Article Text
PDF
Abstract
Objective Coronary CT angiography (CCTA) permits both qualitative and quantitative analysis of atherosclerotic plaque and may be a suitable risk modifier in assessing patients at intermediate risk of atherosclerotic cardiovascular disease. We sought to determine the association of plaque components with long-term major adverse cardiovascular events (MACEs) in asymptomatic intermediate-risk patients, compared with conventional coronary artery calcium (CAC) score.
Methods 100 intermediate-risk patients underwent double-blinded CCTA. Follow-up was conducted at 10 years and data were cross-referenced with the National Death Index. The primary outcome was MACE, which was a composite of death, acute coronary syndrome (ACS), revascularisation and stroke.
Results The median time from CCTA to follow-up was 9.5 years. 83 patients completed follow-up interview and mortality data were available on all 100 patients. MACE occurred in 17 (20.5%) patients, which included 2 (2%) deaths, 8 (10%) ACS, 3 (4%) strokes and 5 (6%) revascularisation procedures. 47 (57%) patients had mixed plaque, which was predictive of MACE (OR 4.68 (95% CI 1.19 to 18.5) p=0.028). The burden of non-calcified and mixed plaque, defined by non-calcified plaque segment stenosis score, was also a predictor of long-term MACE (OR 1.59 (95% CI 1.18 to 2.13) p=0.002). Neither calcified plaque (OR 3.92 (95% CI 0.80 to 19.3)) nor CAC score (OR 1.01 (95% CI 0.999 to 1.02)) was associated with long-term MACE.
Conclusion The presence and burden of mixed plaque on CCTA is associated with an increased risk of long-term MACE among asymptomatic intermediate-risk patients and is a superior predictor to CAC score.
- Computed Tomography Angiography
- Atherosclerosis
- CORONARY ARTERY DISEASE
- RISK FACTORS
- Acute Coronary Syndrome
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
- Computed Tomography Angiography
- Atherosclerosis
- CORONARY ARTERY DISEASE
- RISK FACTORS
- Acute Coronary Syndrome
WHAT IS ALREADY KNOWN ON THIS TOPIC
Plaque components, such as non-calcified plaque and plaque burden, identified on coronary CT angiography (CCTA) have been shown to be prognostically significant in patients presenting with chest pain. The coronary artery calcium (CAC) score provides incremental risk prediction in asymptomatic cohorts, however, it can miss non-calcified plaque and potentially underestimate future risk.
WHAT THIS STUDY ADDS
The presence of mixed plaque on CCTA is associated with an increased risk of long-term major adverse cardiovascular events (MACEs) in asymptomatic patients at intermediate risk of cardiovascular disease. The non-calcified plaque segment stenosis score is a superior predictor of long-term MACE than CAC score.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Contemporary primary prevention strategies using non-invasive imaging may be better focused on identifying non-calcified plaque rather than coronary calcification in light of our study findings. Future randomised, prospective trials are required to determine whether initiation of treatment based on the presence of mixed plaque on CCTA significantly impacts outcomes.
Introduction
Globally, coronary artery disease (CAD) remains the leading cause of death, despite key contemporary advances in pharmacological and interventional therapies.1 Current guidelines emphasise early detection of at-risk individuals to allow for implementation of preventative treatment to impede the progression of CAD from a subclinical syndrome to clinically manifested cardiovascular events.
There is an established relationship between an elevated coronary artery calcium (CAC) score and incremental risk of cardiovascular disease in asymptomatic individuals.2 Accordingly, CAC scoring features in current international guidelines for the prevention of cardiovascular disease to refine risk estimation and inform the need for intensified preventative therapy in patients at intermediate risk of cardiac events.3 4 At present, although coronary CT angiography (CCTA) has been shown in several asymptomatic cohorts to assist in predicting cardiac events,5–9 it is not currently integrated into the guidelines as a risk-modifying tool.
As distinct from CAC scoring, which is a surrogate marker of atherosclerotic burden, CCTA allows for comprehensive qualitative and quantitative assessment of coronary plaque. There is a growing interest in the prognostic significance of non-calcified and mixed plaques, with non-calcified plaque burden shown to be the strongest predictor of adverse cardiovascular events in patients with stable angina in a subanalysis of the Scottish CT of the Heart study, above plaque burden, stenosis severity and CAC score.10 However, the role of CCTA in informing risk stratification in the asymptomatic population has not been established.
