Coronary artery calcification (CAC) quantitation is an emerging technology that has come into widespread clinical use within the past several years. It can now be performed accurately on modified conventional multidetector computed tomographic (MDCT) scanners, which are widely available. In the past, CAC was performed only on electron beam computed tomography (EBCT) scanners, which are not widely available. Coronary computed tomographic arteriography (CTA) is an experimental technique that is not yet in clinical use. This technique uses the latest generation of spiral CT technology and MDCT scanners to noninvasively image the coronary arteries. Both techniques are important noninvasive imaging methods for the identification and stratification of patients with coronary atherosclerosis and have great potential for widespread use in the next several years. The scientific session on coronary artery calcification that discussed these techniques was moderated by William Stanford, MD, of the University of Iowa Hospital and Clinics, lowa City, and Christoph R. Becker, MD, of Neuried, Germany. MDCT vs Sequential CT Techniques Sandra S. Halliburton, PhD, from the Cleveland Clinic Foundation, Cleveland, Ohio,[1] presented the initial paper, which compared coronary calcium scores from MDCT using a spiral imaging technique to a more commonly performed and validated sequential CT technique. Patients were examined with a MDCT scanner using prospective electrocardiograph (ECG)-triggered sequential modes (4 mm by 2.5 mm collimation; 2.5-mm thickness) and retrospective ECG-gated spiral modes (4 mm by 2.5 mm collimation; 3-mm thickness with no slice overlap). Data analysis revealed that the sequential method tended to yield higher scores than the spiral method. There was a significant difference (P < .001) between the mean total Agatston and volume scores obtained with each technique. Of significant concern, 8 of the 40 patients had some calcium depicted on sequential images, whereas no calcium was depicted on spiral images. Because the CAC score varied, the resulting percentile clinical risk rankings also differed in 34 patients. Eleven patients had differences greater than 20%. The researchers concluded that the MDCT acquisition technique has a significant impact on coronary calcium scoring and the resulting risk stratification. Use of EBCT and Single-Slice Helical CT in Calcium Scoring Kishore C. Acharya, PhD, of GE Medical Systems, Milwaukee, Wisconsin,[2] presented the results of a 3-center study, which compared calcium scores obtained with both EBCT and single-slice helical CT (HCT). The HCT studies were performed on a single-slice HCT system capable of subsecond scanning with retrospective ECG gating. Conventional Agatston Janowitz scoring was used for EBCT images, but a modification using a minimum lesion size of 0.25 mm2 was used for helical scans. Results revealed that the mean, median, and standard deviation values were similar for both EBCT and HCT. Although the quantitative data were promising, when the individual patients were stratified to assess for risk of coronary artery disease (CAD). there were substantial differences. All mismatches were within 1 category, and most were underscored by HCT compared with EBCT. The researchers stated that their results improved significantly if the minimum Hounsfield unit (HU) density was lowered for the HCT technique, which resulted in higher Agatston Janowitz scores that agreed better with the EBCT. Coronary Calcium Scoring With EBCT and ECG-Gated MDCT B. M. Ohnesorge, PhD, from Munich University, Germany, and Siemens Medical Systems, Inc,[3] gave the first of 4 presentations from his institution in this session. The study focused on the interexamination reproducibility of calcium scores when compared with EBCT. Patients were examined twice with MDCT. Data from a separate group of patients who underwent scanning earlier with EBCT were used for comparison. Ohnesorge and colleagues compared the variability of EBCT and MDCT with 3-mm, nonoverlapping increments and evaluated the variability of MDCT with overlapping increments (2, 1.5, and l mm). Results revealed that MDCT with nonoverlapping increment showed lower variability than EBCT. The variability of volume scoring with MDCT could be reduced with overlapping increments. They concluded that MDCT can provide continuous volume image data for volumetric calcium scoring with higher reproducibility than EBCT. The use of image data with overlapping incremental results in significantly improved reproducibility compared with nonoverlapping sequential data. Significance of Age and Sex in CAC Progression An interesting evaluation of the influence of sex and age on the rate of progression of CAC scores was presented by Aletha M. Emerick, BS, of the UCLA School of Medicine Center for Health Science, Los Angeles, California.[4] This study compared data from 238 asymptomatic patients who had undergone 2 or more EBCT examinations. The mean interscan interval was 25.9 months. The mean change in CAC score between the first and last scans was 50.7, and the mean rate of change was 2.0 units per month. Men had a much higher rate of change of CAC (2.7 for men, 1.3 for women). When grouped by age, only women in the 40- to 49-year-old age group had a significantly different rate of change in their CAC scores (0.2) than men (0.7) (P = .04). Emerick and coworkers concluded that the difference in rates of change in CAC score between sexes was not significant for subjects age 50 years or older. However, in the 40- to 49-year-old age group, the monthly rate of progression was lower in women than men. They also found that the only risk factor for prediction of rate of change of CAC was the CAC determined from the first EBCT study. Multisector Reconstruction vs Single-Sector Reconstruction in Calcium Scoring A comparison of calcium scores from a MDCT reconstructed using both multisector reconstruction (MSR) and single-sector reconstruction (SSR) algorithms was presented by Curtis H. Coulam, MD, of Stanford University/Lucas MRS Imaging Center, Palo Alto, California.