Diagnostic performance of apparent diffusion coefficient parameters for glioma grading.
Qun WangJiaShu ZhangXinghua XuXiaoLei ChenBainan XuPublished in: Journal of neuro-oncology (2018)
This study was to evaluate the diagnostic performance of fractional anisotropy (FA) and apparent diffusion coefficient (ADC) parameters derived from diffusion tensor imaging in the differentiation between grade II and III gliomas. The records of 60 patients (30 women, 30 men; mean age, 45.4 years) suspected of having gliomas who underwent an ADC image-guided stereotactic biopsy were retrospectively reviewed. The values of FA and ADC were measured, and the sensitivity, specificity, accuracy and area under the curve (AUC) of those parameters were calculated based on the receiver operating characteristic curve analysis. A predictive diagnostic equation was also constructed and evaluated. Significant differences in minimum ADC values were found in the quantitative analysis between the grade III and II glioma groups. The sensitivity, specificity, positive predictive values (PPV), negative predictive values (NPV), accuracy and AUC for identifying grade III and II gliomas at the optimum cut-off value of 0.895 × 10-3 mm2/s of minimum ADC were 81.0, 89.1, 77.3, 91.1, 86.6 and 0.87, respectively. The predictive diagnostic equation was superior to the single minimum ADC indicator with a sensitivity of 90.5%, a specificity of 84.8%, a PPV of 73.1%, an NPV of 95.1%, and an accuracy of 86.6%, respectively. The study provides evidence that minimum ADC values have a superior diagnostic performance in differentiating grade III and II gliomas, and the predictive diagnostic equation may be helpful in the differentiation.
Keyphrases
- diffusion weighted imaging
- contrast enhanced
- diffusion weighted
- high grade
- magnetic resonance imaging
- end stage renal disease
- ejection fraction
- magnetic resonance
- peritoneal dialysis
- skeletal muscle
- pulmonary embolism
- small cell lung cancer
- prognostic factors
- wastewater treatment
- ultrasound guided
- data analysis
- breast cancer risk