Diagnostic value of minimum apparent diffusion coefficient values in prediction of neuroepithelial tumor grading

Abstract

To retrospectively evaluate the diagnostic accuracy of diffusion weighted image (DWI) in the prediction of neuroepithelial tumors grading, and to appraise the apparent diffusion coefficient (ADC) value of neuroepithelial tumors with histologic findings as a reference standard. ADC values in 110 patients with pathologically proved neuroepithelial tumors, including 77 astrocytic tumors, 16 oligodendroglial tumors, 11 oligoastrocytic tumors, and 6 ependymal tumors, were investigated retrospectively. The minimum ADC (MinADC) value of tumors was measured postoperatively on ADC maps, avoiding cystic, necrotic, or hemorrhagic components. The Ki-67 Labeling Index (Ki-67 LI) was determined by immunohistochemistry. The patients were classified into low (WHO II) and high (WHO III or IV) grade groups. Correlation analysis, Student t-test, Welch test, receiver operating characteristic (ROC) analysis, and analysis of variance were used for statistical evaluation. There was a negative correlation between MinADC value and Ki-67 LI (P < 0.001). The mean MinADC value (1.057 x 10(-3) mm(2)/s) of low grade group was higher than that (0.773 x 10(-3) mm(2)/s) of high grade group. The area under the ROC curve (AUC) was 0.809, and the cutoff MinADC value of 0.900 x 10(-3) mm(2)/s for the differentiation between high and low grade neuroepithelial tumors provided the best combination of sensitivity (85.4%) and specificity (71.0%). MinADC value may be a simple and effective optional tool for the prediction of neuroepithelial tumor grading.