[5] The multisector algorithm is a novel approach that uses only small-sector angle of projection views collected during several cardiac cycles, which effectively reduces the temporal resolution to 133 msec (compared with 533 msec for the SSR technique). Coulam and associates determined that coronary blurring was more evident on the SSR images compared with MSR. The calcium score averaged 15% less with MSR compared with SSR (P < .05). They concluded that MSR results in significantly lower calcium scores, improved temporal resolution, and less coronary motion blurring compared with SSR. Optimizing Coronary Visualization With Multislice CT Andreas F. Kopp, MD, from the University of Tuebingen, Tuebingen, Germany, presented his first of 3 papers on coronary CT angiography.[6] The initial paper evaluated the use of multiple differing reconstruction time points within the RR interval. Image quality for depiction of each of the 3 main coronary arteries was evaluated at multiple time points to determine if each should be reconstructed at specific points, which may differ from vessel to vessel. Image quality was determined by review of 3 independent readers. They found that mid-to-late diastole is the best time point for visualizing the left anterior descending (LAD) coronary artery and circumflex artery (RCX). The right coronary artery (RCA) is best visualized when reconstructed in early diastole. It was concluded that an optimization of the time point for reconstruction of ECG-gated multislice coronary CT angiography is mandatory to obtain optimal image quality. Atherosclerotic Plaque Morphology by MDCT Compared With Intracoronary Ultrasound Dr. Kopp’s next presentation focused on another study by his group, which evaluated atherosclerotic plaque morphology by MDCT and compared the results to the “gold standard” intracoronary ultrasound (ICUS).[7] The identification of soft plaque and stratification of its potential to rupture and cause a coronary thrombosis is a goal that will have significant impact on the diagnosis and treatment of coronary artery disease. This paper represents a major step in obtaining that goal. ICUS and contrast-enhanced MDCT scans were analyzed in patients scheduled for ICUS-guided angioplasty. One plaque was selected in each patient, and plaque composition was defined as soft, intermediate, or calcified according to established ICUS criteria (echogenicity) and by using HU density measurements on MDCT. The MDCT and ICUS yielded nearly identical results regarding plaque composition. Some plaques were only detected on CT after reviewing the ICUS data. Nevertheless, this is an exciting result that will need to be investigated further. Potential of ECG-Gated Multislice Spiral CT in Visualizing Noncalcified Atherosclerotic Plaques Dr. Kopp’s final presentation was that of a phantom study performed to evaluate the limitations of MDCT’s ability to differentiate among lipid-rich, fibrous, and calcified coronary plaques.[8] The study showed that noncalcified plaques could be characterized by their mean HU values. Kopp and coworkers determined that lipid-rich plaques could be identified with mean HU values in the interval of -60 to 50 HU, noncalcified fibrous plaques in the interval of 50 to 120 HU, and calcified plaques at greater than 120 HU. Assessment of Inflammation and Atherosclerotic Plaque Volume in the Coronary Arteries Guy J. Friedrich, MD, of University Hospital, Innsbruck, Austria, investigated a possible link between the progression of atherosclerosis and the presence of an active inflammatory process.[9] Friedrich and colleagues correlated the volume of calcified plaque determined by EBCT with the serum level C-reactive protein (CRP), a marker of active inflammation. Patients with symptomatic coronary artery disease underwent EBCT imaging and had CRP levels measured. All patients had an elevated coronary calcium score and 29.5% had an increase of CRP values. However, there was no significant correlation between the level of CRP or the CAC and clinical symptoms. Detecting High-Grade Coronary Artery Stenoses Using Contrast-Enhanced MDCT Toru Sakuma, MD, of Jikei University School of Medicine, Tokyo, Japan,[10] evaluated the detectability of the high-grade coronary artery stenoses using retrospectively ECG-gated contrast-enhanced MDCT. Results were compared to conventional coronary angiography (CCA). Sakuma and associates used a retrospectively ECG-gated reconstruction technique with a temporal resolution of either 125 or 250 msec per image and reconstruction thickness of 1.25 mm. Image analysis was compared with CCA as the “gold standard.” Coronary artery stenoses were present in 19 of 96 anatomic segments using CCA. Fourteen of the 19 stenoses were depicted on MDCT. MDCT also detected 3 false-positive stenoses. The sensitivity of MDCT for detecting high-grade stenoses was 73.7%, and the specificity was 96.1%.