Andrew Sher MD, Abstract Co-Author: Research Grant, Koninklijke Philips NV

Abed Ghandour MD, Abstract Co-Author: Nothing to Disclose

Amar Dhanantwari, Abstract Co-Author: Employee, Koninklijke Philips NV

Luis Alberto Landeras MD, Abstract Co-Author: Institutional Grant support, Koninklijke Philips NV

Prabhakar Rajiah MD, FRCR, Presenter: Institutional Research Grant, Koninklijke Philips NV

Utilize the novel Spectral Detector CT (SDCT) system to retrospectively augment pulmonary artery enhancement on routine chest CT and assess whether resulting image sets enable diagnostic evaluation of pulmonary vasculature. 

29 patients underwent routine contrast-enhanced chest scans with SDCT (Philips Healthcare) following 70 ml IV contrast. 21 studies not in ideal angiographic phase were chosen, defined as not achieving attenuation greater than 200 HU in the main pulmonary artery. Monoenergetic image sets from 40 to 180 keV were retrospectively created. Images were assessed at fixed window level (30) and width (450). A 5-point scale measured subjective evaluation of central and peripheral arterial enhancement and image noise (1=Non-diagnostic, 5=Excellent). An ideal image set was chosen, defined as the highest energy set with attenuation greater than 200 HU in the main PA and a subjective image evaluation of diagnostically acceptable. At ideal energy, mean attenuation, noise, and signal-to-noise ratios of 9 ROIs throughout the pulmonary vasculature as well as subjective image quality were compared to the standard 120 kVp study. Paired t-test and Wilcoxon signed-rank test were used for analysis. 

The mean optimal monoenergetic level was 53 keV ± 7.3. Optimal reconstructions showed higher attenuation than 120 kVp studies (241 ± 81 vs. 133 ±53 HU; p<.001). Image noise was significantly higher in monoenergetic studies (47.7±86.4 vs. 22.6 ±23.7; p<.001), while overall SNR remained significantly higher (13.1 ± 9.5 vs. 10.3 ± 6.4; p <.05). Subjective vascular enhancement within monoenergetic images was higher than 120 kVp images (Central: 4.2 ± .4 vs. 1.9 ± .8; p<.001, Peripheral: 4.6 ±.5 vs. 1.6 ± .7, p<.001). Subjective image noise was higher on monoenergetic images (4.4 ± .7 vs. 4.9 ± .3, p<.05) however all studies maintained diagnostic acceptability.   

Optimal monoenergetic energy reconstruction of routine chest CT obtained on SDCT significantly increased pulmonary arterial enhancement and SNR, achieving attenuation levels comparable with pulmonary angiographic exams and maintaining diagnostic acceptability.   

SDCT enables creation of an optimal image set of the pulmonary vessels that can allow creation of CTPA from a routine or low-dose chest CT.

Sher, A, Ghandour, A, Dhanantwari, A, Landeras, L, Rajiah, P, Evaluation of Monoenergetic Reconstruction on Pulmonary Angiography Using Spectral Detector CT.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14019232.html