Introduction: The Hidden Cardiovascular Burden in Inflammatory Disease
For clinicians managing immune-mediated inflammatory diseases (IMIDs) such as rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and psoriatic disease, the primary focus is often on achieving remission of joint or skin symptoms. However, a parallel and often more lethal process is frequently occurring beneath the surface: accelerated atherosclerosis. It is well-established that chronic systemic inflammation drives cardiovascular (CV) disease, yet current risk stratification tools, such as the Pooled Cohort Equations, often underestimate the actual risk in these populations.
A groundbreaking study published in JACC: Cardiovascular Imaging by Weber et al. explores a novel solution to this diagnostic challenge. By leveraging artificial intelligence (AI) to analyze incidental findings on routine chest computed tomography (CT) scans, researchers have uncovered a high prevalence of coronary artery calcium (CAC) in IMID patients—a finding that carries profound prognostic weight for both major adverse cardiovascular events (MACE) and all-cause mortality.
The Clinical Context: Why Current Risk Models Fall Short
Patients with IMIDs face a cardiovascular risk profile that mirrors that of patients with diabetes mellitus. The pathophysiology is multifaceted, involving traditional risk factors (hypertension, dyslipidemia) exacerbated by chronic cytokine elevation (TNF-alpha, IL-6, IL-17), which promotes endothelial dysfunction and plaque instability.
Despite this known risk, primary prevention remains suboptimal. Many patients with RA or SLE do not meet the traditional thresholds for statin initiation based on conventional calculators, even though their biological risk is significantly higher. There is a desperate clinical need for objective biomarkers of subclinical atherosclerosis that can guide aggressive preventive therapy. Coronary artery calcium (CAC) scoring via dedicated cardiac CT is a gold-standard predictor, but it is not routinely performed in the rheumatology clinic. However, many of these patients undergo chest CTs for other reasons—such as screening for interstitial lung disease or evaluating respiratory symptoms. This presents an opportunity for “opportunistic” screening.
Study Design: Harnessing AI for Opportunistic Screening
The study conducted by Weber and colleagues was a retrospective cohort analysis involving 2,546 individuals aged 40 to 70 years. The participants had a confirmed diagnosis of SLE, RA, or psoriatic disease and no prior history of atherosclerotic cardiovascular disease (ASCVD). All participants had undergone at least one routine chest CT at two major medical centers in Boston between 2000 and 2023.
The Role of Artificial Intelligence
The innovative core of this study was the use of a validated AI algorithm to quantify CAC from non-gated chest CT scans. Traditionally, measuring CAC requires specialized cardiac CT protocols. However, the AI methodology used here allows for the automated detection and quantification of calcium in the coronary arteries on standard chest CTs, which are often performed for non-cardiac indications.
Endpoints and Analysis
The researchers categorized the AI-derived CAC scores (CAC-AI) into three groups: 0, 1-99, and ≥100. The primary outcomes were all-cause mortality and MACE, defined as a composite of nonfatal myocardial infarction, coronary revascularization, nonfatal stroke, or cardiovascular mortality. The median follow-up period was 8.1 years, providing a robust window to observe long-term outcomes. Cox proportional hazards modeling was adjusted for age, sex, and traditional cardiovascular risk factors to ensure the independence of the CAC-AI findings.
Key Findings: High Prevalence and Stark Prognostic Value
The results of the study highlight both a significant medical challenge and a clear path forward for risk stratification.
Prevalence of Subclinical Atherosclerosis
Among the cohort of 2,546 individuals (median age 59 years; 66.5% women), more than half (53%) had a CAC-AI score greater than zero. This indicates that a majority of middle-aged patients with IMIDs already have established coronary atherosclerosis. Perhaps most strikingly, despite this high prevalence of disease, only 6.0% of the cohort were on statin therapy at the time of the CT scan. This represents a massive gap between the presence of disease and the implementation of evidence-based preventive care.
