Coronary Artery Calcium Scores Vary Widely Among Scanners

Kate Johnson

March 14, 2014

VIENNA — Up to 6.5% of people classified as being at intermediate cardiovascular risk could be misclassified, depending on which brand of CT scanner was used to calculate their coronary artery calcium score, a new study suggests.

"We found that calcium scores differed substantially among state-of-the-art CT systems from 4 major vendors. Subsequently, simulation showed that 0.5% to 6.5% of individuals were reclassified," said researcher Martin Willemink, MD, from University Medical Center Utrecht in the Netherlands.

Although a 6.5% reclassification rate might seem modest, "the absolute number of calcium-scoring CT scans is very large, especially since American Heart Association guidelines recommend calcium scoring for asymptomatic adults at low–intermediate and intermediate risk," Dr. Willemink told Medscape Medical News. This comprises approximately 40% of the population in the United States.

He presented the findings here at the European Congress of Radiology 2014.

The research team evaluated images from 15 ex vivo human hearts scanned with CT scanners from 4 vendors: GE Healthcare, Philips, Siemens, and Toshiba.

For each heart, an Agatston score was calculated using the software of the respective vendor. Calcium scoring protocols and radiation dose levels were similar for the 4 scanners.

"We found very large differences in Agatston scores for the same heart on different scanners," he reported.

We found that calcium scores differed substantially among state-of-the-art CT systems from 4 major vendors.

The researchers then used the ex vivo Agatston scores to simulate the effects of the different scanners on 432 older people participating in the Rotterdam Study, a prospective population-based study.

All subjects had known Agatston scores and were classified as being at intermediate risk, Dr. Willemink explained. Linear regression analysis was used to obtain a conversion factor to compare scores from each scanner.

The largest variation in calcium score was between Siemens and GE Healthcare scanners, where median scores ranged from 332 to 469 (P < .05).

Dr. Willemink reported that "14 individuals who were initially classified by Siemens as low risk were reclassified as intermediate risk by another vendor, and another 14 were reclassified from intermediate to high risk."

Table. Risk Classification of Patients by Scanner

Risk Category Siemens, n GE Healthcare, n
Low (calcium score <50) 196 182
High (calcium score >650) 70 84


"Theoretically, a subject could be classified as low risk at hospital A and not receive any therapy, and as high risk at hospital B and receive therapy," he said. "The calcium scoring scan protocols should be adjusted so that different CT scanners result in similar calcium scores."

The impact of these findings on individuals is potentially "dramatic," noted Matthew Budoff, MD, professor of medicine at the Los Angeles Biomedical Research Institute and one of the authors of the 2010 AHA/ACCF Guideline for Assessment of Cardiovascular Risk in Asymptomatic Adults (J Am Coll Cardiol. 2010;56:e50-e103).

"I think the study presented is valid," he told Medscape Medical News. However, it was primarily based on scans of ex vivo hearts and should be confirmed in live subjects. "We need in vivo data to prove there is a problem," he said.

In addition, "the motion of the coronary arteries changes with heart rate changes; this was not taken into account because the hearts were not moving. With coronary motion in vivo, this may be an even bigger problem. It certainly suggests that, until more research is done, patients getting follow-up scans should get the scan done on similar equipment."

Dr. Budoff was involved in a previous study of nearly 100 patients that found "very high concordance" among 3 scanners (J Comput Assist Tomogr. 2009;33:175-178). "Perhaps the newer scanners are not being well calibrated for coronary artery calcium scanning. There may be too much focus on new adaptations with CT scanners and the vendors might not be careful enough with calcium score results," he said.

In fact, "this study has limited relevance in the current environment," said Donald Lloyd-Jones, MD, professor of preventive medicine at the Northwestern University Feinberg School of Medicine in Chicago, who was speaking on behalf of the American Heart Association.

"Recent evidence-based recommendations from the American Heart Association/American College of Cardiology and from the Joint British Societies advise consideration of statin therapy for individuals at risk levels well below those evaluated by Dr. Willemink's team," Dr. Lloyd-Jones told Medscape Medical News. Therefore, "the issue of reclassification for the old intermediate-risk group is largely moot."

Dr. Lloyd-Jones is a coauthor of the 2013 American College of Cardiology/American Heart Association Guideline on the Assessment of Cardiovascular Risk (J Am Coll Cardiol. Published online November 12, 2013).

But Dr. Budoff said he disagrees with Dr. Lloyd-Jones. "The new guidelines advocate for calcium scoring when risk ascertainment is uncertain with Framingham risk."

"Whatever the risk cutpoint, we need a consistent result," he said.

Dr. Willemink and Dr. Lloyd-Jones have disclosed no relevant financial relationships. Dr. Budoff is a consultant for GE Healthcare.

European Congress of Radiology (ECR) 2014: Abstract B-0601. Presented March 8, 2014.


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.