Introduction: The aims of the study were to compare the diagnostic performance of a combination of virtual non-contrast (VNC) images and arterial images obtained from a single-phase dual-energy CT (DECT) acquisition and standard non-contrast and arterial images from a biphasic protocol and to study the potential radiation dose reduction of the former approach.
Methods: All DECT examinations performed for evaluation of parathyroid adenomas during a 13-month period were retrospectively reviewed. An initial single-energy unenhanced acquisition was followed by a dual-energy arterial phase acquisition. "Virtual non-contrast images" were generated from the dual-energy acquisition. Two independent and blinded radiologists evaluated three different sets of images during three reading sessions: single arterial phase, single-phase DECT (virtual non-contrast and arterial phase), and standard biphasic protocol (true non-contrast and arterial phase). The accuracy of interpretation in lateralizing an adenoma to the side of the neck and localizing it to a quadrant in the neck was evaluated.
Results: Sixty patients (mean age, 65.5 years; age range, 38-87 years) were included in the study. The lateralization and localization accuracy, sensitivity, and positive predicted value (PPV) and negative predicted value (NPV) of the different image datasets were comparable. The combination of VNC and arterial images was more specific than arterial images alone to lateralize a parathyroid lesion (OR = 1.93, p = 0.043). The use of the single-phase protocol resulted in a calculated radiation exposure reduction of 52.8 %.
Conclusions: Virtual non-contrast and arterial images from a single DECT acquisition showed similar diagnostic accuracy than a biphasic protocol, providing a significant dose reduction.
Keywords: Computed tomography; Dual energy; Hypercalcemia; Parathyroid adenoma; Radiation dose.