Gliomas are the most common malignant brain tumor. Glioblastoma, WHO grade IV astrocytoma, is the most common subtype and unfortunately also the most aggressive subtype with median survival in population based cohorts being only 10 months. Extensive surgical resections followed by postoperative fractioned radiotherapy and concomitant and adjuvant temozolomide prolong survival and is the standard treatment. The investigators think there is significant potential in individualized surgical decision-making in glioblastoma management. The idea that some patients are amendable to radical surgery, while others should be treated more conservatively, is not controversial in other fields of oncology. The current concept in all patients with glioblastoma is maximum safe resection of the contrast enhancing tumor, but this may in selected cases be extended to simply maximum safe resection tailored to the patient and extent of disease at hand. Densely proliferating tumor cells have been found from at an average of 10 mm beyond the margins of contrast enhancement in high-grade gliomas. There are now several case series, using various definitions of supramarginal resection, but they have in common that they report a benefit of resection with a margin. This potential benefit also comes together with an associated neurological risk, making this approach unethical and simply not feasible in the patients with glioblastoma as a whole. Objective of this study is: To investigate if resection with a margin, that is significantly beyond the radiological contrast enhancement, improves survival in selected patients with glioblastoma.