Glioma grading: sensitivity, specificity, positive and negative predictive values of diffusion and perfusion imaging

• Published in: Journal of neuro-oncology 2009-08, Vol.94 (1), p.87-96
• Main Authors: Arvinda, H R, Kesavadas, C, Sarma, P S, Thomas, B, Radhakrishnan, V V, Gupta, A K, Kapilamoorthy, T R, Nair, S
• Format: Article
• Language: English
• Subjects: Adult; Brain Neoplasms - diagnosis; Brain Neoplasms - physiopathology; Cerebrovascular Circulation - physiology; Diffusion Magnetic Resonance Imaging - methods; Female; Glioma - diagnosis; Glioma - physiopathology; Humans; Magnetic Resonance Imaging; Male; Middle Aged; ROC Curve; Sensitivity and Specificity
• ISSN: 0167-594X; 1573-7373
• DOI: 10.1007/s11060-009-9807-6

The purpose of our study was to determine the statistical significance of thresholds of relative cerebral blood volume (rCBV), apparent diffusion coefficient (ADC) and ADC ratios in grading cerebral gliomas. In this retrospective study, 51 patients with histopathologically confirmed primary cerebral gliomas who had undergone conventional MR imaging, dynamic contrast-enhanced T2*-weighted perfusion MR imaging and diffusion MR imaging were included. A retrospective blinded analysis of the imaging findings including the perfusion and diffusion parameters was done. The rCBV measurements were obtained from regions of maximum perfusion. Minimum ADC values were obtained from the region of maximum hypointensity within the tumor and from the corresponding opposite white matter. Tumor grade determined with the two methods were then compared with the histopathologic grade. Mann-Whitney tests were performed to compare the DWI and PWI between tumor types. Receiver operating characteristic analyses were performed to determine optimum thresholds for tumor grading and also to calculate the sensitivity, specificity, PPV, and NPV for identifying high-grade gliomas. Statistical analysis demonstrated a threshold value of 2.91 for rCBV to provide sensitivity, specificity, PPV, and NPV of 94.7, 93.75, 90.0, and 96.8%, respectively, in determining high-grade gliomas. An ADC value of 98.50 mm(2)/s was defined as a threshold below which tumors were classified as high-grade gliomas and a sensitivity, specificity, PPV, and NPV of 90, 87.1, 81.81 and 93.10% respectively, were obtained. Significant differences were noted in the rCBV ratios, ADC and ADC ratios between low- and high-grade gliomas (P < 0.0001). Combining PWI and DWI with conventional MR imaging increases the accuracy of pre-operative imaging grading of glial neoplasms. The rCBV measurements had the most superior diagnostic performance in predicting glioma grade. Absolute ADC values or ADC ratios were also helpful in preoperative grading of gliomas. Threshold values can be used in a clinical setting to evaluate tumors preoperatively for histologic grade and provide a means for guiding treatment and predicting postoperative patient outcome.