Characteristics and Role of Apparent Diffusion Coefficient (ADC) Value in Musculoskeletal Soft Tissue Tumor

Introduction: Musculoskeletal soft tissue tumors (STTs) consist of a group of malignant and benign tumors that present challenges in diagnosis and therapy. Adding diffusion-weighted imaging and apparent diffusion coefficient (DWI/ADC) ​​provides tissue characteristics that are different from those obtained by conventional MR techniques. Quantitative examination of the diffusion of water molecules in tissue is expressed as ADC value, which is expected to be more helpful in differentiating malignant and benign soft tissue masses. This study was carried out to determine the characteristics and role of ADC values ​​in musculoskeletal STTs. Methods: This research was conducted retrospectively. Analysis was performed using MRI results of patients with musculoskeletal STTs who met the inclusion and exclusion criteria. Three regions of interest (ROI) were assessed in the tumor area with the strongest contrast enhancement for DWI mapping, ADC, and ROI ADC value placement. An Independent T-test test was performed to determine the mean difference of ADC in each tumor category. Receiver Operating Characteristics (ROC) were generated to find the optimal cut-off value. Then, a diagnostic test was carried out to determine the sensitivity, specificity, PPV, and NPV of the ADC value. Results: There were 41 research subjects with musculoskeletal STTs with male predominance, in which 75.61% of cases were malignant. DWI/ADC showed that 78.05% of patients with musculoskeletal soft tissue tumors had restricted diffusion areas. In this study, the ADC cutoff value to differentiate benign and malignant musculoskeletal STT is 0,84 x 10-3mm2/s with 77.4% sensitivity and 90% specificity, 96% positive predictive value and 56.3% negative predictive value. Conclusion: In differentiating malignant and benign musculoskeletal STTs, the ADC value provides fair sensitivity, very good specificity, good positive predictive value, and low negative predictive value.