Clinical Values of Dose Reduction Techniques in Computed Tomography (CT) Coronary Artery Imaging With 64-row Multi-Slice Helical CT

Five hundred patients will be recruited over one and half years from our institution and other hospitals for suspected coronary artery disease. All patients will give informed consent and our Institutional Review Board approved the study. To obtain relatively consistent image noise, patients will be limited to body height and weight of 160 cm to 175 cm,and 55 kg to 75 kg, respectively. Patients with acute heart failure, severe arrhythmia and severe renal dysfunction will be excluded from the cardiac studies.

All patients underwent CT coronary artery scans using a 64-row multi-detector CT (MDCT) scanner (Light Speed VCT, GE Healthcare)in supine position. The five hundred patients were divided into conventional (C)and low dose (L) scan groups with 250 patients in each group. The patient distributions in terms of height, weight, body mass index (BMI)and heart rate were similar for the two groups with no statistically significant difference. Patients will be given bocker of about 25-50 mg orally when necessary one-hour before the cardiac scans to control heart rates and reduce their variations. Patient heart rates during CT scans were 43 to 81. A single scout scan shall be performed on every patient to cover the whole heart area, which is about 12cm. CT calcification scoring scan was performed for every patient before the CT angiography (CTA) according to the standard protocols at our institution. Timing bolus scans were performed before CTA to determine the scan start delay time: A total of 15 ml contrast agent was injected intravenously. Cine scans were used to monitor the contrast enhancement in the aorta. An ROI was placed on the ascending aorta and contrast enhancement curve as function of time was measured to determine the blood circulation time between vein of the elbow and aorta. This time was then used to determine the optimal scan delay time for CTA. The retrospective ECG-gated helical CTA scan will be then performed with predetermined scan delay time and with minimum scan range to just cover the whole heart. 70 ml contrast agent will be injected at 4.5 ml/s injection rate intravenously followed by 30ml of saline at the same injection rate.Standard scan technique shall be used for the CTA scans for the C group including: 120kV,640mA,0.35s rotation speed, 0.625 mm slice thickness, body bowtie, and 0.20-0.26 pitch selected automatically by the scanner based on the patient heart rates. For patients in L group combinations of dose reduction techniques will be applied including the use of application-specific cardiac bowtie,and adaptive post-processing filter C2,and ECG mA modulation with full tube current of 450mA for cardiac phases of 40-80%and 90mA for the rest of the cardiac phases. Cardiac bowtie provides more x-rays near the center and less x-rays towards the edge compared to a typical body bowtie to optimize x-ray distribution for cardiac applications. Our previous studies have demonstrated its advantages of image noise reduction for the heart as well as overall dose (CTDI)reduction. The lower tube current selection for the patients in L group shall be based on previous studies which indicated that the noise reduction advantage of the adaptive post-processing filter C2 might be converted into the use of lower mA to achieve similar image noise as with higher mA in clinical applications.

All images will be reconstructed with standard segment reconstruction algorithm. For the L group with lower tube current, we will apply the adaptive post-processing filter C2 to reduce image noise. Two experienced radiologists quantitatively evaluated the image quality blindly in terms of the number of coronary arteries and smaller branches identified, the sharpness of vessel edges and contrast agent clarity in the vessels using scores of 1-4 with 4 being the highest. Score 4 would indicate that at least 13 coronary artery segments could be identified,and up to the 3rd branch. Also, all identified vessels have sharp edge and clear delineations of contrast agent in the vessels. If any one of the above parameters is not satisfactory, then the quality score would be 3 and 2 if any two parameters are not as good. Image noises will be measured for the two sets with ROI placed at the center of the ascending aorta in the same place for both groups. CTDI values will be recorded for each group and converted into effective dose (ED) using European DLP method. We shall perform statistical analyses on the quality scores,the noise measurements and the effective dose measurements from the two sets using t-test with the spss10.0 software.