First scan, then treat: 10 years of the SCOT-HEART study

Abstract

The widespread use of coronary computed tomography (CT) to exclude significant coronary stenoses and, more broadly, to achieve a more accurate assessment of cardiovascular risk through the evaluation of atherosclerosis has gained momentum in recent years, largely driven by growing criticism of the traditional ischaemia-centred paradigm. The SCOT-HEART studies were among the first large randomized trials to promote the use of coronary CT in patients presenting with angina. The information provided by coronary CT translated into improved clinical outcomes through the adoption of more aggressive medical therapy, particularly lipid-lowering treatment, which was selectively prescribed to patients with evidence of atherosclerosis. The results of the SCOT-HEART 2 primary prevention trial are eagerly awaited and may further influence clinical practice. Finally, photon-counting CT, representing the latest technological advance, together with the application of artificial intelligence–based image analysis algorithms, may constitute an additional step forward in the use of CT imaging.

In recent years, the search for myocardial ischaemia as a tool for preventing coronary events has been relegated to a secondary role compared with the identification of coronary atherosclerosis.¹ The development of this new prognostic approach has been supported by the negative results of previous studies investigating the detection and subsequent treatment of ischaemia by means of percutaneous coronary intervention (PCI).²

Even invasive studies incorporating intracoronary functional assessment using fractional flow reserve or instantaneous wave-free ratio² have failed to demonstrate that identifying ischaemia-producing coronary lesions and treating them with angioplasty leads to a reduction in hard clinical endpoints, such as myocardial infarction or sudden cardiac death.

Further support for imaging techniques aimed at detecting atherosclerosis arises from the observation that subclinical atherosclerosis in peripheral or coronary vascular beds is highly prevalent, even in the absence of traditional cardiovascular risk factors, and is closely associated with coronary disease. According to the PESA study, atherosclerosis was detectable in 58% of individuals classified as low coronary risk based on risk charts (FHS-Risk Score),³ and in 86% of those with a moderate risk profile.

Use of coronary CT

It is within this conceptual framework that coronary computed tomography (CT) has emerged as a non-invasive technique for the detection of coronary atherosclerosis, with the aim of tailoring and personalizing treatments designed to reduce cardiovascular risk.

Calcium score

In recent years, particularly in the United States, coronary calcium scoring by CT has been widely adopted. This technique is used to identify and quantify coronary calcifications through the Agatston score, based on the concept that coronary calcification represents a surrogate marker of atherosclerosis.⁴ Calcium is, in fact, a major component of atherosclerotic disease, although overall it is considered a stabilizing element of the atherosclerotic plaque.²

However, the use of calcium scoring has several limitations, as it does not evaluate the coronary lumen nor the lipid-rich component of atherosclerosis, which is the feature most closely associated with higher cardiovascular risk. Despite these limitations, calcium scoring has proven to be a valuable prognostic tool, particularly for ruling out atherosclerosis; indeed, the absence of coronary calcium identifies individuals at extremely low risk of cardiovascular events.⁵

The widespread availability in recent years of high-definition, low-radiation coronary CT scanners has progressively relegated calcium scoring to a secondary role, favouring contrast-enhanced coronary CT, which allows highly accurate assessment of coronary anatomy.⁶

Secondary prevention studies using CT

The SCOT-HEART study⁷ was the first large randomized trial to promote the use of coronary CT in patients with angina. This was an open-label, multicentre, parallel-group study involving patients presenting with angina-type chest pain. Participants were randomized (1:1) to receive optimal medical therapy plus coronary CT (performed in most cases with 320-detector row scanners) or optimal medical therapy alone.

At 5-year follow-up, the primary endpoint—death from coronary heart disease or non-fatal myocardial infarction, analysed according to the intention-to-treat principle—was significantly reduced in the CT group (2.3% vs. 3.9%; hazard ratio 0.59; 95% confidence interval 0.41–0.84; P = 0.004).

At extended follow-up of 10 years,⁸ death from coronary heart disease or non-fatal myocardial infarction remained less frequent in patients undergoing coronary CT compared with those receiving standard care alone (6.6% vs. 8.2%; HR 0.79; 95% CI 0.63–0.99; P = 0.044). Importantly, the rate of coronary revascularization procedures was similar between groups [15.2% vs. 15.3%; HR 1.00 (0.86–1.17); P = 0.99], whereas the prescription of therapies aimed at stabilizing atherosclerosis (lipid-lowering agents) or preventing its complications (antiplatelet therapy) was more frequent in the CT group (55.9% vs. 49.0%; odds ratio 1.17; 95% CI 1.01–1.36; P = 0.034).

The SCOT-HEART studies^7,8 support the hypothesis that identification of coronary atherosclerosis by CT improves cardiovascular prevention in both the medium and long term among patients with stable chest pain.