In this study, we present the 10-year follow-up of a cohort of 100 patients at intermediate risk of cardiovascular events who underwent screening CCTA, in whom both the patients and their treating physicians were blinded to the CCTA results. We sought to directly compare the utility of the two predominant plaque components imaged with CCTA (non-calcified plaque and coronary calcification) in the prediction of future cardiovascular events.
Methods
Patient selection
The current study is the long-term follow-up of a cohort of patients at intermediate 10-year risk of cardiovascular events who underwent double-blinded CCTA scanning. The patient selection process and CCTA protocol have been previously described.11 In brief, prospective subjects were identified from a database registry managed by the Baker Heart Research Institute in Melbourne, Australia, which contained clinical information on over 20 000 volunteers. Participants were classified at intermediate risk of CAD according to the Framingham cardiovascular risk assessment,12 which at that time was defined as a 10%–20% 10-year risk of myocardial infarction (MI) or coronary death. Exclusion criteria included established CAD, anginal symptoms, diabetes mellitus, chronic atrial fibrillation, significant renal dysfunction (estimated glomerular filtration rate <30 mL/min/1.73 m2) and contrast allergy.
Coronary CT angiography
All CCTA scans were conducted at the Alfred Hospital in Melbourne, Australia between July 2010 and November 2011. CCTA was performed using a 64-slice CT scanner (Light-Speed VCT, GE Healthcare, USA) with ECG gating. Oral or IV beta blockers were administered for heart rate control, aiming <60 beats per min. Sublingual nitroglycerin (400 µg) was administered for coronary vasodilatation in subjects with systolic blood pressure >100 mm Hg. A non-contrast scan was performed first to obtain CAC score. Following initial scout imaging, a bolus of intravenous iodinated contrast (Visipaque, GE Healthcare) was administered to confirm the correct timing of contrast CCTA image acquisition. Subsequently, a 60–100 mL contrast bolus was given at 5 mL per second followed by a 20 mL contrast bolus and then 40 mL of normal saline at a rate of 3 mL per second.
Image analysis was performed by two experienced CCTA readers who were blinded to the subjects’ clinical data on a dedicated cardiac workstation (Advantage Workstation V.4.2, GE HealthCare). In the event of significant discrepancy between readers, a third experienced reader adjudicated on results. Using unformatted axial slices, maximum intensity projection, three-dimensional reconstructions and multiplanar reformatted coronary images, assessment for coronary artery plaque was performed using a 16-segment model. All available coronary segments were assessed, irrespective of size. CAC score was calculated using the Agatston method, with any area of hyperattenuation (>130 Hounsfield units (HUs)) classified as calcium. Coronary plaque was evaluated by the degree of luminal stenosis, ranging from minimal (<25%), mild (25%–49%), moderate (50%–69%) to severe (>70%). Plaque composition was classified as calcified, non-calcified or mixed (figure 1). Calcified plaque was defined as any structure within the wall of the coronary artery distinct from the lumen with a density of 130 HU. Any plaque classified as calcified was cross-referenced against the non-contrast CAC scan. Non-calcified plaque was defined as any structure within vessel wall with a lower CT density than the contrast-enhanced lumen but greater than the surrounding tissue.13 Plaque comprising both calcified and non-calcified plaque was classified as mixed. Obstructive plaque was defined as any plaque causing a >50% stenosis. All segments containing plaque were evaluated for the presence of high-risk plaque features, including spotty calcification (calcification <3 mm in all planes,14 the napkin ring sign (non-calcified plaque with a higher attenuation peripheral ring,15 positive remodelling (remodelling index >1.115) and low attenuation (density <30 HU).
Characterisation of coronary artery plaque by coronary ct angiography. Examples of CCTA imaging of non-calcified (A), calcified (B) and mixed (C, D) coronary artery plaques. CCTA, coronary CT-angiography.
Plaque burden was evaluated with the segment stenosis score (SSS) for total plaque, for any calcified plaque and for any non-calcified and mixed plaque. The SSS was calculated by adding the maximum stenosis score for each coronary artery segment (minimal=0, mild=1, moderate=2 and severe=3) to obtain a score ranging from 0 to 48 points. Segment involvement score (SIS) was used to determine the number of segments with plaque, irrespective of stenosis severity, ranging from 0 to 16. Segments-at-risk score (SRS) was calculated, whereby all segments subtended by a severe proximal plaque were classified as severe to generate a score between 0 and 48.16
All CCTA results were blinded both to the patient and treating clinician. Unblinding was reserved for the presence of either severe left main or triple vessel CAD due to the concern that historical studies of coronary artery bypass graft (CABG) have shown a mortality benefit in these anatomic groups.17 18
Follow-up
Standardised follow-up was performed via telephone interview and consultation of medical records between October 2020 and October 2021. Patients who were uncontactable with no medical records were cross-checked with the National Death Index (NDI) to confirm mortality status. The primary outcome was a composite endpoint of death, acute coronary syndrome (ACS), coronary revascularisation (either with percutaneous coronary intervention or CABG) and cerebrovascular accident (CVA).
