Introduction Coronary artery calcium (CAC) scoring, developed and validated using computed tomography (CT), is a well-established technique for quantifying coronary atherosclerosis and assessing cardiovascular risk, thereby guiding preventive therapeutic strategies. Similarly, abdominal aortic calcium (AAC) is a biomarker indicating the degree of calcification within the abdominal aorta, reflecting systemic atherosclerosis. AAC has been shown to predict cardiovascular morbidity and mortality.
AAC can be assessed using two common methods: Kidney, Ureter, and Bladder (KUB) plain film radiography and CT. Studies have demonstrated a correlation between AAC and CAC scores, with AAC suggesting the presence of asymptomatic coronary artery disease (CAD). This highlights the potential importance of quantifying AAC in predicting CAD.
Coronary Computed Tomography Angiography (CCTA) enables direct visualization of coronary artery lumen stenosis, atherosclerotic plaque composition, and high-risk plaque features, such as low attenuation, spot calcification, positive remodeling, and the napkin ring sign. CCTA is particularly valuable in patients with stable chest pain, given its high sensitivity and specificity in detecting CAD. Furthermore, it plays a critical role in guiding decisions regarding coronary revascularization procedures. Beyond assessing coronary artery stenosis and atherosclerotic plaque characteristics, analyzing plaque volume and burden provides prognostic value in stratifying the risk of acute coronary syndrome (ACS) in patients with stable chest pain undergoing CCTA.
The Coronary Artery Disease Reporting and Data System (CAD-RADS) is a comprehensive framework designed to standardize the assessment of disease severity and guide treatment decisions in patients with CAD. Initially published in 2016, CAD-RADS employs a categorical system based on coronary artery stenosis and high-risk plaque characteristics as modifiers. The 2022 update, CAD-RADS 2.0, incorporates plaque burden as a new subclassification and assesses lesion-specific ischemia using CT fractional flow reserve (CT-FFR) or myocardial CT perfusion (CTP). CAD-RADS aims to standardize CCTA reporting and improve communication with referring physicians, including recommendations.
Previous studies have indicated a correlation between AAC and CAC, with AAC quantification linked to traditional cardiovascular risk factors and cardiovascular events. Additionally, CT-based AAC quantification can predict future cardiovascular events in asymptomatic adults. However, few studies have explored whether AAC correlates with non-calcified coronary artery plaques or the degree of coronary artery stenosis. This study aims to investigate the correlation between CT-based AAC scores and coronary plaque burden and stenosis, using the CAD-RADS 2.0 classification system.
Materials and Methods Study Population This prospective cohort study enrolled patients aged 60 years and older who presented to the cardiovascular department at Kaohsiung Chang Gung Memorial Hospital between June 2021 and May 2023. Inclusion criteria were patients with chest pain or discomfort who were scheduled to undergo cardiac CT for CAC and CCTA, as determined by the treating physician for the evaluation of CAD. Exclusion criteria included: (1) severe allergies to contrast agents, (2) metallic implants in the lumbar spine, (3) renal function abnormalities (estimated glomerular filtration rate \[eGFR\] \< 60), (4) abdominal aortic aneurysms, and (5) prior coronary artery stent placements or bypass surgery. The study received approval from the Research Ethics Committee of Kaohsiung Chang Gung Memorial Hospital, Taiwan. Written informed consent was obtained from all participants. Initially, 80 patients were enrolled; however, two withdrew for personal reasons, one was excluded due to the discovery of an incidental abdominal aortic aneurysm, and nine were excluded due to prior coronary artery stent placement. Consequently, 68 patients were included in the final analysis.
Patient demographics were reviewed, including age, gender, and the presence of hypertension, diabetes mellitus, lipid profiles (total cholesterol, HDL, and LDL levels from the closest pre-CCTA tests), smoking status, and body mass index (BMI). Hypertension was defined as a history of antihypertensive medication use or resting blood pressure readings of ≥ 140 mmHg systolic or ≥ 90 mmHg diastolic. Diabetes mellitus was defined as a fasting plasma glucose level ≥ 126 mg/dL or prior use of oral hypoglycemic agents or insulin. Smoking status was based on any history of smoking at least one cigarette daily before CCTA.
The data collected were used to assess 10-year atherosclerotic cardiovascular disease (ASCVD) risk, categorized into three groups: low to borderline risk (\< 7.4%), intermediate risk (7.5% to 19.9%), and high risk (≥ 20%) according to the guidelines of the American College of Cardiology.
Image Acquisition CCTA and Non-Enhanced Abdominal CT Scan Protocol CCTA was performed using a 640-slice multislice CT scanner (Canon Aquilion One Genesis, Canon Medical Systems, Japan). All participants received sublingual nitroglycerin (0.3 mg) to promote coronary vasodilation prior to imaging. The first phase involved an unenhanced, prospective electrocardiogram (ECG)-gated volume scan for CAC assessment, configured to 120 kVp and 50 mAs with a field of view (FOV) of 16 cm to ensure comprehensive cardiac coverage. Imaging was performed with a rotation time of 0.275 seconds and a slice thickness of 2 mm. Following this, an abdominal aorta scan was performed using a non-ECG-gated sequential scan mode, maintaining identical settings of 120 kVp and 2 mm slice thickness. The inferior aspect of the L5 vertebral body endplate was used as the caudal extent of the abdominal volume to be imaged, covering the iliac bifurcation of the infrarenal abdominal aorta while minimizing radiation exposure to the pelvic genital organs.
