CAC Tests Boost CAD Prediction in Younger Patients at Risk for ASCVD

Preexisting risk factors, like diabetes, might justify earlier CAC tests, but whether that can improve outcomes remains unclear.

CAC Tests Boost CAD Prediction in Younger Patients at Risk for ASCVD

Results of a coronary artery calcium (CAC) scan can augment risk prediction in younger adults with atherosclerotic cardiovascular disease (ASCVD) risk factors, especially diabetes, and support the use of the test in certain groups of patients younger than those currently recommended in the guidelines, authors of a new study say.

Based on the clinical risk equations intended to estimate the probability of a CAC score > 0, investigators conclude that men and women with diabetes, for example, could undergo CAC scans at ages 37 and 50, respectively, as compared with ages 42 and 58 in those without. They say their analyses help to identify groups of patients at especially high risk for premature coronary disease in whom CT tests might glean additional information.

Just how much ordering a test would prove useful, however, beyond targeting those preexisting risk factors—and whether doing so actually leads to less disease—remains unclear.

The most recent American College of Cardiology/American Heart Association guidelines give a class IIa indication for CAC scoring to help drive treatment decisions in patients at intermediate risk for developing ASCVD, typically focused on adults 40 to 75 years old. But most of the data looking at CAC scoring has been in older patients. Recent findings, however, have demonstrated that any CAC in those under 45 with chest pain is a concern. And since existing ASCVD risk equations depend heavily on age, the value of CAC tests in younger adults remains unclear.

The current study zeros in on younger adults with existing ASCVD risk factors, which can help “inform the expected prevalence of CAC,” lead author Omar Dzaye, MD, PhD (Johns Hopkins University School of Medicine, Baltimore, MD), explained to TCTMD. “This can be used to determine an appropriate age to consider the first CAC testing to especially identify those individuals most susceptible to very early premature atherosclerosis.”

Commenting on the study for TCTMD, Giuseppe Biondi-Zoccai, MD (Sapienza University of Rome, Italy), agreed that the age suggestions proposed in the study seem “reasonable,” but he argued that the study’s overall contribution is in highlighting the importance of primary prevention, not rationalizing the use of CAC. “Even if you don't perform this test, the need to address the risk factors is there,” he said, adding that many of his European colleagues don’t feel as strongly about performing CAC scans as some do in the United States. “This study reinforces the need for aggressive primary prevention. That's the main take-home message.”

Risk Factors and CAC

For the study, published in the October 19, 2021, issue of the Journal of the American College of Cardiology, Dzaye and colleagues included more than 22,000 adults aged 30-50 years (mean age 43.5 years; 25% women) from the CAC Consortium who were referred for CAC scanning at one of four US institutions between 1991 and 2010. Just over one-third had scores of greater than zero, with a median score of 20.

All but 7.3% of participants had a low 10-year risk using the pooled cohort equations, but 49.6% had dyslipidemia and 49.3% had a family history of CHD. Hypertension was reported in 20.1% and diabetes in 4.0%. Eleven percent were active smokers.

Median and mean CAC scores consistently increased with age, with a steeper curve observed for men than for women. As such, men and women without traditional risk factors in the cohort were expected to develop CAC at ages 42.3 and 57.6, respectively. Diabetes, however, had the strongest single influence on CAC development, lowering this threshold by an average of 6.4 years to ages 36.8 for men and 50.3 for women. Family history of CHD, smoking, dyslipidemia, and hypertension each resulted in earlier CAC appearance by 3.7, 3.3, 4.3, and 3.7 years, respectively. In combination with family history, dyslipidemia was linked to earlier CAC by 7.9 and hypertension by 7.3 years.

Dzaye said the results were internally and externally validated, although the latter remain unpublished.

He also acknowledged the limitations of the study: the cohort included only one-quarter women and 12.3% nonwhite participants, and half had family histories of CHD. “Therefore this data set does not represent the general population,” he said. However, “coincidentally” the ages their study suggests for CAC score referral are in line with what many of the authors had already been doing in clinical practice, Dzaye added.

Critically, there are no longitudinal data, the authors acknowledge: “Future prospective studies with multiple CAC scans that begin in young adulthood and in more ethnically diverse populations will be required for reproducibility.”

Limited Generalizability

In an accompanying editorial, Tasneem Naqvi, MD (Mayo Clinic, Scottsdale, AZ), and Tamar S. Polonsky, MD (University of Chicago, IL), write that “the study has filled an important clinical gap, providing highly actionable data that could help guide clinical decision-making for ASCVD prevention.” However, they say the limitations regarding the study cohort itself as well as the fact that participants only had a single CAC scan do limit its generalizability.

Biondi-Zoccai had additional concerns, noting that patient management has changed since the end of the study period, risk factor data were missing for about 20% of the cohort, and the study does not delve into the severity or control of individual risk factors. “They label patients has having a risk factor, but we don't know whether that risk factor was managed in that patient,” he said. “So, for instance, you could have diabetes uncontrolled or diabetes well controlled.”

Ultrasound, which results in less radiation and is more often used in Europe, may be a good tool for identifying CAC in younger patients, especially because it is easily repeatable, according to the editorialists. However, Dzaye argued that minimal but clinically relevant CAC scores below 10 “might be under the detection level of ultrasound.” Different populations might be best targeted by one or the other, but “I believe a mix of ultrasound and CAC [scoring] might be the solution,” he added.

Biondi-Zoccai, on the other hand, said he and his colleagues often perform ultrasound to look at CAC, and that a CAC scan is not the only way to gauge risk, noting that many of the authors are “very well-established” proponents of CAC testing. This study, by focusing squarely on the added value of CAC scans in this group, doesn’t incorporate the full spectrum of options, he asserted.  

“In a way, it's like I want to show you how to prepare a pizza and you need to learn how to prepare a pizza and then you'll be able to eat a very good pizza,” he said. “But I'm limiting you in your choice of food, because maybe you don't want to eat pizza.”

What’s also missing from the findings is a link between CAC scores and clinical events, Biondi-Zoccai continued, and whether taking the added step of getting CAC scans in these patients leads to meaningful differences in events down the road. “These are not patients followed over time,” he said. “In a way, it's reinforcement that we need to perform coronary artery calcium scores, but this is not necessarily agreed upon by everybody.”

Still, for a physician to be able to tell a patient that smoking will age their hearts by a specific number of years, so they anticipate the atherosclerosis, is “going to be very motivating to adopt treatments and also to maintain adherence,” he said.

Going forward, Biondi-Zoccai said he “would like to see how we can change trajectories by performing the coronary artery calcium scoring by managing these patients more aggressively [plus] see confirmation that these temporal changes, this earlier atherosclerotic labeling, is also associated with early events. Because this is the main rationale, but it's not necessarily established."

  • Dzaye, Biondi-Zoccai, Naqvi, and Polonsky report no relevant conflicts of interest.