Multi-parametric MR imaging using pseudo-continuous arterial-spin labeling and diffusion-weighted MR imaging in differentiating subtypes of parotid tumors

Purpose: To evaluate multi-parametric MR imaging using pseudo-continuous arterial-spin labeling (pCASL) and diffusion-weighted imaging (DWI) in differentiating subtypes of parotid tumors. Material and methods: This study was done on 61 consecutive patients (27 M, 34 F aged 15-75 ys: mean 42 ys) with parotid tumors that underwent pCASL and DWI with the calculation of tumor blood flow (TBF) and apparent diffusion coefficient (ADC). Results: The mean TBF and ADC of parotid malignancy (n = 20) (53.1 +/- 6.9 mL/100 g/min and 0.66 +/- 0.1 x 10(-3) mm(2)/s) was a significant different (P = 0.001) than benign tumors (n = 41) (22.7 +/- 2.7 mL/100 g/min and 1.28 +/- 0.4 x 10(-3) mm(2)/s) respectively. Multi-parametric TBF and ADC were used to differentiate malignant from benign tumors has an AUC (AUC) of 0.97, and the accuracy of 90%. There was a significant difference (P = 0.001) TBF between Warthin tumors (WT) (26.7 +/- 9.7 mL/100 g/min) and malignancy. Selection of TBF of 30.5 mL/100 g/min to differentiate malignancy from WT revealed an AUC of 0.95 and the accuracy of 87%. There was a significant difference in TBF and ADC of pleomorphic adenomas (PA) from malignancy (P = 0.001). Multi-parametric TBF and ADC were used to differentiate PA from parotid malignancy revealed an AUC of 1.00. There was a significant difference in TBF and ADC of PA and WT (P = 0.001). Multi-parametric MRI using TBF and ADC used to differentiate PA from WT have an AUC of 0.99, and the accuracy of 92%. Conclusion: Multi-parametric MR imaging using pCASL and DWI is useful for differentiating benign parotid tumors from parotid malignancy and PA from WT and TBF helps in discrimination of WT from parotid malignancy.