Association of Cardiovascular Disease Risk Factor Burden With Progression of Coronary Atherosclerosis Assessed by Serial Coronary Computed Tomographic Angiography

Key Points Question Is the risk factor burden of cardiovascular disease, as assessed by atherosclerotic cardiovascular disease risk score, associated with coronary plaque progression and the development of adverse plaque characteristics? Findings In this cohort study of 1005 adult patients from an international multicenter registry who underwent serial coronary computed tomographic angiography, the progression of coronary atherosclerotic plaque volume and the development of adverse plaque characteristics was greater in patients with a high atherosclerotic cardiovascular disease risk score. Meaning The study findings suggest that the overall cardiovascular disease risk burden is associated with the progression of coronary atherosclerosis; the progression of fibrofatty plaque and low-attenuation plaque and the development of adverse plaque characteristics appear to be accelerated in patients with a high risk of atherosclerotic cardiovascular disease.

EXPOSURES Association of baseline ASCVD risk burden with plaque progression.

MAIN OUTCOMES AND MEASURES
Noncalcified plaque, calcified plaque, and total plaque volumes (mm 3 ) were measured. Noncalcified plaque was subclassified using predefined Hounsfield unit thresholds for fibrous, fibrofatty, and low-attenuation plaque. The percent atheroma volume (PAV) was defined as plaque volume divided by vessel volume. Adverse plaque characteristics were defined as the presence of positive remodeling, low-attenuation plaque, or spotty calcification.

Introduction
Coronary computed tomographic (CT) angiography allows quantitative measurement of multiple components of coronary atherosclerotic plaque and assessment of adverse plaque characteristics. [1][2][3][4] In addition, the development and progression of coronary atherosclerotic plaque across the entire coronary tree can be evaluated using serial coronary CT angiography scans. 5,6 Serial assessment of coronary artery plaques through coronary CT angiography provides clinical information regarding the progression of disease and the risk of experiencing future adverse cardiovascular events. 7,8 Although several studies have reported an association between individual cardiovascular disease (CVD) risk factors and plaque progression, [9][10][11][12] the cumulative consequences of multiple risk factors for plaque progression and the development of adverse plaque characteristics have not been well characterized. Current guidelines recommend the application of the 10-year atherosclerotic CVD (ASCVD) risk score, 13 a validated model that incorporates age, sex, and traditional CVD risk factors to estimate the likelihood of cardiovascular events over 10 years. 14 We aimed to explore the association of CVD risk factor burden, as measured by the 10-year ASCVD risk score, with coronary plaque progression and the development of adverse plaque characteristics in a large international longitudinal cohort using serial coronary CT angiography.

Study Population
The study population was acquired using data from the Progression of Atherosclerotic Plaque Determined by Computed Tomographic Angiography Imaging (PARADIGM) study. The PARADIGM registry has been previously described. 15 In brief, the registry is a prospective international multicenter dynamic observational database designed to evaluate the association between serial coronary CT angiography findings and clinical presentation. Baseline data for the PARADIGM registry   represent adult participants who received serial coronary CT angiography scans between December   24, 2003, and December 16, 2015, with follow-up through November 24, 2016. A total of 2252   consecutive adult participants underwent serial coronary CT angiography scans at an interval of 2 or   more years at 1 of 13 centers in 7 countries (Brazil, Canada reporting guideline for cohort studies. Among 2252 consecutive participants, we excluded 754 patients with coronary CT angiography images that were inadequate for quantitative plaque analysis of the entire coronary tree, 282 patients who had a previous coronary revascularization, 133 patients who experienced an adverse cardiovascular event (defined as a myocardial infarction or revascularization) between serial coronary CT angiography scans, and 78 patients for whom the ASCVD risk score could not be calculated. After exclusions, 1005 patients were included in the current analysis. Baseline demographic characteristics, including age, sex, smoking status, and presence of hypertension, diabetes, or dyslipidemia, were collected at the baseline and follow-up coronary CT angiography scans. The 10-year ASCVD risk score, which was calculated using the pooled cohort equation 13 based on information obtained at baseline coronary CT angiography, was used to assess CVD risk factor burden. Participants' risk factor burdens were categorized as low (<7.5%), intermediate (7.5%-20.0%), or high (>20.0%). 14

