Discriminating glioblastoma from solitary brain metastases on 3 Tesla magnetic resonance imaging: the roles of fractional anisotropy and mean diffusivity

D.-H. Nguyen, T.-D. Le, H.-V. Nguyen, V.-A. Nguyen-Thi, D.-H. Nguyen, H. Dong-Van, M.-D. Nguyen

Department of Radiology, Hanoi Medical University, Hanoi, Vietnam. bsnguyenminhduc@pnt.edu.vn

---

• Neurology
• Radiology

OBJECTIVE: This study determined the diagnostic value of diffusion tensor imaging (DTI) sequences using fractional anisotropy (FA) and mean diffusivity (MD) for discriminating glioblastoma (GBM) from solitary brain metastases (SBM) using 3 Tesla magnetic resonance imaging (MRI).

PATIENTS AND METHODS: A retrospective study was conducted, including 40 patients who underwent biopsy or surgery and received a histological diagnosis of GBM or SBM between August 2020 and December 2021. All preoperative examinations were performed on 3 Tesla MRI using conventional and DTI sequences. Three regions of interest (ROIs) were placed to measure a solid tumor component, peritumoral edema, and the opposite normal white matter to evaluate FA and MD values. Parametric and nonparametric statistical tests were used to determine differences between GBM and SBM. The diagnostic value of significantly different parameters between the two tumor entities was analyzed using the receiver operating characteristic (ROC) curve.

RESULTS: The FA value for peritumoral edema (eFA) in GBM cases was significantly larger than that in SBM cases (p < 0.05), with no significant difference in MD values. The FA and MD values for the solid tumor component (sFA and sMD, respectively) and the ratio of the sFA value to the FA value of the opposite normal white matter (rFAs/n) in GBM cases were significantly larger than those in SBM cases (p < 0.05). Combining the sFA and sMD values provided the highest area under the ROC curve (AUC) value of 0.96, with a sensitivity, specificity, positive predictive value, and negative predictive value of 85.2%, 100%, 85.2%, and 87.1%, respectively, for distinguishing GBM from SBM.

CONCLUSIONS: MRI parameters, including sFA, sMD, eFA, and rFAs/n, are useful for differentiating between GBM and SBM. The combination of sFA and sMD may increase the diagnostic performance of MRI for these two tumor entities.

Free PDF Download