To explore the optimal monochromatic level for the observation of coronary in-stent lumen by dual-layer spectral CT (DLCT).

Forty-nine patients with 74 stents after percutaneous coronary intervention (PCI) who underwent coronary CTA (CCTA) examinations by a DLCT between January 2016 and September 2017 were retrospectively enrolled. A total of 12 groups of images including 60-120 keV (kilo electron voltage) images with 10 keV interval, 140-200 keV images with 20 keV interval and conventional images. In-stent lumen diameter of proximal, mid and distal portion was measured. Difference of CT values between in-stent lumen and ascending aorta was used to describe as blooming artifact, and noise of in-stent lumen as image noise. Then Likert 5-point scale was performed to evaluate images noise, enhancement of in-stent lumen, blooming artifact and diagnostic confidence. Differences of objective and subjective parameters among conventional and various monochromatic images were compared by Friedman test.

In the diameter measurement of the proximal, middle and distal segments of the stent, the difference between the images of each group was statistically significant (χ² = 427.270, 426.375, 400.981, P< 0.001). The diameter of the lumen measured by 120-200 keV single-level image was larger than that measured by 60-100 keV single-level image, and the difference was statistically significant (P< 0.05). In the comparison of CT difference between the stent lumen and ascending aorta, the difference between the images of each group was statistically significant (χ² = 242.193, P< 0.001), and 100-200 keV single-level images were lower than the conventional images, the difference was statistically significant (P< 0.05). In the comparison of noise values, the difference between the images of each group was statistically significant (χ² = 420.161, P< 0.001), and the difference was statistically significant (P< 0.05). In the subjective scores of noise, enhancement, halo artifact and diagnostic confidence, there were statistically significant differences among the groups (χ²= 333.827, 455.989, 276.824, 399.497,P< 0.001). The noise score of 100-200 keV single-level image was higher than that of conventional image, the difference was statistically significant (P< 0.05). The enhancement score of 60 keV was significantly higher than that of other images (P< 0.05). The halo artifact score of 100-200 keV single level image was higher than that of 60-90 keV image, the difference was statistically significant (P<0.05). The scores of 90-120 keV single-level images were higher than those of other single-level images, and the difference was statistically significant (P< 0.05).

CCTA examinations can be effectively performed by DLCT in patients after PCI in clinical settings, and 120 keV is recommended as the optimal monochromatic image for the observation of in-stent lumen.

Tomography， X-ray computed; Coronary vessels; Stents