Miho Horie MSc, Presenter: Former Employee, Toshiba Corporation

Sam Santiago, Abstract Co-Author: Nothing to Disclose

Theresa Goode, Abstract Co-Author: Nothing to Disclose

Yasser Karimzad BSc, Abstract Co-Author: Nothing to Disclose

Narinder S. Paul MD, Abstract Co-Author: Research funding, Toshiba Corporation

Purpose: To assess the accuracy of calculating the Agatston score (AS) and assigning risk stratification for patients having CT coronary artery calcium scores (CCS) with standard dose and low dose filtered back projection (FBP) and Adaptive Iterative Dose Reduction (AIDR).

50 patients, age 54y±6, BMI 29±4 referred for CCS, were scanned twice using volume CT, prospective ECG gated acquisition and 240-320 x 0.5mm detector rows. The standard dose scan used a weight based protocol (S1, 120kV, 87.5-140mAs) and was followed by a low dose scan (S2, 120kV, 17.5-105mAs). Both image series were reconstructed with FBP and AIDR, anonymised, randomized and read by two expert readers (R1, R2) who evaluated the AS and assigned a risk stratification; AS ≤10, minimal; 11-100 mild; 101-400, moderate; and >400, high risk. A 3 point scale (1=non diagnostic, 3=excellent) was used for qualitative assessment of image quality (IQ). Pearson Correlation coefficient was computed between CCS scores.

IQ scores: S1- FBP= 2.7±0.5, AIDR= 2.8±0.4; S2- FBP= 2.0±0.7 and AIDR= 2.6±0.6. AS: S1- FBP = 154 (0-1414), AIDR = 143 (0-1296); S2 - FBP = 164 (0-1606), AIDR = 140 (0-1282). The Pearson Correlation coefficient for all the image series showed a high degree of correlation r=0.99 (p<0.001). Risk stratification: Using FBP S1 protocol, patient grouping was: minimal=28, mild=8, moderate=8 and high risk=6. AIDR S1 caused no change in risk stratification. FBP S2 protocol resulted in one patient changing category from minimal to mild risk (AS 10 to 18) and AIDR S2 protocol, resulted in one patient changing category from moderate to mild risk (AS 105 to 96). When FBP S1 AS=0, FBP S2 showed AS > 0 in 11 cases. AIDR S2 reduced the number of false positive cases from 11 to 4. Image noise (IN) for S2 scans reduced from FBP S2 IN=31.2 HU (19.3-50.5) to AIDR S2=19.4 (14.4-24.8). Mean DLP (mGy*cm) and effective radiation dose (mSv) was S1= 151.8±27.8, 2.13mSv and S2= 68.3±29.4, 0.96mSv.  

Use of AIDR in sub-mSv CCS results in a significant improvement in image noise with a 64% reduction in false positive results compared to FBP.

Robust and accurate sub-mSv CCS is possible when AIDR is utilized.

Horie, M, Santiago, S, Goode, T, Karimzad, Y, Paul, N, Utility of Iterative Reconstruction in Sub-mSv Scan Coronary Artery Calcium Score.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12043630.html