Erschienen in:
23.11.2016 | Neuro
Dual-energy CT of the brain: Comparison between DECT angiography-derived virtual unenhanced images and true unenhanced images in the detection of intracranial haemorrhage
verfasst von:
Matteo Bonatti, Fabio Lombardo, Giulia A. Zamboni, Patrizia Pernter, Roberto Pozzi Mucelli, Giampietro Bonatti
Erschienen in:
European Radiology
|
Ausgabe 7/2017
Einloggen, um Zugang zu erhalten
Abstract
Objective
To evaluate the diagnostic performance of virtual non-contrast (VNC) images in detecting intracranial haemorrhages (ICHs).
Methods
Sixty-seven consecutive patients with and 67 without ICH who underwent unenhanced brain CT and DECT angiography were included. Two radiologists independently evaluated VNC and true non-contrast (TNC) images for ICH presence and type. Inter-observer agreement for VNC and TNC image evaluation was calculated. Sensitivity and specificity of VNC images for ICH detection were calculated using Fisher’s exact test. VNC and TNC images were compared for ICH extent (qualitatively and quantitatively) and conspicuity assessment.
Results
On TNC images 116 different haemorrhages were detected in 67 patients. Inter-observer agreement ranged from 0.98–1.00 for TNC images and from 0.86–1.00 for VNC images. VNC sensitivity ranged from 0.90–1, according to the different ICH types, and specificity from 0.97–1. Qualitatively, ICH extent was underestimated on VNC images in 11.9% of cases. Haemorrhage volume did not show statistically significant differences between VNC and TNC images. Mean haemorrhage conspicuity was significantly lower on VNC images than on TNC images for both readers (p < 0.001).
Conclusion
VNC images are accurate for ICH detection. Haemorrhages are less conspicuous on VNC images and their extent may be underestimated.
Key points
• VNC images represent a reproducible tool for detecting ICH.
• ICH can be identified on VNC images with high sensitivity and specificity.
• Intracranial haemorrhages are less conspicuous on VNC images than on TNC images.
• Intracranial haemorrhages extent may be underestimated on VNC images.