Incidental CAC Tied to Worse Outcomes in Systemic Inflammatory Diseases

WASHINGTON, DC—Incidental coronary artery calcium (CAC) is often seen on chest CTs in patients with immune-mediated inflammatory diseases who don’t have atherosclerotic cardiovascular disease (ASCVD), with the finding indicating a greater risk of poor outcomes, researchers report.

An artificial intelligence (AI) algorithm picked up CAC in more than half of patients with psoriasis, systemic lupus erythematosus (SLE), or rheumatoid arthritis who had undergone a prior noncardiac chest CT scan, Brittany Weber, MD, PhD (Brigham and Women’s Hospital and Harvard Medical School, Boston, MA), told attendees here at the 2024 Society of Cardiovascular Computed Tomography (SCCT) meeting.

Any amount of calcium was associated with greater risks of all-cause death and MACE (death, MI, stroke, or revascularization), with the relationships strengthening at higher CAC scores. In fact, Weber said, patients with a CAC-AI score of 100 or more carried a CV risk similar to what has been seen in secondary prevention cohorts.

“These data demonstrate that AI-based calcium scoring could offer additional prognostic information, which could better guide the initiation or the intensification of our prevention therapies,” she said. She noted that ASCVD risk scores can underestimate risk in patients with systemic inflammatory diseases, a population in whom chest CTs are commonly ordered for noncardiac indications.

Leveraging AI

Prior studies have shown that patients with systemic immune-mediated inflammatory diseases have greater risks of CVD, Weber said, noting that the subspecialty of cardio-rheumatology is emerging to improve recognition of heart disease in these patients. There are no clear guidelines on how to assess cardiovascular risks in these patients, however. “This requires multidisciplinary care models and that’s just, unfortunately, not available at the majority of academic medical centers—hence the need for innovative solutions,” Weber said.

The detection of coronary plaque can enhance CVD prevention, and this can be done in multiple ways, she said, pointing to coronary CT angiography, calcium scoring, or the evaluation of prior chest CTs obtained for other indications.

Ultimately what we’re all trying to achieve here is prevention. Brittany Weber

The current study explored that last option, with investigators using a US Food and Drug Administration-cleared AI algorithm from Nanox to look for CAC on previously obtained noncardiac, nongated chest CTs from 2,528 patients (median age 59 years; 66.6% women) with psoriasis, SLE, or rheumatoid arthritis and no known ASCVD who were scanned between 2010 and 2020.

In this cohort, the most common CV risk factors were hypertension (32%), hyperlipidemia (21%), and active tobacco use (17%). The heaviest risk factor burden was in patients with psoriasis, though those with SLE had the highest rate of chronic kidney disease. Lipid profiles tended to be better in patients with SLE versus the other two inflammatory conditions.

Using the AI algorithm, the researchers found that more than half of patients had some degree of CAC, with only 46.6% having a CAC-AI score of 0. Scores of 1-99, 100-299, and 300 or more were found in 30.3%, 10.1%, and 13.0%, respectively. CAC severity was greater in patients with psoriasis than in those with the other two conditions.

Through a median follow-up of 7.4 years, the presence of CAC was associated with adverse outcomes. Risks were greater with increasing levels of CAC, but the increases were significant even for those with CAC-AI scores of 1-99: HR 1.4 (95% CI 1.1-1.9) for all-cause mortality and HR 1.5 (95% CI 1.2-1.9) for MACE. The HRs for both of those outcomes were 2.5 in patients with a CAC-AI score of 300 or higher.

Unanswered Questions

Commenting for TCTMD, Jonathan Weir-McCall, MBChB, PhD (University of Cambridge, England), pointed to several open questions. He said that in a future study he’d like to see a control group of patients without systemic inflammatory diseases, which would allow for a better assessment of the impact of incidental CAC in these patients versus the general population.

In addition, a real difference in the risk of adverse outcomes between the two highest categories of CAC-AI score did not emerge until after 10 years, at which point patient numbers were very small, he said. And that’s difficult to explain, he added.

Finally, Weir-McCall said, the interplay between immune-mediated inflammatory diseases, standard CV prevention therapies, CAC, and risk of adverse outcomes remains unclear.

The study “speaks really well to AI-automated extraction of this information from scans and hopefully using it to better guide patient care,” he said, noting that all the scans in this study were already performed for other indications.

But there’s a need to understand how therapies like statins interact with immune-mediated inflammatory diseases in terms of reduced CV risk, according to Weir-McCall. “Because of course it’s inflammatory-driven, so will our standard preventative therapies work or do we need to look at alternate approaches to mitigate this risk?”

Weber said in the future she hopes to do a prospective randomized clinical trial to understand the impact of notifying patients with immune-mediated inflammatory diseases and the clinicians—often rheumatologists—caring for them about incidental CAC findings, including any effects on recommendations for primary prevention therapies.

“Because ultimately what we’re all trying to achieve here is prevention,” Weber stressed.