It should nevertheless be noted that, although the design of the SCOT-HEART study⁷ required a diagnosis of chronic coronary syndrome, typical effort-induced angina was present in only a limited proportion of patients (36% in the CT group vs. 35% in the control group). In the remaining cases, angina was classified as atypical or absent (64% vs. 65%, respectively). The study therefore occupies an intermediate position between a primary and a secondary prevention trial.

Coronary CT improved the diagnostic accuracy for atherosclerotic coronary disease.⁹ Coronary artery disease was identified in 63% of patients, while the remaining 37% had no detectable coronary lesions. Atherosclerosis was classified as moderate in 38% of cases and significant (>50% stenosis) in 25%.

Notably, the diagnosis of atherosclerotic disease changed in 7% of patients assigned to CT compared with 1% of those in the control group, whereas the diagnosis of angina attributable to epicardial coronary disease changed in 23% of patients undergoing CT vs. 1% in the control group (P < 0.001 for both comparisons).

Coronary CT primarily guided referral to invasive coronary angiography in patients with stenoses deemed significant on CT imaging. Accordingly, during the first 6 months after enrolment, the proportion of patients treated with percutaneous coronary intervention (PCI) was higher in the CT group (10.5% vs. 7.0%), but became similar between the two groups at 5-year follow-up (11.2% vs. 9.7%; P = 0.06). It can therefore be stated that the use of coronary CT did not result in a more aggressive interventional strategy, as evidenced by the comparable number of angioplasty procedures in the two groups. Moreover, it would have been difficult to attribute the observed clinical benefit to interventional procedures, considering that previous studies conducted in similar populations and based on ischaemia detection¹ had excluded a reduction in cardiovascular events with angioplasty in patients with chronic ischaemic heart disease.

The information provided by coronary CT translated into clinical benefit through the adoption of more aggressive medical therapy, particularly lipid-lowering treatment, which was selectively prescribed to patients with evidence of atherosclerosis. In the SCOT-HEART study, at 5-year follow-up, lipid-lowering and/or antiplatelet therapy was prescribed in 52% of patients in the CT group compared with 45% in the control group.

Not all randomized trials investigating the clinical utility of coronary CT have confirmed the superiority of an atherosclerosis-based approach. The PROMISE trial¹⁰ randomized 10 003 symptomatic patients to coronary CT vs. functional testing (exercise electrocardiography, myocardial perfusion imaging, or stress echocardiography). The composite primary endpoint—death, myocardial infarction, hospitalization for unstable angina, or major procedural complications—occurred at a median follow-up of 25 months in 3.3% of patients in the CT group and in 3.0% of those undergoing functional testing (adjusted hazard ratio 1.04; 95% CI 0.83–1.29; P = 0.75). The PROMISE trial therefore demonstrated that, in symptomatic patients with suspected coronary artery disease requiring non-invasive testing, an initial strategy based on coronary CT angiography did not improve prognosis at 2 years compared with functional testing.

Several differences exist between the PROMISE and SCOT-HEART studies. First, the prevalence of typical angina was markedly lower in PROMISE than in SCOT-HEART (12% vs. 35%, respectively). Second, the two trials differed in their study design. Unlike PROMISE, SCOT-HEART encouraged the use of preventive therapies (antiplatelet and/or lipid-lowering agents) in the presence of coronary artery disease even when non-obstructive. In SCOT-HEART, such therapies were prescribed at 5 years in a substantial proportion of patients (52% in the CT group vs. 45% in the control group).

Further studies will be required to clarify the effectiveness of preventive strategies based on coronary imaging with CT, particularly in the setting of primary prevention.

Primary prevention studies using CT

The SCOT-HEART 2 study (Scottish Computed Tomography of the Heart)¹¹ will evaluate whether screening with coronary CT angiography is more effective than conventional cardiovascular risk assessment in guiding the use of primary prevention therapies and in reducing the risk of myocardial infarction. The study will enrol at least 6000 asymptomatic individuals aged between 40 and 70 years with at least one coronary risk factor.

Participants assigned to the standard-of-care arm will be advised to correct modifiable risk factors and to initiate statin therapy if the estimated 10-year cardiovascular risk exceeds 10%. In the CT arm, risk factor modification will be recommended in the absence of atherosclerosis; statin and antiplatelet therapy will be prescribed in the presence of non-obstructive coronary artery disease; and, in cases of significant coronary disease, angiotensin-converting enzyme inhibitors will also be initiated, with consideration given to referral for invasive coronary angiography and possible revascularization.