Statistical analysis
Continuous variables are expressed as mean±SD and categorical variables are expressed as number (percentage).
The relationship between plaque type and severity scores and major adverse cardiovascular event (MACE) was examined with logistic regression, producing ORs with 95% CIs, and adjusted for age and gender. CAC score was examined as a categorical variable, in increments of 10.
The decision to compare groups based on the presence or absence of mixed plaque was made on the basis that mixed plaque was an independent predictor of MACE. Although non-calcified plaque was also a predictor of MACE, there was only one patient who had non-calcified plaque in isolation.
Continuous variables were compared between groups using Kruskal-Wallis equality-of-populations rank test or t-test as appropriate. Categorical variables were compared using Pearson’s χ2 test.
Primary outcome analysis was performed using Kaplan-Meier survival estimates, with the log-rank test used for the comparison between those with and without mixed plaque, as well as differences according to overall plaque status (which included no plaque, mixed plaque and calcified plaque only).
Statistical analysis was performed by using Stata/MP V.14.2 for Windows. Values of p<0.05 were considered to represent statistical significance.
Results
Complete follow-up data were available on 83 participants. One subject declined to participate and the remaining 16 participants were uncontactable with no information within their medical records. Cross-referencing with the NDI confirmed that all 16 subjects were alive.
The baseline characteristics are summarised in table 1. The mean age was 61±6 years and 83% were male. The mean body mass index (BMI) was 28±5 kg/m2 and 48% had a family history of CAD. Almost half (47%) had a history of smoking. Only 12% of patients were taking antiplatelet therapy and 33% were on lipid-modifying therapy. Regarding antihypertensive use, 13% were taking beta blockers, 10% were taking calcium channel blockers (CCBs) and 43% were taking either an ACE inhibitor or an angiotensin II receptor blocker (ACEi/ARB). The mean 10-year Framingham risk score was 12%±6%. There were no significant differences between baseline characteristics among those with and without mixed plaque.
Baseline characteristics
The CCTA findings are displayed in table 2. 68 patients (87%) had coronary plaque, of which 47 had non-obstructive plaque and 21 had obstructive (>50%) plaque. Regarding plaque composition, 54 patients (65%) had calcified plaque, which included 17 patients with calcified plaque only. There were 47 patients (57%) with mixed plaque. Nine patients (11%) had non-calcified plaque, of which just one had non-calcified plaque only. The mean CAC score for the cohort was 302±509. The mean SSS was 3.1±4.8, mean SIS 3.8±3.3 and mean SRS 3.3±5.7. The mean non-calcified SSS was 1.7±3.0. 22 patients (27%) had high-risk plaque features on CCTA, of which 4 had low attenuation plaque, 1 had positive remodelling and 18 had spotty calcification. No patients had a napkin ring sign on CCTA.
Coronary CT angiography results
When CCTA findings were grouped according to mixed plaque status, all plaque scores were significantly higher among patients with mixed plaque (table 3).
Coronary CT angiography findings according to presence of mixed plaque
There was excellent agreement between reporters in the detection of plaque subtypes, with an interobserver variability of 0.93.
Table 4 shows 10-year adverse events and medication use. The median time from original CCTA to follow-up was 9.5 years. MACE occurred in 17 (20.5%) patients. Two patients (2%) had died, one from a cardiovascular cause. Eight patients (10%) had an ACS. Five patients (6%) underwent coronary revascularisation. Three patients (4%) had a CVA. MACE was more common among patients with mixed plaque (30% vs 8%, p=0.016), as was coronary revascularisation (11% vs 0%, p=0.04). Regarding medication use, 24 (29%) patients were taking aspirin and 37 patients (45%) were taking statins. ACEi/ARB use was highest, among 47 (57%) patients, while 13 patients (16%) were on beta blockers and 19 patients (23%) were taking CCB. Patients with mixed plaque had significantly higher rates of aspirin (38% vs 17%, p=0.02) and statin (53% vs 33%, p=0.05) use at follow-up.
