Treatment results of glioblastoma (GB) during the last 30 years in Tohoku University were analyzed to identify any improvements in patient outcome in all 332 histologically proven cases of newly diagnosed GB treated consecutively in our department between 1982 and 2011. These 30 years was divided into 5 treatment eras, Group 1 (1982-1988, without preoperative evaluation by magnetic resonance [MR] imaging, n = 46), Group 2 (1989-1996, with preoperative MR imaging, n = 41), Group 3 (1997-1999, additionally underwent intraoperative functional brain mapping and neuronavigation system, n = 38), Group 4 (2000-August 2006, underwent 30 Gy of whole brain radiation followed by 30 Gy of extended local accelerated hyperfractionated radiation therapy, n = 96), and Group 5 (September 2006-2011, adjuvant usage of temozolomide [TMZ], n = 111). Overall survival (OS) was calculated from the date of surgery to the death from any cause. The median survival time/2-year OS/5-year OS of Groups 1 to 5 were 10.7 months/10.9%/0%, 17.3 months/26.2%/6.9%, 15.9 months/23.7%/5.3%, 20.1 months/34.8%/15.5%, and 20.9 months/45.5%/19.7%. The prognosis for patients with GB improved significantly after the introduction of MR imaging. Younger GB, defined as patients aged below 60 years, or total tumor resection with all ages in Group 5 had 5-year 0S of 31.0% and 30.1%, respectively. The prognosis of GB was improved significantly after the introduction of TMZ for elderly GB, recursive partitioning analysis class 5, or totally resected GB. Introduction of MR imaging and TMZ, and total resection of the tumor were important in the improvement of outcome for patients with GB.