Purpose: To evaluate the value of diffusion-weighted imaging and dynamic contrast-enhanced MRI for the diagnosis of parotid gland tumors.

Methods: Retrospective review of patients with surgically treated parotid tumors between January 2009 and June 2020, who underwent a preoperative parotid gland MRI including standard morphological sequences, diffusion-weighted echoplanar imaging with apparent diffusion coefficient measurement and T1-weighted gadolinium-enhanced dynamic MRI sequences with Fat Saturation. The lesion was classified between malignant vs benign and precisions regarding its histological type were given when possible. Imaging findings were compared with pathology results.

Results: Inclusion of 133 patients (mean age: 53 years). Multiparametric MRI had a sensitivity of 90.3%, a specificity of 77.5%, an overall accuracy of 80.5%, a positive predictive value of 54.9% and a negative predictive value of 96.3% to differentiate benign parotid tumor from malignant ones. Specificity (85.5%) and positive predictive value (67.6%) were improved for cases, where anatomical and functional MRI characteristics were conclusive and consistent with clinical findings.

Conclusions: Combining diffusion-weighted and gadolinium-enhanced dynamic sequences, in addition to morphological ones enables high (> 90%) sensitivity to detect malignant parotid gland tumors. It also gives the possibility to characterize pleomorphic adenomas and Warthin tumors and to avoid fine-needle aspiration in cases of typical imaging presentation and reassuring clinical findings.

Download full-text PDF

Source
http://dx.doi.org/10.1007/s00405-022-07590-6DOI Listing

Publication Analysis

Top Keywords

parotid gland
16
gadolinium-enhanced dynamic
12
diffusion-weighted gadolinium-enhanced
8
dynamic mri
8
gland tumors
8
positive predictive
8
mri
6
parotid
6
diffusion-weighted
4
dynamic
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!