10-year follow-up
The relationship between the CCTA derived plaque features and the primary endpoint was evaluated by logistic regression and controlled for age and gender, shown in table 5. Mixed plaque was a significant predictor of MACE with an OR of 4.68, 95% CI 1.19 to 18.5 (p=0.028). Non-calcified plaque was also significant predictor of MACE with an OR of 5.56, 95% CI 1.21 to 25.6 (p=0.028). The presence of any calcified plaque was not associated with an increased risk of MACE (OR 3.92 (95% CI 0.80 to 19.3) p=0.093), nor was the presence of calcified plaque only (OR 0.40 (95% CI 0.08 to 2.02) p=0.267). The presence of high-risk plaque features was strongly associated with an increased risk of MACE (OR 7.00, 95% CI 2.01 to 24.4, p=0.002).
Relationship of coronary CT-angiography findings to major adverse cardiovascular events, by logistic regression
Neither the presence (OR 1.40 (95% CI 0.42 to 4.62) p=0.586) nor number (OR 1.27 (95% CI 0.96 to 1.69) p=0.096) of obstructive plaques was associated with an increased risk of long-term MACE. SSS, SRS and SIS were all predictors of long-term MACE. CAC score was not an independent predictor of long-term MACE (OR 1.01, 95% CI 0.999 to 1.02, p=0.092). Conversely, the burden of non-calcified and mixed plaque, as quantified by the non-calcified SSS, was highly predictive of MACE (OR 1.59, 95% CI 1.18 to 2.13, p=0.002).
Figure 2 shows the Kaplan-Meier curve for event-free survival by mixed plaque status, demonstrating that the presence of mixed plaque was associated with a significantly increased risk of long-term MACE (p=0.03). Figure 3 demonstrates Kaplan-Meier curves for MACE according to plaque status. Mixed plaque was associated with a significantly increased risk of MACE compared with no coronary plaque (p=0.037). By contrast, the presence of calcified plaque only was not associated with an increased risk of MACE compared with an absence of coronary plaque (p=0.177 by log-rank test).
Kaplan Meier curve for event-free survival by mixed plaque status. Kaplan-Meier curve demonstrating reduced event-free survival among patients with mixed plaque compared with those without mixed plaque (p=0.03 by log-rank test).
Kaplan-Meier curve for major adverse cardiovascular events according to plaque type. Kaplan-Meier curve demonstrating reduced event-free survival among patients with mixed plaque (in green) compared with those with calcified plaque (red) and no plaque (blue).
Discussion
Our findings demonstrate that the presence and burden of mixed plaque on CCTA in asymptomatic subjects at intermediate risk of cardiovascular events is associated with a markedly increased risk of long-term MACE, with nearly one in three patients with any mixed plaque experiencing the primary endpoint at 10 years. Furthermore, quantification of mixed and non-calcified plaque burden with the non-calcified SSS was a superior predictor of long-term MACE than CAC score, which was not associated with increased long-term events. These findings suggest that CCTA, through evaluating plaque composition and burden, may be a promising adjunct screening tool in the intermediate-risk asymptomatic population, a group that is currently challenging to manage. The finding of mixed plaque was a significant risk modifier in this cohort and, if validated in larger studies, its presence may prompt consideration of intensification of primary preventative therapy.
Historically, it has been challenging to demonstrate the long-term prognostic role of screening CCTA once adverse features, such as high-grade stenoses or diffuse non-calcified plaque, have been uncovered due to the flow-on effect on physician practice, which often prompts referral for further investigation and treatment initiation. The influence of adverse CCTA findings on physician behaviour has been previously described,19 and this represents a significant limitation to prior studies evaluating the screening potential of CCTA. The double-blinded design of our study allowed for an examination of the natural history of CCTA detected subclinical atherosclerosis, which, in conjunction with the long follow-up period, gives this study unique value.
Non-calcified SSS outperformed CAC score in the prediction of long-term MACE. A key limitation of CAC scoring is that it fails to account for non-calcified plaque and can, therefore, overlook potentially high-risk atherosclerosis. Previous studies have shown it is possible to have develop adverse cardiovascular events without coronary calcification.20 21 Recently, the ICONIC (Incident Coronary Syndromes Identified by Computed Tomography) study showed that a quarter of ACS events in patients with no prior history of CAD occurred in those with a CAC score of 0.21 By contrast, non-calcified plaque correlates with invasively and histologically identified vulnerable plaque features and provides additive prognostic value in symptomatic patients.10 22–24 Non-calcified plaque is also a potentially treatable entity, with two large recent trials demonstrating substantial regression of non-calcified plaque burden with statin therapy on serial CCTAs.25 26 As such, non-calcified SSS may be a more attractive parameter than CAC score as a risk predictor and marker of treatment response.