Coronary CT Angiography
All testing, data acquisition, and image postprocessing for coronary CT angiography were performed in accordance with the Society of Cardiovascular Computed Tomography guidelines. 16 The coronary CT angiography scans were acquired in all centers using a scanner with 64 or more detector rows.
Baseline and follow-up data sets from each center were transferred to an offline workstation for analysis using semiautomated plaque analysis software (QAngio CT Research Edition, version 2.1.9.1; Medis Medical Imaging Systems) with manual correction as needed. Independent blinded readers who were experienced with coronary CT angiography (Core Cardiovascular Training Statement [COCATS] level 3 certification) analyzed all images. Segments were matched between baseline and follow-up coronary CT angiography scans using branch points as landmarks. For longitudinal comparisons of coronary CT angiography images, both baseline and follow-up coronary segments were coregistered using fiduciary landmarks, including distance from ostia or branch vessel takeoffs.
Plaques were qualitatively assessed for adverse characteristics, including positive remodeling, low-attenuation plaque, or spotty calcification. A remodeling index was defined as the maximal lesion vessel diameter divided by the proximal reference vessel diameter. Positive remodeling was defined as a remodeling index greater than 1.1, and low-attenuation plaque was defined as any voxel less than 30 Hounsfield units (HUs) within an individual coronary plaque. 7,17 An intralesion calcific plaque less than 3 mm in length that composed less than 90 degrees of the lesion circumference was defined as spotty calcification. 17 Development of adverse plaque characteristics was defined as the presence of a new lesion with adverse plaque characteristics on the follow-up coronary CT angiography scan or the development of adverse plaque characteristics from a lesion without adverse plaque characteristics on the baseline coronary CT angiography scan.
Plaque volumes (measured in mm 3 ) of all coronary segments were obtained and summed to generate the total plaque volume on a per-patient level. Atherosclerotic plaque volume was further subclassified by composition, employing predefined intensity cutoff values in HU, including low-attenuation plaque (−30 to 30 HU), fibrofatty plaque (31-130 HU), fibrous plaque (131-350 HU), and calcified plaque (>350 HU). 18,19 Percent atheroma volume (PAV) was defined as total plaque volume divided by vessel volume. 2 The PAV was also calculated for each subtype of plaque composition. Rapid plaque progression was defined as an increase from baseline total PAV of more than 0.59% per year (the mean value of total PAV progression in the study population) on the follow-up coronary CT angiography scan.

Statistical Analysis
Continuous variables were reported as mean (SD) or median (interquartile range [IQR]) and were compared using a t test or a Wilcoxon rank sum test (as appropriate) for 2-group comparisons and 1-way analysis of variance or a Kruskal-Wallis test (as appropriate) for comparisons of more than 2 groups. Categorical variables were reported as numbers and percentages and compared using the Pearson χ 2 test. The association between ASCVD risk score and plaque progression was assessed using linear regression analyses and reported as correlation coefficient (β). All statistical tests were 2-sided and performed on independent or unpaired groups, with P < .05 considered statistically significant. All statistical analyses were performed using Stata software, version 13 (StataCorp LLC).
The mean (SD) 10-year ASCVD risk score was 11.3 (9.9). The baseline characteristics of the study population according to 10-year ASCVD risk groups are shown in Table 1

Plaque Progression and ASCVD Risk
The median interval between coronary CT angiography scans was 3.
The results of linear regression analyses for the association between the annualized PAV progression of total plaque and the ASCVD risk score are shown in Table 3. In the multivariate analysis, the ASCVD risk score was significantly correlated with the annualized PAV progression of total plaque (β = 0.108; SE = 0.238; P < .001) after adjusting for statin use, PAV, and the presence of adverse plaque characteristics at the baseline coronary CT angiography scan.
The annualized PAV progression according to plaque components is described in Figure 1 and   (Figure 2). The incidence of adverse plaque characteristics did not significantly differ between the low-risk and intermediaterisk groups.

