The impact of diffusion and perfusion-weighted imaging on glioma grading

Determining the grade of a glioma is extremely important for treatment planning and prognosis prediction. The study aimed to evaluate the usefulness of multiparametric perfusion-weighted imaging (PWI) and diffusion-weighted imaging (DWI) in preoperative glioma grading. In this retrospective study, 63 individuals with brain tumors histologically confirmed, of which 23 had low-grade gliomas (LGGs) and 40 had high-grade gliomas (HGGs) were involved. We conducted this paper on apparent diffusion coefficient (ADC) maps using the entire tumor volume method, allowing us to use all ADC values of the tumor. Small-sample regions of interest (ROIs) were drawn to collect parameters of relative cerebral blood flow (rCBF), cerebral blood flow (CBF), and relative cerebral blood volume (rCBV), from both the tumor core and peritumoral edema. The PWI and DWI metrics were compared to identify the most accurate distinguishing HGGs and LGGs, analyze receiver operating characteristics (ROC), and evaluate the diagnostic performance using solitary parameters and combined. In diffusion MRI, there were significant differences in minimum ADC and mean ADC between LGGs and HGGs (p<0.05), with the larger area under the curve (AUC) of 0.898 found for mean ADC at a cut-off value of 1.275, with sensitivity of 82.6 % and specificity of 90 %. The maximum ADC value did not differ significantly (p>0.05). All perfusion parameters in both the tumor core and peritumoral edema area were significantly greater values in cases of HGG compared to LGG (p<0.001), with the highest AUC of 0.946 found for solid tumor rCBV value (rCBVt), the cut-off is 3.585, sensitivity of 85 % and specificity of 100 %. Combining mean ADC and rCBVt provided an excellent AUC of 0.975, a sensitivity of 92.5 %, and a specificity of 91.3 % for differentiating between HGGs and LGGs. Perfusion and diffusion MRI are valuable in discriminating between high-grade and low-grade gliomas, with the major criterion in the decision-making process being the combined mean ADC and rCBVt parameters.