Diagnostic performance of diffusion tensor imaging for preo- perative glioma grading

Area of Science:

• Neuroimaging
• Oncology
• Radiology

Background:

• Glioma grading is crucial for treatment planning and prognosis.
• Accurate differentiation between low-grade glioma (LGG) and high-grade glioma (HGG) remains a clinical challenge.
• Diffusion tensor imaging (DTI) offers potential for non-invasive glioma characterization.

Purpose of the Study:

• To evaluate the diagnostic performance of fractional anisotropy (FA) and mean diffusivity (MD) values for glioma grading.
• To determine the utility of absolute and relative FA and MD values in distinguishing LGG from HGG.
• To assess the combined diagnostic efficacy of selected DTI parameters.

Main Methods:

• Retrospective analysis of 42 patients with histologically confirmed glioma.
• Preoperative 3 Tesla MRI including DTI and conventional sequences.
• Measurement of FA and MD values in the tumor core, peritumoral area, and normal white matter.
• Receiver operating characteristic (ROC) curve analysis to assess diagnostic performance.

Main Results:

• Tumor MD (tMD), relative MD (rMDt/w), and relative peritumoral FA (rFAp/w) showed significant differences between LGG and HGG groups.
• The combination of tMD, rMDt/w, and rFAp/w achieved an area under the curve (AUC) of 89%.
• This combined parameter set demonstrated high specificity (100%) and positive predictive value (100%) for distinguishing LGG from HGG.

Conclusions:

• The studied DTI parameters (tMD, rMDt/w, rFAp/w) are valuable for differentiating LGG and HGG.
• Combining these indices enhances diagnostic specificity and accuracy in glioma grading.
• DTI-derived metrics show promise as non-invasive biomarkers for glioma classification.