Role of diffusion-weighted MR imaging in the initial diagnosis of soft tissue tumours

Abstract

Background

Soft tissue tumours (STT) constitute a heterogeneous group of lesions frequently studied by Magnetic Resonance Imaging (MRI). It has not yet been clearly established whether the inclusion of apparent diffusion coefficient (ADC) diffusion-weighted imaging (DWI) maps would better determine tumour aggressiveness.

Purpose

To assess the diagnostic value of quantitative ADC DWI maps in the initial diagnosis of STT; and to determine whether the inclusion of DWI provides more valuable information than conventional sequences alone.

Material and methods

Retrospective study of patients with histologically proven STT. Conventional morphological MRI sequences and the DWI sequence were analysed. The ADC was quantified using a region of interest (ROI) that covered the largest sectional area (global ADC) and another that selected the area of ​​greatest restriction (selected ADC). Differences in ADC values ​​were analysed between both benign and malignant lesions and high and low-grade sarcomas. A multivariate analysis was performed to determine the ability of ADC ​​to adequately diagnose the nature of STTs when associated with other morphological characteristics.

Results

84 patients with STT, of which 40 were benign and 44 malignant. The malignant group included 10 low-grade sarcomas, 23 high-grade sarcomas, 4 non-sarcomatous neoplasms and 7 sarcomas with no histological grading. The ADC values ​​were significantly higher in benign lesions for the selected ADC. Significantly higher selected ADC values ​​were also obtained in low-grade sarcomas. In the multivariate analysis, the highest diagnostic precision values were obtained when morphological features and ADC were included, with a sensitivity, specificity, and area under the curve (AUC) of 84, 75 and 91%, respectively.

Conclusion

The inclusion of ADC DWI values ​​improves the diagnostic accuracy of MRI for STTs, especially when used in combination with conventional MRI sequences.