Varying CT Systems, Varying Risk Classification

September 3, 2014

New CT scanners’ variability affects cardiovascular risk classification.

Patients with intermediate-risk cardiovascular disease who undergo coronary artery calcification scoring with state-of-the-art computed tomography (CT) scanners may be inappropriately reclassified into high- or low-risk disease categories because of the high variability of Agatston scores produced by these new systems, a new study has found.

“These differences in coronary calcium scores result in modest reclassification in risk categories, and hence, treatment strategies for 0.5%-6.5% of individuals from the general population who were determined previously to be at intermediate cardiovascular risk,” wrote Martin J. Willemink, MD, from the department of radiology and Utrecht University Medical Center, and colleagues in Radiology. “Depending on the vendor, the use of a different state-of-the-art CT system can result in modest cardiovascular risk reclassification, which can result in substantial absolute numbers with adherence to current guidelines.”

Current standards of care call for patients with cardiovascular disease classified as intermediate risk to undergo further evaluation with CT to assess coronary artery calcification. This assessment can help to determine if the patient is at risk for future cardiovascular events. The American Heart Association currently recommends that intermediate-risk or low-to-intermediate risk patients undergo scanning for measurement of coronary calcium expressed as Agatston scores.

Although a consensus standard currently exists for Agatston scoring, the introduction of new CT systems and software may have affected the standardization of these scores. In their study, Willemink and colleagues assessed state-of-the-art CT systems from four vendors to determine intervendor variability.

“First, we analyzed the differences in Agatston scores by scanning 15 human cadaveric hearts in a realistic human chest phantom with the state-of-the-art CT scanners from the four major vendors,” the researchers wrote. “On the basis of these results we simulated the effects of Agatston score difference on risk reclassification of 432 participants at intermediate risk from the Rotterdam Study.”

Fourteen of the 15 cadaveric hearts had coronary calcification. There was a significant different in Agatston scores found between the different CT systems tested (P<.05). The median Agatston scores found were:

• 353 for Phillips

• 410 for Toshiba

• 469 for GE

• 332 for Siemens

Looking at the data from the 432 asymptomatic people from the Rotterdam study, 45.4 percent were reclassified as low-risk as a result of their Agatston score, and 16.2 percent were reclassified as high risk. Using the Siemens scanner as a reference the largest difference in reclassification was 6.5 percent found with the GE scanner.

“This means that up to 6.5% of individuals at intermediate risk would be reclassified differently because of the use of a system from a different CT vendor,” the researchers wrote.

In the original study, 196 individuals were classified as low-risk compared with 205 using the reclassification with the reference scanner. Similarly, only 70 individuals were reclassified as high risk compared with 84 using the reference scanner.

“Although the number of reclassifications is modest, given the large population at intermediate risk, the absolute number can be substantial,” the researchers wrote.