Association with Adverse Outcomes
The prognostic value of the AI-detected calcium was clear and dose-dependent:
1. Low Burden (CAC-AI 1-99): Even a small amount of calcium was associated with a significant increase in risk. These patients had an adjusted Hazard Ratio (aHR) of 1.41 for all-cause mortality and 2.05 for MACE compared to those with a score of zero.
2. High Burden (CAC-AI ≥100): The risks were even more pronounced in this group, with an aHR of 2.45 for all-cause mortality and 3.24 for MACE.
These findings suggest that the detection of any calcium in this population should be viewed as a high-risk marker, potentially warranting a shift in clinical management.
Mechanistic Insights: Inflammation and Calcification
The relationship between IMIDs and CAC is not merely incidental. Chronic inflammation accelerates the transition of vascular smooth muscle cells into osteoblast-like cells, leading to the deposition of calcium in the arterial wall. In patients with RA or SLE, the inflammatory milieu may lead to the development of “vulnerable” plaques that, while perhaps less calcified than those in older non-IMID patients, are more prone to rupture. The fact that even low CAC scores (1-99) were so strongly predictive of MACE in this study suggests that in the context of systemic inflammation, any visible calcium is a sign of a high-risk vascular environment.
Clinical Implications: Closing the Treatment Gap
The most immediate takeaway for clinicians is the potential for “opportunistic” risk stratification. If a patient with an inflammatory disease undergoes a chest CT for any reason, the coronary arteries should be scrutinized. The presence of calcium, as identified by AI or expert radiologic review, provides objective evidence that the patient is at elevated cardiovascular risk, regardless of what their traditional risk score might suggest.
A Call to Action for Statins
The 6% statin use rate observed in this study is a sobering statistic. Given the high hazard ratios associated with CAC in this population, there is a strong argument for more aggressive lipid-lowering therapy. Future guidelines may need to incorporate incidental CAC findings as a “risk enhancer” that triggers the initiation of statins in IMID patients who would otherwise fall below the treatment threshold.
Study Limitations and Considerations
While the study is robust, some limitations must be considered. As a retrospective analysis, it can show association but not direct causality. Additionally, the cohort was drawn from two major academic centers, which may not represent all practice settings. The use of non-gated CT scans, while practical for opportunistic screening, is slightly less precise than dedicated gated cardiac CTs, though the AI algorithm used has been validated to mitigate these discrepancies.
Conclusion: Integrating AI into Preventive Rheumatology
The study by Weber et al. demonstrates that AI-detected coronary artery calcium is a powerful and underutilized tool in the management of patients with immune-mediated inflammatory diseases. In a population where cardiovascular risk is high and often invisible, the ability to extract prognostic information from routine imaging represents a significant step forward in personalized medicine.
By recognizing that more than half of these patients have subclinical atherosclerosis—and that this calcium is a direct harbinger of future MACE and death—clinicians can move toward a more proactive, preventive model of care. The integration of AI tools to automate this detection could ensure that no opportunity for cardiovascular risk reduction is missed, ultimately saving lives in a population that has been historically undertreated.
References
1. Weber BN, Biery DW, Petranovic M, et al. Prevalence and Prognostic Value of Incidentally Detected Coronary Artery Calcium Using Artificial Intelligence Among Individuals With Immune-Mediated Inflammatory Diseases. JACC Cardiovasc Imaging. 2025;S1936-878X(25)00521-2. doi:10.1016/j.jcmg.2025.08.020.
2. Agca R, Heslinga SC, Rollefstad S, et al. EULAR recommendations for cardiovascular disease risk management in patients with rheumatoid arthritis and other forms of inflammatory joint diseases: 2015/2016 update. Ann Rheum Dis. 2017;76(1):17-28.
3. Greenland P, Blaha MJ, Budoff MJ, Gaziano JM, Hecht HS, Lauer MS. Coronary Calcium Score and Cardiovascular Risk. J Am Coll Cardiol. 2018;71(4):434-447.
