Nakano Yoshiteru
   Department   School of Medicine  Neurosurgery, Clinical Medical Sciences
Article types journal article
Language English
Peer review Peer reviewed
Title Evaluation of Peritumoral Brain Parenchyma Using Contrast-Enhanced 3D Fast Imaging Employing Steady-State Acquisition at 3T for Differentiating Metastatic Brain Tumors and Glioblastomas.
Journal Formal name:World neurosurgery
Abbreviation:World Neurosurg
ISSN code:18788769/18788750
Domestic / ForeginForegin
Volume, Issue, Page 1-11頁
Author and coauthor Yamamoto Junkoh, Kakeda Shingo, Shimajiri Shohei, Nakano Yoshiteru, Saito Takeshi, Ide Satoru, Moriya Junji, Korogi Yukunori, Nishizawa Shigeru
Publication date 2018/08
Summary BACKGROUND:Metastatic brain tumors and glioblastomas are the 2 of the most common brain neoplasms in adults. However, distinguishing solitary metastatic brain tumors from glioblastomas on conventional magnetic resonance imaging remains particularly challenging. Thus, we aimed to retrospectively assess the role of contrast-enhanced fast imaging employing steady-state acquisition (CE-FIESTA) imaging in distinguishing between metastatic brain tumors and glioblastomas.MATERIALS AND METHODS:Forty-three patients with metastatic brain tumors and 14 patients with glioblastomas underwent conventional magnetic resonance imaging and CE-FIESTA before surgery. First, 1 neuroradiologist and 1 neurosurgeon classified the CE-FIESTA findings for the peritumoral brain parenchyma by consensus. Next, the 2 neuroradiologists performed an observer performance study comparing tumor shape classification (smooth or irregular margins), a classic imaging finding, with the CE-FIESTA classification of the peritumoral brain parenchyma.RESULTS:The CE-FIESTA findings for the peritumoral brain parenchyma were classified as follows: type A, no hyperintense rim; type B, partial hyperintense rim; and type C, extended hyperintense rim. With regard to the diagnosis of metastatic brain tumors, the observer performance study demonstrated that the mean sensitivity, specificity, and accuracy of an extended hyperintense rim classification (type C) on CE-FIESTA images were 95.3%, 85.7%, and 93.0%, respectively. The accuracy of the CE-FIESTA classification was significantly higher than that of the tumor shape classification.CONCLUSIONS:CE-FIESTA images may provide useful information for distinguishing metastatic brain tumors from glioblastomas, especially when focusing on differences in the peritumoral brain parenchyma.
DOI 10.1016/j.wneu.2018.08.147
PMID 30165229