If validated with larger prospective studies, our findings may support a role for CCTA as a screening tool for asymptomatic individuals at elevated risk of atherosclerotic cardiovascular disease (ASCVD). The asymptomatic intermediate-risk group is a cohort in whom management is challenging and our results suggest that defining the presence and burden of non-calcified plaque provides superior predictive value to CAC scoring. However, these results do not support the use of CCTA in population-based screening as widespread indiscriminate use of CCTA is costly and incurs radiation and contrast exposure. Indeed, population-based CAC scoring has recently been shown to have little influence on altering long-term mortality in a cohort of 46 611 asymptomatic men in Denmark.27 The Swedish Cardiopulmonary Bioimage Study, in which investigators performed CCTA and CAC scores on over 25 000 asymptomatic individuals, found that screening CCTA provided additive predictive value to CAC scoring, but only in those at intermediate risk.28
Current guidelines advocate using the CAC score to guide statin therapy, but this study shows the presence and burden of non-calcified plaque to be a more substantial modifier of risk. It must be acknowledged, however, that the mean CAC score in patients with mixed plaque in our study was elevated, which meant that many of these patients may have qualified for statin therapy on the basis of the CAC score alone. Prior studies have shown that statin use in patients with non-obstructive plaque on CCTA is associated with reduced all-cause mortality,29 an effect which is more pronounced among patients with higher plaque burden.30 Interestingly, in our cohort, statin use was significantly higher among patients with mixed plaque at 10-year follow-up. This may relate to increased rates of MACE in this cohort, although we did not collect data on the timing of statin initiation in relation to MACE.
In the recently published Copenhagen General Population Study, blinded CCTA was performed on 9533 asymptomatic participants, identifying that subclinical atherosclerosis is prevalent among the general population and predictive of adverse outcomes. This study demonstrated that the presence of obstructive subclinical atherosclerosis conferred an eightfold increase in risk of developing MI at 3.5 years compared with those with normal coronary arteries.9 Only a CAC score of >300 was predictive of the composite of death and MI (relative risk (RR) 2.67, 95% CI 1.81 to 3.95), but a CAC score of <299 was not. The presence of non-calcified plaque was also associated with the composite outcome (RR 1.81, 95% CI 1.20 to 2.73), but the authors did not quantify non-calcified plaque burden. Limiting this study was the relatively short follow-up period and the homogeneous patient cohort, which consisted of healthier, higher socioeconomic status participants. Because our study population was composed of intermediate-risk participants with higher BMI and higher rates of cardiac comorbidities, our results are reflective of a real-world patient cohort, and therefore, may be more applicable to guiding management.
The results of this study must be considered within the context of its limitations. Primarily, it is limited by its observational design. Furthermore, the cohort was small with low overall event rates. In addition, long-term follow-up data were missing on 17% of patients, although mortality data were available on all.
In conclusion, the presence and burden of mixed plaque on CCTA is associated with an increased risk of long-term MACE among patients at intermediate risk of ASCVD. Non-calcified SSS is a superior predictor of long-term MACE than CAC score, highlighting the potential future role of CCTA as a risk-modifying tool when screening intermediate-risk patients.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Alfred Human Research Ethics Committee (approval ID 435/20). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
Dr Josephine Warren is supported by a National Health and Medical Research Council Postgraduate Scholarship. Dr Jason Bloom is supported by National Health and Medical Research Council & National Heart Foundation Postgraduate Scholarships. A/Prof Dion Stub is supported by National Health and Medical Research Council & National Heart Foundation Fellowships. Professor Andrew Taylor is supported by a National Health and Medical Research Council Investigator Fellowship.
References
Footnotes
Twitter @drjosiewarren
Contributors JW was responsible for ethics application, follow-up interviewing, data analysis and manuscript writing. AE conducted the original CTCA study and provided assistance with data management and CT analysis. JB assisted with statistical analysis. NS assisted with CT analysis. PL and HK facilitated the CT scanning and assisted with the management of the CT studies. SP assisted with statistical analysis. DS assisted with manuscript preparation and data analysis. AT was responsible for the conception of the study and was the primary supervisor to JW in all of the aforementioned tasks. AT is the guarantor for the study.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.