Purpose: CT artifacts from port-systems are a common problem in staging- and restaging-examinations and reduce image quality and diagnostic assessment. The purpose of this study was to investigate the reduction of these artifacts using virtual monoenergetic images (VMI) from dual-energy spectral-detector CT (SDCT) in comparison to conventional CT-images (CI).
Method: 50 SDCT-datasets of patients with artifacts from port-chamber and port-catheters were included in this IRB-approved, retrospective study. CI and VMI (range, 40-200 keV, 10 keV increment) were reconstructed from the same acquisition. The quantitative image analysis was performed ROI-based assessing mean and standard deviation of attenuation (HU) in most pronounced hypo- and hyperdense artifacts surrounding to the port-chamber and the distal end of the port-catheter in the superior vena cava. Subjectively, artifact reduction and diagnostic assessment of surrounding soft tissue were rated on 5-point Likert-scales.
Results: In comparison to CI, VMI of higher keV-values showed strong reduction of hypo- and hyperattenuating artifacts around the port-chamber and port-catheter (CI/VMI200keV: hypodense -104.7 ± 124.7HU/10.8 ± 58.1HU and -101.6 ± 101.5HU/-36.7 ± 32.9HU; hyperdense 240.8 ± 151.6HU/79.6 ± 81.3HU and 108.6 ± 129.3HU/25.9 ± 31.9HU; all p < 0.001). Image noise could also be reduced significantly. The subjective analysis showed significantly reduced artifacts around the port-chamber and port-catheter (CI/VMI200keV: hypodense 3(1-4)/5(4-5) and 3(2-4)/5(4-5); hyperdense 3(1-4)/5(4-5) and 3(2-3)/5(3-5); all p < 0.001) and improved diagnostic assessment of pectoral/subclavian soft tissue for VMI of ≥100keV. Ratings for diagnostic assessment were best between 140-200 keV. Overall interrater agreement was high (ICC = 0.79).
Conclusions: Higher keV VMI enabled a significant reduction of artifacts from port-systems around the chamber and the catheter leading to improved assessment of surrounding soft tissue.
Keywords: Artifacts; Neoplasm staging; Port system; X-ray computed tomography.
Copyright © 2019. Published by Elsevier B.V.