Qualitative assessment of atherosclerosis by CT

Coronary CT has proven highly effective in identifying coronary lesions and quantifying luminal narrowing. Beyond the dichotomous diagnosis of absence vs. presence of disease, CT allows quantification of stenosis severity and assessment of whether a lesion is haemodynamically significant. A further expected advance of the technique lies in the characterization of atherosclerotic plaque composition.

Anatomopathological studies and in vivo intracoronary imaging have clarified the morphology of lesions responsible for sudden cardiac death.^12,13 Longitudinal studies using invasive imaging techniques, such as optical coherence tomography (OCT) and near-infrared spectroscopy–intravascular ultrasound (NIRS-IVUS), have demonstrated that identification of high-risk (vulnerable) coronary plaques enables prediction of future cardiac events.^2,14 Thin fibrous caps, a large lipid component, reduced luminal area, and inflammatory cell infiltration characterize patients at increased risk of myocardial infarction or sudden death.

The use of non-invasive coronary imaging with CT is particularly promising for primary prevention pathways. Its main advantage lies in its non-invasive nature, albeit at the expense of a spatial resolution that remains inferior to that of intracoronary imaging techniques.

CT can identify several features associated with plaque vulnerability, including positive vascular remodelling, low-attenuation plaque indicative of a lipid–necrotic core, and spotty intraplaque microcalcifications.^15,16

A post hoc analysis of the SCOT-HEART trial¹⁷ evaluated the impact of high-risk morphological plaque markers (positive remodelling and/or low attenuation), in addition to significant stenoses and coronary calcification, in 1769 patients. Cardiac death and/or non-fatal myocardial infarction occurred three times more frequently in patients with high-risk morphological features (adverse plaque) (4.1% vs. 1.4%; P < 0.001; HR 3.01; 95% CI 1.61–5.63; P < 0.001). Moreover, patients with both significant stenosis and adverse plaque features exhibited the highest risk, with a tenfold increase in the primary endpoint compared with those with normal coronary arteries (HR 11.50; 95% CI 3.39–39.04; P < 0.001).

A quantitative estimate of atherosclerosis with prognostic relevance is also available, namely total plaque volume. In the ISCHEMIA trial, plaque volume was associated with the risk of cardiac death or myocardial infarction and proved to be a more powerful predictor of risk than ischaemia itself.¹⁸

Photon-counting CT and artificial intelligence

The adoption of photon-counting CT technology provides further improvements in image definition. Conventional CT relies on detectors that convert X-ray photons into light, which is then transformed into an electrical signal by photodiodes to generate images. In contrast, photon-counting CT does not require scintillation detectors; photons interact directly with a semiconductor material, generating an electrical signal.

Recent studies have confirmed that photon-counting CT improves image quality while requiring a lower radiation dose.^19,20 This technology enhances visualization of coronary stents, the coronary lumen, coronary calcifications, and non-calcified plaque. Importantly, it substantially reduces the blooming effect that hampers luminal assessment in the presence of heavy calcification.^19,20

Comparative studies with intracoronary imaging techniques, such as OCT and IVUS with NIRS, will be required to fully elucidate the potential of this technology.

The application of artificial intelligence (AI) algorithms appears to further improve the prediction of high-risk plaques by quantifying stenosis severity and refining morphological assessment. The CREDENCE study²¹ compared CT-based evaluation with quantitative coronary angiography in 303 patients with chronic coronary syndrome. AI-assisted CT demonstrated high accuracy in identifying significant stenoses, with sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of 94%, 68%, 81%, 90%, and 84%, respectively. Good agreement with core laboratory assessment was also observed in the identification of predominantly lipid-rich plaques.²¹

Similar conclusions were reported by Choi et al. in the CLARIFY study,²² which included 232 patients undergoing CT analysed using FDA-approved software. An excellent correlation was observed between OCT-derived parameters assessed by the AI software and those evaluated by core laboratory personnel.

Other AI-based coronary CT angiography approaches relying on radiomics appear capable of analysing high-risk plaques and improving the diagnostic accuracy of the technique.²³ Through radiomics, quantitative data are extracted by analysing numerous image features, while textural information is quantified by assessing the spatial distribution of Hounsfield unit values or signal intensities and their interrelationships.

In a subanalysis of the SCOT-HEART study,²⁴ the application of radiomics to CT improved the assessment of atherosclerosis and enabled a more precise estimation of cardiovascular risk.

Funding

None declared.

Data availability

No new data were generated or analysed in support of this research.

Disclaimer

This paper was originally published in the Italian language as ‘Prima guarda con la TAC e poi cura: Lo Scot-Heart study compie 10 anni’, in the Volume degli Atti del Congresso “Conoscere e Cuare il Cuore 2026”, published by Centro per la Lotta contro l'Infarto for distribution at the CCC Conference. This paper was translated by Dr. Mario Albertucci, representative of the CLI Foundation, and republished with permission.