Differentiation between Progression from Pseudoprogression in Glioblastoma Using Diffusion Weighted Imaging Techniques

Purposes:
the purpose of this study is to investigate the diagnostic value of different diffusion parameters, especially the MK parameter, and compare it with the results of MRS and pathology as a standard for distinguishing true recurrence of glioblastoma from pseudoprogression.
Methods:
19 glioblastoma patients who had a new lesion with absorption of contrast material after surgery and radiotherapy were subjected to conventional MRI, DKI and multivoxel MRS imaging. Based on pathology findings and radiologist’s opinion, 13 patients had true recurrence and 6 patients had pseudoprogression of glioblastoma. FA, MD and
MKparametersbasedontheareaswithabsorption of contrast material in T1 images after injection, once in voxels with a volume of approximately
2.34 cm3 and corresponding to the MRS voxel and again in a smaller voxel with a volume of
0.004 cm3 was calculated. The mean differences of diffusion parameters between true recurrence and pseudoprogression of glioblastoma were compared using the Mann-Whitney test. The receiver operating characteristic curve was used to evaluate the differential detection efficiency of each parameter.
Findings and conclusions:
MK parameter between these two groups was significant with p-value=0.01. ROC analysis showed that MK parameter with AUC=0.93 with
sensitivity of 69% and specificity of 100%, also FA with AUC=0.67 and sensitivity of 67% and specificity of 69.2% and MD with AUC=0.64 with sensitivity of 83.3% and specificity of 53.8% It is able to distinguish between the true recurrence of glioblastoma and its false recurrence. The result of MK parameter in 78% of patients was consistent with MRS and
pathology results. This value for MD and FA parameters was 63 and 53%, respectively.
MK parameter has more sensitivity, specificity, accuracy, NPV and PPV than other studied diffusion parameters in differentiating true recurrence from pseudoprogression of glioblastoma.