A Single CT Scan Reveals Calcium Score and Coronary Anatomy

Reed Miller

April 12, 2011

April 11, 2011 (New Orleans, Louisiana) — Coronary calcium can be accurately quantified with a single standard-contrast computed tomography (CT) scan, thereby eliminating the added radiation exposure of a follow-up noncontrast scan, according to research presented at the American College of Cardiology 2011 Scientific Sessions last week [1].

Dr James Otton (St Vincent's Hospital, Sydney, Australia) and colleagues presented results from their Prospective Evaluation of an Algorithm for Coronary Calcium Estimation, a study of their algorithm for quantifying extraluminal coronary calcium in the major epicardial vessels from a contrast-enhanced CT scan.

Measuring calcium score from a single contrast CT has been tried before, but it has usually been impossible to accurately measure calcium across the whole range of Agatston scores, Otton told heartwire ."The key to [our] success was realizing, first, that it is not possible to adequately separate contrast from arterial-wall calcium using solely the traditional axial measurement approach, and second, that traditional calcium-score thresholds don't apply to high-resolution contrast-enhanced coronary CT."

Instead, Otton et al tested coaxial analysis--looking at the coronary artery in short axis--which allows for precise separation of the contrast-filled lumen and the vessel wall and can be mostly automated with standard software tools, he explained. The researchers also found that a lot of noncalcified material appears in the Hounsfield range of 130 to 200 Hounsfield units (HU) on high-resolution coronary CT scans, while the true calcium generally greatly exceeds 200 HU. This finding is supported by previous intravascular ultrasound (IVUS) studies.

Calcium-containing arterial segments are selected and the automatic contours inspected in arterial short axis. After any necessary adjustments, the calcium volume >320 HU is recorded. The calcium volume can then be converted to the Agatston score with a simple linear conversion factor. [Source: Dr James Otton]

All of the patients in the study underwent both calcium scoring and contrast scanning with a wide-volume 320-detector-row scanner using standard clinical protocols. The investigators developed an algorithm to quantify extraluminal coronary calcium in the major epicardial vessels using a semiautomated technique. The initial parameters were estimated based on 100 contrast and corresponding Agatston calcium-score measurements. Then the researchers prospectively validated their coaxial technique in a further 100 patients according to prespecified inclusion criteria.

The calcium scores measured in the study ranged from 0 to 4842, and correlation of the contrast-scan–derived calcium score with the score measured using the usual noncontrast method was extremely high (r 2=0.99). Using standard coronary calcium "bands" (0–10, 11–100,101–400, >400), they found that agreement of the virtual calcium scores derived with the new method was near perfect ({:kappa:}=0.88), as was the agreement between different "raters" who evaluated the scans. The authors hope to be able to publish the final results of the validation study soon.

The main benefit of this technique is that it exposes the patient to lower radiation doses than the standard two-scan method, currently a major concern among physicians and patients, Otton said. The expected radiation savings are between 30% and 50% of total protocol dose, and doses of less than 1 mSv are possible in selected patients.

"The method can be used when a combined calcium score and contrast coronary scan would have been performed to reduce the radiation dose, and it can be used when only a contrast-enhanced coronary CT is planned, such as for EP procedures, to provide quantitative prognostic information 'for free,' " Otton told heartwire .

The most common scenario in which Otton uses this technique is a patient undergoing contrast CT for chest-pain evaluation. "While it is possible to 'guesstimate' a person's risk from the contrast-enhanced CT, using the calcium score provides a well-validated risk quantification and allows evidence-based and rational selection of therapies and future monitoring," he said.

One current limitation of this technique is that it depends on the accuracy of the existing software tools, which sometimes require manual adjustments. "The overall process involves selecting the calcium in each artery, adjusting where necessary, noting the total volume above 320 HU, and then multiplying by the conversion factor to get the Agatston calcium score, so a few software adjustments need to be made," Otton explained. However, he expects that the process could be made to be fully automated in the near future.

Commenting on the study, Dr Matthew Budoff (Los Angeles Biomedical Research Institute, CA) suggested that Otton et al present analysis of how their method can distinguish between zero and low calcium scores.  "Given our data from the MESA study, a calcium score of 1 to 10 carries a threefold risk compared with zero. A zero score carries a very good negative predictive power for no events for the next five to 10 years," Budoff told heartwire . "They need to show the differences in patients with scores below 100 to better validate their technique."

Otton is an applicant for a patent related to computer-assisted measurement of coronary calcium deposits.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.