Hunting for a coronary artery disease diagnosis in asymptomatic patients with diabetes mellitus: if, how and when | Cardiovascular Diabetology

Cardiovascular disease (CVD) is one of the primary causes of morbidity and mortality in individuals with T2DM [1]. It is associated with accelerated atherosclerosis, including atherosclerotic coronary artery disease (ACAD), carotid artery disease, and lower extremity artery disease (LEAD) [2]. To mitigate the progression of cardiovascular disease (CVD) in type 2 diabetes mellitus (T2DM) patients, it is essential to manage glucose levels effectively, alongside controlling atherosclerotic risk factors through lifestyle changes and pharmacological interventions [2]. Moreover, the timely detection of CVD, including ACAD, LEAD, and carotid artery disease, is critical for individuals with T2DM [3].

The PADDIA/CADDIA survey [4], an European medical research collaboration, evaluated the current management and screening practices for ACAD and peripheral artery disease (PAD) in individuals with T2DM across Europe, using the 2013 ESC/EASD (European Society of Cardiology/European Association for the Study of Diabetes) guidelines [5] as a benchmark.

One major point of debate of the survey was when physicians should screen for ACAD, carotid artery disease, or LEAD. Among the six potential answers for each question, the majority of physicians (58% to 79%) reported that they screen for CVD only when symptoms suggestive of CVD are present. A smaller proportion of physicians (29%, 32%, and 37%) would consider screening all patients with T2DM for ACAD, carotid artery disease, and LEAD, respectively [4].

The effectiveness and methods for detecting silent ACAD are still subjects of debate.

The aim of our review is to shed light on the rationale, if any, for the screening of asymptomatic ACAD and to determine the most effective method to use.

ACAD in diabetic patients: prevalence, clinical characteristics and prognosis

A meta-analysis by the Emerging Risk Factors Collaboration demonstrated that diabetes confers about a two-fold excess risk for a wide range of vascular diseases, independently from other conventional risk factors [6].

A population study showed that cardiovascular death risks were similar between men without diabetes mellitus but a prior myocardial infarction (MI) and men with diabetes mellitus but no prior MI (hazard ratio, 2.44 and 2.42 in the 2 groups, respectively; P=0.60), and in women, cardiovascular death was slightly, albeit significantly, higher in those without diabetes mellitus but with a prior MI (hazard ratio, 2.62 and 2.45 in the 2 groups; P<0.001) [7].

Insulin resistance develops before pre-diabetes and diabetes, progressively worsening over time, while hyperglycemia emerges during pre-diabetes and escalates with diabetes. Insulin resistance, along with impaired insulin signaling, hyperinsulinemia, and hyperglycemia, promotes elevated free fatty acids, advanced glycation end products, PKC activation, oxidative stress, mitochondrial dysfunction, and epigenetic changes [8]. These collectively drive endothelial dysfunction and inflammation, activating vascular smooth muscle cells (VSMCs), endothelial cells, and monocytes [8]. In diabetes, oxidized LDL accumulates in vulnerable vessel areas. Monocytes infiltrate the intima, ingest lipids, and become foam cells, releasing inflammatory cytokines such as TNF-α and interleukins. Stress responses like ER stress and inflammasome activation further promote macrophage and VSMC proliferation, migration, and dedifferentiation. VSMCs secrete collagen to form fibrous caps, stabilizing plaques, but inward remodeling narrows arteries.[8]. Advanced plaques may become unstable as fibrous caps thin and macrophage clearance fails, forming necrotic cores that exacerbate inflammation and thrombosis. Unstable lesions can rupture, causing acute thrombus formation, platelet aggregation, intraplaque hemorrhage, and vascular occlusion.[8].

In a meta-regression analysis, every 1% reduction in HbA1c was associated with 26% and 35% decreases in the logarithm of hazard ratio (HR) of major adverse cardiovascular events (MACE) (P = .044; R² = 0.65). The magnitude of HbA1c reduction can be a surrogate for the cardiovascular benefits of treatment with antidiabetic treatment [9].

In the ADVANCE trial the joint effects of routine blood pressure lowering and intensive glucose control reduced the relative risk of cardiovascular death by 24% versus standard therapy or placebo [10].

The Strong Heart Study showed that patient with higher quartile of LDL in 2034 subjects had higher HR for fatal and non fatal CVD [11]. In the STENO-2 trial [12] the composite endpoint of death from CV causes, nonfatal MI, coronary artery bypass graft (CABG), percutaneous coronary intervention (PCI), stroke, amputation, or surgery for PAD was lower (HR=0.47, 95% CI, 0.24-0.72; p=0.008) in patients in the arm with intensive combined therapy versus conventional therapy.

