Relative percentage signal intensity recovery of perfusion metrics-an efficient tool for differentiating grades of glioma
No Thumbnail Available
Date
2015
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
BRITISH JOURNAL OF RADIOLOGY
Abstract
Objective: Glioma classification and characterization may be facilitated by a multiparametric approach of perfusion metrics, which could not be achieved by conventional MRI alone. Our aim is to explore the potential of relative percentage signal intensity recovery (rPSR) values, in addition to relative cerebral blood volume (rCBV) and relative cerebral blood flow (rCBF) of first-pass T-2* dynamic susceptibility contrast (DSC) perfusion MRI, in differentiating high-and low-grade glioma. Methods: This prospective study included 39 patients with low-grade and 25 patients with high-grade glioma. rPSR, rCBV and rCBF were calculated from the first-pass T-2* DSC perfusion MRI. rPSR was calculated using standard software and validated with dedicated perfusion metrics analysis software. The statistical analysis was performed using analysis of variance and receiver operating characteristic (ROC) curves. Results: Variation in rPSR, rCBV and rCBF values between low-and high-grade gliomas were statistically significant (p<0.005). The ROC curve analysis for each of them yielded 96% sensitivity and 71.8% specificity; 88% sensitivity and 69.2% specificity; and 72% sensitivity and 66.7% specificity. The area under the curve (AUC) from the ROC curve analysis yielded 0.893, 0.852 and 0.702 for rPSR, rCBV and rCBF, respectively. The rPSR calculation with the validation software yielded 92.3% sensitivity and 72% specificity with an AUC of 0.864. Conclusion: rPSR inversely correlates while rCBV and rCBF values directly correlate with the tumour grade. Furthermore, the overall diagnostic performance of rPSR is better than rCBV and rCBF values. Advances in knowledge: rPSR of T-2* DSC perfusion is an indicator of blood-brain barrier status and lesion leakiness, which has not been explored yet compared with the usual haemodynamic parameters, rCBV and rCBF.
Description
Keywords
Radiology, Nuclear Medicine & Medical Imaging
Citation
88 ,1052;-