preoperative performance sta- tus, tumor location and adjuvant treatment, are still matter of controversy. In this study we performed a retrospective analysis to evaluate the survival outcomes after reoperation Abstract The aim of the present study is to evaluate the impact of extent of resection at initial and repeat craniot- omy on overall survival of patients with recurrent glioblas- toma. The authors retrospectively reviewed the records of all adults patients who underwent repeat resection of recur- rent glioblastoma following radiation and chemotherapy at an academic tertiary-care institution between 2011 and 2015. We evaluated the survival outcomes with regard to extent of resection considering both the initial and repeat resections. The role of possible prognostic factors that may afect survival after repeat resection, including age, pre- operative performance status, tumor location and adjuvant treatment, was evaluated using Cox regression analyses. Forty-eight patients were included in this study. The overall median survival of 14 patients who had subtotal resection at recurrence after initial subtotal resection did not statisti- cally difer from seven patients who had gross-total resec- tion at recurrence after initial subtotal resection (18 months vs. 22 months, p = 0.583). The overall median survival of 13 patients who had gross-total resection at recurrence after initial gross-total resection was signiicantly increased compared with survival of 13 patients who had subtotal resection at recurrence after initial gross-total resection (47 months vs. 14 months, p = 0.009). A Cox proportional * Paolo Perrini paolo.perrini@unipi.it 1 Department of Neurosurgery, Azienda Ospedaliero Universitaria Pisana (AOUP), Via Paradisa 2, 56127 Pisa, Italy 2 Division of Radiation Oncology, Azienda Ospedaliero Universitaria Pisana (AOUP), Pisa, Italy 3 Department of Clinical and Experimental Medicine, Section of Statistics, University of Pisa, Pisa, Italy