|
|
|
| Selecting optimal monochromatic level of spectral CT imaging with low concentration contrast media for improving imaging quality of hepatic veins |
| TONG Zi-bin, LIU Ai-lian, LIU Yi-jun, TIAN Shi-feng, ZHANG Long-min, LIU Jing-hong, PU Ren-wang |
| Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian Liaoning 116000, China |
|
|
|
|
Abstract Objective: To investigate the effect of spectral CT monochromatic images with low concentration contrast media for improving imaging quality of the hepatic veins. Methods: One hundred and fifty-one patients from March 2013 to January 2014 in our hospital underwent spectral CT examination of the full abdomen. According to the first scan time, patients(BMI＞25 kg/m2) were randomly divided into group A with spectral CT optimal monochromatic imaging and low concentration contrast media(270 mgI/mL) and group B with conventional imaging and high concentration contrast media(350 mgI/mL). During portal phase, conventional images and optimal monochromatic images with slice thickness of 1.25 mm were obtained, and volume rendering(VR) and maximum intensity projection(MIP) were created to show the hepatic venography. Independent t-test was used to compare the hepatic venography. And the liver parenchyma, image noise, vein-to-liver contrast-to-noise ratio(CNR) between the two image data sets were compared. The overall imaging quality was evaluated on a five-point scale by two radiologists. Inter-observer agreement in subjective image quality grading was assessed by Kapppa statistics. Mann-Whitney U test was used to compare the image quality score of the hepatic venography between the two groups. The CTDIvol of the two groups were recorded, and the difference between group A and group B was compared by independent t-test. The total amount of iodine was calculated and compared. Result: The monochromatic images at 49～51 keV were found to demonstrate the optimal CNR for the hepatic venography. In group A, CT values of the left hepatic vein, the middle hepatic vein, the right hepatic vein, liver parenchyma, the difference between hepatic veins and liver parenchyma, and the image noise were (335.29±68.45) HU, (343.16±73.95) HU, (327.80±76.22) HU, (187.74±31.98) HU, (147.39±48.91) HU, (155.26±56.68) HU, (139.90±55.19) HU and (39.12±10.04) HU, respectively. Those in group B were (194.82±33.60) HU, (196.62±36.11) HU, (190.60±33.61) HU, (115.89±16.94) HU, (78.93±27.92) HU, (80.73±28.44) HU, (74.71±26.03) HU and (33.82±8.94) HU, respectively. The vein-to-liver CNRs for group A and group B were 3.93±1.80, 4.35±2.29, 3.91±2.08, 2.40±0.85, 2.44±0.82 and 2.26±0.75, respectively. The data of group A were all significantly higher than group B(P＜0.05). Kappa statistics demonstrated a good inter-observer agreement(Kappa values＞0.8). The image quality scores of the hepatic venography of the two groups were 3.43±0.58 and 3.48±0.68. There was no significant difference between the two groups. The CTDIvol of the two groups were (16.87±0.00) mGy and (15.74±5.26) mGy, respectively, and there was no significant difference between the two groups. The total amount of iodine of the two groups were 27 000 mg and 35 000 mg. Group A was 22.86% lower than group B. Conclusion: For patients with BMI＞25 kg/m2, in the premise of reducing the concentration of contrast agent, spectral CT monochromatic imaging can improve the imaging quality of the hepatic veins without significant increase of radiation exposure.
|
|
Received: 12 August 2014
|
|
|
|
Corresponding Authors:
LIU Ai-lian
|
|
|
|
|
|
|
2015, Vol. 26 
