• Xiaojuan Li, Hua Jin, Ying Lu, Joonmi Oh, Susan Chang, and Sarah J Nelson.
• Magnetic Resonance Science Center, Department of Radiology, University of California, San Francisco, 1 Irving St, San Francisco, CA 94143, USA.
• NMR Biomed. 2004 Feb 1; 17 (1): 10-20.

AbstractAlthough MR imaging (MRI) and MR spectroscopic imaging (MRSI) have been applied in the diagnosis and treatment planning for brain tumors, their prognostic significance has not yet been determined. The goal of this study was to identify pre-treatment MRI and MRSI parameters for patients with malignant glioma that may be useful in predicting survival. Two populations of patients with newly-diagnosed malignant glioma were examined with MRI and three-dimensional proton ((1)H) MRSI. Thirty-nine patients (22 grade 3 and 17 glioblastoma multiforme, GBM) were studied prior to surgery, and 33 GBM patients were studied after surgery but prior to treatment with radiation and chemotherapy. Signal intensities of choline (Cho), creatine (Cr), N-acetyl aspartate (NAA), and lactate/lipid (LL) were estimated from the spectra. Recursive partitioning methods were applied to parameters that included age, histological grade, MRI and MRSI variables to generate survival trees. Patients were grouped into high and low risk categories and the corresponding Kaplan-Meier curves were plotted for comparison between groups. The parameters that were selected by recursive partitioning as being predictive of poor outcome were older age, larger contrast enhancement, higher Cho-to-Cr, higher Cho-to-NAA, higher LL and lower Cr-to-NAA abnormalities. The survival functions were significantly different between the sub-groups of patients obtained from the survival tree for both pre-surgery and post-surgery data. The results of this study suggest that pre-treatment MRI and three-dimensional (1)H-MRSI provide information that predicts outcome for patients with malignant gliomas and have drawn attention to variables that should be examined prospectively in future studies using these techniques.Copyright 2004 John Wiley & Sons, Ltd.

33 keywords...

hide…