Virtual non-enhanced dual-energy CT reconstruction may replace true non-enhanced CT scans in the setting of suspected active hemorrhage. Issue 109 (December 2018)

• Record Type: Journal Article
• Authors: Walter, Sven S.; Schneeweiß, Sven; Maurer, Michael; Kraus, Mareen S.; Wichmann, Julian L.; Bongers, Malte N.; Lescan, Mario; Bamberg, Fabian; Othman, Ahmed E.
• Is Part Of: European journal of radiology. Issue 109(2018)
• Page Start: 218
• Page End: 222
• Publication Date: 2018-12
• Subjects: Dual energy CT -- Hemorrhage -- Bleeding -- Virtual unenhanced -- True non-enhanced; Medical radiology -- Periodicals; Radiology -- Periodicals; Radiologie médicale -- Périodiques; Medical radiology; Periodicals; 616.075705
• DOI: 10.1016/j.ejrad.2018.10.026
• Languages: English
• ISSNs: 0720-048X

Abstract

Abstract: Objective: To assess virtual non-enhanced dual energy computed tomography (DECT) reconstruction as a potential alternative to true non-enhanced thoracic and abdominal CT scans in patients with suspected active hemorrhage. Methods and Materials: Ninety-seven consecutive patients with suspected active bleeding, who underwent routine triphasic CT with intravenous contrast, were included. All patients underwent single energy non-enhanced (TNE) and arterial phase CT as well as venous DECT. Using commercially available software, venous dual energy images were used to generate virtual non-enhanced images (VNE). TNE and VNE were evaluated regarding subjective image quality and diagnostic confidence on a 5-point-scale. The consistency of interpretation with respect to presence and type of hemorrhage between conventional triphasic CT including TNE and biphasic CT with simulated VNE was assessed by two blinded independent readers. Results: Among the included subjects (mean age: 66.3 ± 14.5 years, 63.9% male), 37.1% were diagnosed with acute hemorrhage. VNE yielded significantly lower image quality but superior reduction of artefacts as compared to TNE (p ≤ 0.001). Diagnostic confidence was rated high for both TNE and VNE with a significant superiority of TNE (p = 0.003). Triphasic and biphasic CT datasets yielded perfect intrareader agreement regarding presence and type of hemorrhage. The interrater agreement was almost perfect (K = 0.955). Radiation dose was significantly reduced using biphasic dual energy CT with VNE (p ≤ 0.001). Conclusion: Virtual non-enhanced images yield comparable diagnostic confidence and perfect agreement with true non-enhanced images for assessment of acute hemorrhage. This suggests that true non-enhanced images could be replaced by virtual non-enhanced images resulting in a significant reduction of radiation dose.