Practice variations in salivary gland imaging and utility of virtual unenhanced dual energy CT images for the detection of major salivary gland stones

Abstract

Background

There are variations in computed tomography (CT) protocols used for evaluating sialolithiasis, with some protocols including acquisitions before and after IV contrast administration. Dual-energy CT (DECT) can be used to generate virtual unenhanced (VUE) images, potentially precluding the need for an uninfused scan.

Purpose

In this study, we performed a survey in order to assess variations in the imaging approach for sialolithiasis and evaluated the accuracy of DECT VUE images for the detection of salivary stones.

Material and Methods

Practice variations were evaluated by an online survey of the membership of the American Society of Neuroradiology. We then identified 28 patients with salivary gland calcifications matched with 28 negative controls that had both an unenhanced and a contrast-enhanced acquisition performed as DECT. A total of 123 major salivary gland calcifications and 85 tonsilloliths were evaluated and the true unenhanced series was used as gold standard.

Results

The survey revealed substantial variations in CT protocols used for sialolithiasis. On a per-patient basis, DECT VUE had 96.4% sensitivity and 100% specificity. On a per-calcification basis, sensitivity and specificity was 100% for stones > 2 mm but dropped for smaller calcifications. The false-negative cases corresponded to clinically insignificant, intra-glandular parotid calcifications. Inter-reader agreement was excellent (0.9256).

Conclusion

This study confirms that there are significant variations in CT protocols used for evaluation of sialolithiasis. A single contrast-enhanced DECT acquisition with reconstruction of VUE may represent an attractive, streamlined alternative and enable the elimination of the true unenhanced phase and associated radiation exposure.