For the enhanced CCTA scans, intravenous contrast medium was delivered using a tailored volume of Omnipaque 350 (Iohexol, GE Healthcare, Chicago, IL, USA), calculated at a 1:1 ratio relative to the patient's body weight, capped at a maximum dose of 65 milliliters (mL). Contrast infusion was performed at a rate of 4-5 mL/s, followed by a saline flush of 50 mL at the same rate. Image acquisition was synchronized with the cardiac cycle using automated peak enhancement detection in the ascending aorta, triggering the scan upon reaching a threshold of 100 Hounsfield units (HU), with imaging beginning after a 5-second delay.
The data acquisition protocol was tailored to each patient's heart rate and rhythm. Specifically, a prospectively ECG-triggered volume scan was used to account for heart rates and rhythms, capturing predefined scanning areas during R-R intervals, especially in instances of frequent ectopic beats or irregular heart rates noted during pre-scan ECG monitoring. Imaging system parameters included a gantry rotation time of 0.275 seconds, collimation of 0.5 mm, tube voltage settings ranging from 100 kV to 120 kV, and a tube current-time product between 200 mAs and 450 mAs, with adjustments based on the patient's BMI.
Image Analysis CAC and AAC were quantitatively assessed using the Vitrea workstation (version 7.4.0.462, Vital Images, Plymouth, MN) following the Agatston method. Calcification was identified as a plaque ≥ 1 mm² with a density exceeding a threshold of 130 HU, which was automatically quantified for both AAC and CAC areas. The AAC and CAC scores were calculated by multiplying the area of each region by a weighted score corresponding to the highest density of calcification: 1 for 130-199 HU, 2 for 200-299 HU, 3 for 300-399 HU, and 4 for densities ≥ 400 HU. Three abdominal aortic coverage areas were analyzed, with the aortoiliac bifurcation serving as the caudal boundary (Figure 1). The first coverage (AAC-all) included the entire abdominal aorta from the superior rim of the celiac trunk origin to the aortoiliac bifurcation. The second coverage (AAC-8cm) extended 8 cm cranially from the aortoiliac bifurcation, comprising 40 slices of 2 mm each, while the third coverage (AAC-5cm) extended 5 cm cranially from the aortoiliac bifurcation, comprising 25 slices of 2 mm each.
The CAD-RADS 2.0 classification system was used to assess the severity of coronary stenosis across five distinct categories: CAD-RADS 0 (no visible stenosis, 0% maximal coronary stenosis), CAD-RADS 1 (minimal stenosis, 1-24%), CAD-RADS 2 (mild stenosis, 25-49%), CAD-RADS 3 (moderate stenosis, 50-69%), CAD-RADS 4 (severe stenosis, subdivided into 4A \[70-99%\] and 4B \[left main \>50% or three-vessel obstructive disease\]), and CAD-RADS 5 (100% occlusion). The plaque burden classification within CAD-RADS 2.0 includes a "P" designation, ranging from P1 to P4, to quantify overall plaque volume. Two assessment methodologies, CAC scoring and the segment involvement score (SIS), were employed: P1 indicated mild plaque burden (CAC of 1-100 or SIS ≤ 2), P2 indicated moderate plaque burden (CAC of 101-300 or SIS 3-4), P3 indicated severe plaque burden (CAC of 301-999 or SIS 5-7), and P4 indicated extensive plaque burden (CAC ≥1000 or SIS ≥8). The most severe classification from these two assessments determined the P subclassification. The study cohort was categorized into two distinct groups based on AAC-all scores of 300, adapted from the CAC score of 300 to differentiate plaque burden into mild to moderate and severe to extensive categories.
The outcome measures in this study were categorized based on the 2022 Coronary Artery Disease-Reporting and Data System (CAD-RADS), which classifies coronary artery stenosis and plaque burden into distinct categories without the use of specific measurement units. Coronary artery stenosis was graded using a numeric scale from 0 to 5, with higher numbers indicating greater severity of stenosis. Plaque volume was categorized using a classification system ranging from P1 to P4, with P4 representing the highest plaque burden.
Statistical Analysis Statistical analysis was performed using SPSS version 22.0 (IBM SPSS, Chicago, IL, USA). Interobserver agreement for CAD-RADS 2.0 was evaluated using Cohen's kappa coefficient, categorized as poor (\< 0.20), fair (0.20-0.39), moderate (0.40-0.59), substantial (0.60-0.79), and excellent (≥ 0.80). Prior to analysis, data distribution normality was assessed using the Kolmogorov-Smirnov test. Quantitative data were summarized as mean ± standard deviation or median (interquartile range, IQR) if not normally distributed. A significance level of p \< 0.05 was set. Continuous variables were compared using independent t-tests or Mann-Whitney U tests, while chi-square or Fisher's exact tests were applied to categorical variables as appropriate. One-way ANOVA or Kruskal-Wallis one-way ANOVA was used to assess differences in AAC across CAD-RADS categories and plaque burden subclassifications. Spearman correlation analysis was used to examine the relationship between CAC scores and various AAC score coverage areas, with coefficients ranging from very weak to very strong. The effectiveness of different AAC scores was assessed using receiver operating characteristic (ROC) curves to identify optimal cut-offs for predicting CAD-RADS categories and coronary plaque burden, with CAD-RADS categories stratified into groups 0-2 and 3-5, and coronary plaque burden divided into groups P1-P2 (mild to moderate) and P3-P4 (severe to extensive).
