ORIGINAL RESEARCH ADULT BRAIN Comparison of the Diagnostic Accuracy of DSC- and Dynamic Contrast-EnhancedMRI in the Preoperative Grading

BACKGROUNDANDPURPOSE: Dynamic contrast-enhancedMR imaging parameters can be biased by poormeasurement of the vascular input function. We have compared the diagnostic accuracy of dynamic contrast-enhanced MR imaging by using a phase-derived vascular input function and “bookend ” T1 measurements with DSC MR imaging for preoperative grading of astrocytomas. MATERIALSANDMETHODS: This prospective study included48patientswith anewpathologicdiagnosis of an astrocytoma. PreoperativeMR imaging was performed at 3T, which included 2 injections of 5-mL gadobutrol for dynamic contrast-enhanced and DSC MR imaging. During dynamic contrast-enhanced MR imaging, both magnitude and phase images were acquired to estimate plasma volume obtained from phase-derived vascular input function (Vp_) and volume transfer constant obtained from phase-derived vascular input function (Ktrans_) as well as plasma volume obtained from magnitude-derived vascular input function (Vp_SI) and volume transfer constant obtained from magnitude-derivedvascular input function (Ktrans_SI). FromDSCMR imaging, corrected relativeCBVwascomputed. FourROIswereplacedover the solidpart of the tumor, and the highest value among the ROIs was recorded. A Mann-Whitney U test was used to test for difference between grades. Diagnostic accuracy was assessed by using receiver operating characteristic analysis. RESULTS: Vp _ and Ktrans_ values were lower for grade II compared with grade III astrocytomas (P .05). Vp_SI and Ktrans_SI were not significantly different between grade II and grade III astrocytomas (P .08–0.15). Relative CBV and dynamic contrast-enhanced MR imaging parameters except forKtrans_SIwere lower for grade III comparedwith grade IV (P .05). In differentiating low- andhigh-grade astrocytomas,w