Multimodality fMRI with perfusion, diffusion-weighted MRI and 1 H-MRS in the diagnosis of lympho-associated benign and malignant lesions of the parotid gland.

Abstract

Differential diagnosis of the mucosa-associated lymphoid tissue lymphoma (MALToma) and tumor-like benign lymphoepithelial lesion (BLEL) in the parotid gland is difficult. To distinguish MALToma and BLEL with multimodality MRI including hydrogenproton magnetic resonance spectroscopy (1 H-MRS), diffusion-weighted imaging (DWI-MR), and dynamic contrast-enhanced (DCE-MR), and evaluate each sequence. Retrospective. Twenty-five patients with parotid tumor-like BLEL and 20 with parotid MALToma. 1.5-T/T1 WI, T2 WI, single-voxel 1 H-MRS, DWI-MR, and DCE-MR. All MR images were interpreted and agreed upon by two radiologists who were blinded to clinical information and histopathologic results. The imaging diagnoses were then compared to the histopathologic results. Youden index was used to determine the optimized threshold value. The receiver operating characteristic (ROC) curve was used to evaluate the diagnostic efficiency of different functional (f)MRI methods. Fisher's exact test indicated a significant difference between the 1 H-MRS images of the two lesions (P < 0.001). The sensitivity, specificity, and accuracy of positive choline (Cho) peak in 1 H-MRS of parotid MALToma were 80%, 76%, and 77.7%, respectively. The mean apparent diffusion coefficient (ADC) was 0.992 × 10-3 mm2 /s in patients with parotid tumor-like BLEL and 0.634 × 10-3 mm2 /s in patients with parotid MALToma, and the difference was statistically significant (t-test, P < 0.001). Choosing the Youden index as 0.669 × 10-3 mm2 /s, the sensitivity, specificity, and accuracy of the assay were 78.9%, 95.8%, and 88.4%, respectively. Assuming that time-intensity curve (TIC) type I indicated parotid MALToma (positive), and type II and type III indicated parotid tumor-like BLEL (negative), the sensitivity, specificity, and accuracy of time-to-peak (TTP) and initial slope of increase (ISI) in diagnosing MALToma were 94.1%, 95.2%, and 94.7%, respectively. Combining methods of TTP, ADC, and Cho peak reached the highest AUC (1.000). Combined use 1 H-MRS, DWI-MR, and DCE-MR increased the accuracy of the differential diagnosis between these lesions to 100%. Cho peak in 1 H-MRS, ADC less than 0.669 × 10-3 mm2 /s, TIC type I together indicated parotid MALToma. 3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:423-432.