Diagnostic Value of Dynamic Contrast-Enhanced (DCE), Dynamic Susceptibility Contrast (DSC) and Arterial Spin Labeling (ASL) Perfusion MRI for Differentiation of High-Grade and Low-Grade Gliomas

Abstract

Objective: To evaluate dynamic contrast-enhanced (DCE), dynamic susceptibility contrast (DSC), and arterial spin labeling (ASL)-derived perfusion magnetic resonance imaging (MRI) parameters as a non-invasive technique for differentiating between high-grade gliomas (HGGs) and low-grade gliomas (LGGs), and to determine the diagnostic value of each parameter. Materials and Methods: Twenty-four patients with histopathologically confirmed HGGs or LGGs underwent DCE-, DSC-, and ASL-magnetic resonance perfusion (MRP). Retrospective qualitative and quantitative assessment of MRP-derived parameters, including DCE-Ktrans, DCE-Ve, DCE-Vp, DSC-rCBV, DSC-rCBF, and ASL-rCBF were performed, and the diagnostic value of each parameter was determined using ROC analysis. Results: Of the 24 patients enrolled in the present study, which included 10 LGGs and 14 HGGs, DCE-derived Vp showed the best diagnostic performance for differentiating between HGGs and LGGs (AUC 0.833, cutoff >0.0002 mL/100 g, p=0.018, 100% sensitivity, 28.6% specificity), followed by DCE-derived Ktrans (AUC 0.75, cutoff >0.024 min-1, p=0.011, 58.3% sensitivity, 100% specificity) and DSC-derived normalized rCBV (AUC 0.75, p=0.021, cutoff >1.15, 100% sensitivity, 37.5% specificity). The ASL-derived normalized rCBF showed no statistically significant difference between HGGs and LGGs (AUC 0.457, p=0.757). Conclusion: DCE-derived Vp, DCE-derived Ktrans, and DSC-derived rCBV are helpful perfusion MRI parameters for differentiating HGGs and LGGs, with DCE-derived Vp showing the best diagnostic performance in the present study.