Rapid Plaque Progression
A total of 77 of 169 patients (45.6%) with a high risk of ASCVD experienced rapid plaque progression.
Patients with high ASCVD risk and rapid plaque progression were more frequently men (50 of 77

Discussion
In this study, we found that overall CVD risk burden was associated with the progression of coronary atherosclerosis. A high risk of ASCVD was associated with more rapid progression of coronary atherosclerosis, including calcified plaque, fibrofatty plaque, and low-attenuation plaque volumes, on serial coronary CT angiography scans as measured by the PAV. Furthermore, the incidence of new adverse plaque characteristics increased among patients who had a high risk of ASCVD compared with those who had a low to intermediate risk.
Previous studies of patients who underwent serial coronary CT angiography examinations have reported that clinical factors and laboratory values are associated with the rate of plaque progression. [10][11][12] The presence of conventional risk factors, such as diabetes 9,10 or high low-density lipoprotein cholesterol levels, 11,12 are associated with accelerated plaque progression. Our findings expand on these earlier findings by indicating that patients with a higher CVD risk factor burden, assessed with a validated global risk score, have accelerated plaque progression compared with patients with a lower risk. Most patients with CVD do not present with a single CVD risk factor, 20,21 and physicians typically integrate multiple conventional CVD risk factors in clinical practice. 14,22 Therefore, the current findings provide practical insights on the development and evolution of atherosclerotic plaques based on a patient's overall CVD risk burden as assessed by the ASCVD risk score.
A growing body of evidence suggests that noncalcified plaque components are more closely associated with CVD risk than calcified plaque components. The 2018 Incident Coronary Syndromes Identified by Computed Tomography (ICONIC) study observed that patients with acute coronary syndrome associated with coronary plaques had substantially larger noncalcified plaque volumes, most notably low-attenuation plaque volume, before they developed acute coronary syndrome compared with a similar degree of stenotic plaque in patients without acute coronary syndrome, while there was no difference in total plaque and calcified plaque volumes. 21 In the current study, we found that although plaque progression occurred in patients with low to intermediate risk, the progression of high-risk noncalcified plaque components, including fibrofatty plaque and low-attenuation plaque volumes, was significantly accelerated in patients with a high CVD risk factor  burden compared with patients with a low to intermediate risk factor burden. However, it is unknown whether interventions in these patients were associated with improvements in clinical outcomes.

JAMA Network Open | Cardiology
Because coronary atherosclerosis is a dynamic process, with plaques gaining or losing adverse plaque characteristics over time, the development of adverse plaque characteristics may be an important step from subclinical atherosclerosis to an acute coronary syndrome event. 7,23 In a study of 449 patients who underwent serial coronary CT angiography scans, the development of high-risk plaque features (ie, positive remodeling and low-attenuation plaque) was independently associated with acute coronary syndrome. 7 In the current study, we observed that a high CVD risk factor burden indicated not only a high prevalence of adverse plaque characteristics at the baseline coronary CT angiography scan but also a significant acceleration in the development of new adverse plaque characteristics.
Although plaque progression has been reported to be associated with an increased risk of developing CVD, methods and thresholds to assess clinically significant plaque progression were The current study results indicated that up to two-thirds of low-risk patients (ie, those with <7.5%) had coronary artery plaques at the baseline coronary CT angiography scan. In addition, the progression of noncalcified plaque components and the incidence of high-risk plaque were only modestly different between the low and intermediate ASCVD risk groups. These findings suggest that the arbitrary cutoffs of 5.0% to 7.5% (borderline) or 7.5% to 20.0% (intermediate) of the 10-year ASCVD score are limited in their ability to accurately differentiate the actual ASCVD risk among individuals in the low-risk to intermediate-risk populations. This finding is consistent with the current guideline, which emphasizes the need for risk-enhancing factors to reclassify ASCVD risk in the borderline to intermediate-risk group. 14 Noninvasive coronary imaging, such as coronary calcium scans or coronary CT angiography scans, may help to personalize risk assessment and shared decision-making regarding the intensity needed for the preventive strategy. [25][26][27] Limitations This study has several limitations. Although the PARADIGM registry is, to our knowledge, the largest serial coronary CT angiography study to date, there may be unmeasured confounding factors which have implications for the results of this study. We did not have detailed medication information or measures of patient adherence during the interval between scans, which could be used to perform a more refined analysis of the association of statin therapy with plaque progression. The ASCVD risk score was originally validated for 10-year outcomes in asymptomatic patients; therefore, the ASCVD risk score may not completely reflect risk in symptomatic patients who are referred for coronary CT angiography. We included patients who had coronary CT angiography scans with sufficient image quality to allow quantitative assessment of both the baseline and follow-up scans for the purpose of assessing plaque progression in the entire coronary tree. Compared with the excluded patients, the analyzed patients had a lower prevalence of cardiovascular risk factors and medication receipt (eTable 3 in the Supplement). Thus, the potential consequences of selection bias for the generalizability of the findings should be considered. Patients who underwent revascularization