A coronary artery calcium score–weighted clinical likelihood strategy may have improved patient classification and reduced unnecessary coronary computed tomography angiography in patients with stable chest pain and low clinical likelihood of coronary artery disease, according to a multicenter registry analysis.
Study Design
In the study, the investigators compared calcium score–based strategies with universal coronary computed tomography angiography (CCTA) in patients with a risk factor–weighted clinical likelihood of coronary artery disease (CAD) between 5% and 15%. They analyzed data from the CCTA Improves Clinical Management of Stable Chest Pain registry, a prospective multicenter cohort of patients undergoing CCTA as the initial cardiovascular imaging test for suspected chronic coronary syndrome.
Among 51,396 registry participants, the investigators identified 19,682 patients with stable chest pain and a risk factor–weighted clinical likelihood of coronary artery disease (CAD) between 5% and 15%.
The mean age of the participants was 58 years and 58% identified as male. Cardiovascular risk factors included hypertension (55%), hyperlipidemia (40%), smoking (37%), and diabetes (17%).
The investigators compared three diagnostic strategies:
-
Coronary artery calcium score (CACS) strategy: referral when calcium score was greater than 0
-
CACS-weighted clinical likelihood strategy: combines calcium score with traditional risk factors
-
All CCTA strategy: all patients undergo angiography.
Although all of the patients underwent CCTA, the investigators modeled how calcium score–based strategies would classify patients into referral and deferral groups.
The primary endpoint was major adverse cardiovascular events (MACE), defined as all-cause mortality or nonfatal myocardial infarction.
Clinical Outcomes
During a median follow-up of 76 months, 9.8% of the patients experienced a MACE, including 4.3% deaths and 6.0% nonfatal myocardial infarction.
Compared with the deferral group, the patients in the referral group defined by the combined model had:
-
5.3 times the risk of MACEs
-
4.6 times the risk of all-cause mortality
-
6.2 times the risk of nonfatal myocardial infarction
-
5.5 times the risk of late revascularization.
The calcium score–only strategy showed a 4.1 greater likelihood of MACEs between the referral and deferral groups.
Diagnostic Classification
The CACS-weighted model also improved the diagnostic classification of CAD.
Net reclassification improvement compared with the calcium score strategy ranged from about 6.7% to 18.7%, depending on the diagnostic endpoint, including stenosis detected by CCTA or invasive coronary angiography.
The combined strategy showed a higher diagnostic yield and correct down-classification rates with lower testing utilization and omission diagnosis.
For instance, detection of coronary artery stenosis of 50% or greater on CCTA showed a 41.7% diagnostic yield with the CACS-weighted strategy compared with 36.8% using the calcium score strategy and 21.8% when all patients underwent CCTA.
Comparison With Universal Imaging
Compared with universal CCTA, the CACS-weighted model reduced imaging utilization.
The investigators estimated that:
-
Three calcium score tests were required to avoid one angiography study.
-
Seventeen calcium score tests were needed to miss one necessary angiography study.
Across diagnostic endpoints, the number needed to test to miss one finding ranged from 5.6 to 124.6.
Limitations
The investigators noted that the registry’s observational design meant that all patients underwent CCTA, which may have introduced verification bias when evaluating alternative strategies. Functional testing and intracoronary imaging weren't routinely used, and the findings applied to stable chest pain rather than acute presentations.
In addition, the CACS-weighted strategy reduced angiography use but increased calcium scoring scans, and the cost-effectiveness of this approach requires further study.
“Compared with CACS strategy, CACS-[weighted clinical likelihood model-based strategy] have more potential to effectively and safely guide subsequent clinical management,” wrote lead study author Peng Wang, of the Department of Cardiology at Tianjin University Chest Hospital and Tianjin Key Laboratory of Cardiovascular Emergency and Critical Care in China, and colleagues.
The researchers reported no competing interests.
Source: BMC Medical Imaging