A pooled analysis harmonized individual-level data from 112 cohort studies conducted in 34 countries to assess the effects of the traditional risk factors (body mass index (BMI), current smoking, systolic blood pressure, non-high density lipoprotein (HDL) cholesterol, diabetes) on the 10-years incidence of cardiovascular disease. Traditional risk factors explained the 57.2% of cardiovascular disease in women and 52.6% in men; therefore 42.8% and 47.4% were the residual risk in women and men respectively [13].

The CONFIRM multicenter Registry showed that the presence of DM is associated with increased prevalence, extent and severity of ACAD and with a worse outcome [14].

In a large evaluation with CT angiography, diabetic patients displayed a significant plaque burden, predominantly concentrated in the left anterior descending (LAD) artery and the proximal segments of each coronary vessel [15]. Plaque composition analysis showed a notably high proportion of calcified plaques. Additionally, the prevalence of obstructive stenosis was as prevalent as non-obstructive stenosis [15]. Seventy-five percent of diabetic patients had multi-vessel disease, with plaques affecting various coronary segments, suggesting that ACAD in symptomatic diabetic patients was extensive [15].

The rate of future adverse events was significantly higher in patients with coronary plaques than in those with a normal CT scan [16]. This is due to the potential of all plaque types to lead to acute or chronic obstructive stenosis [15]. Non-calcified plaques, which are unstable and particularly prone to rupture, were frequently observed in patients with diabetes and acute coronary artery syndrome.

A recent study revealed that patients with diabetes are more likely to present with atypical chest pain, which increases the likelihood of delays in both diagnosis and treatment [17].

Silent myocardial ischemia is common and tends to be more severe, frequent, and prolonged in individuals with diabetes.

One study found that 8.2% of first myocardial infarction were silent, and these individuals had more than a three-fold higher risk of mortality and MACE [18]. Another study showed a prevalence of silent myocardial infarction of 28.5% in diabetic patients and 21.5% in non-diabetic patients, with a mean infarct size representing 11.8% of the left ventricle [19].

In a multi-ethnic, community-based study in the US (MESA study), using cardiac magnetic resonance imaging (CMR), the prevalence of myocardial scarring suggestive of previous myocardial infarction was 7.9%, with a mean age of 68 years. Interestingly, 78% of these cases were undetected on electrocardiograms (ECG) and clinical assessments, and none had a history of acute coronary syndrome (ACS) [20]. Silent myocardial infarction is frequently under-suspected and underdiagnosed, emphasizing the need for heightened awareness, especially in high-risk individuals. Early diagnosis is essential to prevent complications and reduce the risk of premature mortality in the long term.

The clinical definition of Primary Prevention is to prevent the onset of ACAD in individuals who have not yet been diagnosed with the disease and the clinical definition of Secondary Prevention consists in to prevent the progression of established ACAD, or occurrence/recurrence of events in individuals who have already been diagnosed with the disease [21].

Cardiovascular risk stratification: risk scores and modifiers

Recent evidence suggests that the risk of ACAD in T2DM is not universally the same as in patients with a history of CV disease, but rather is highly variable. A meta-analysis of 13 epidemiological studies, including 45,108 patients with and without diabetes, found that in T2DM patients without prior ACAD, the risk of ACAD was 43% lower compared to individuals without diabetes but with a history of myocardial infarction [22]. In a large population-based cohort [23], which included 1,586,061 adults aged 30 to 90 years, followed for 10 years, the ACAD risk was significantly lower in T2DM patients without ACAD compared to patients with ACAD but without diabetes: HR: 1.70 (95% CI 1.66–1.74) versus 2.80 (95% CI 2.70–2.85). Therefore, it is probable that a group with a reduced risk of ACAD exists within T2DM, particularly among patients under 40 years of age and those with a shorter duration of the disease.

Categorizing diabetes into distinct CV risk groups enables the identification of individuals who could benefit from more aggressive CV prevention. As a result, it may be beneficial to develop targeted strategies for detecting and intensively treating patients at higher risk, while it could be both reasonable and cost-effective to apply more moderate therapies to those with lower cardiovascular risk.

Risk calculators assess overall CV risk by weighing independent risk factors in a mathematical formula, generating a score based on absolute risk for various outcomes. This allows clinicians to estimate a patient’s individual risk and make treatment decisions accordingly. Currently, there are at least 110 different cardiovascular risk score calculators, with 45 specifically designed for patients with diabetes [24].

Owing to discrepancies in databases, varying combinations of CVD outcomes, and differing mathematical models, considerable variability exists. It is essential to keep in mind that the validation of these scores is confined to the specific characteristics of the population in which they were tested.

The 2023 ESC Guidelines for managing CVD [5] in patients with diabetes recommend using the SCORE2-Diabetes algorithm to assess 10-year CVD risk in patients aged ≥40 years with Type 2 diabetes who do not have a history of ACAD or severe target organ damage (TOD). The SCORE2-Diabetes algorithm combines traditional CVD risk factors (such as age, smoking status, systolic blood pressure, and total and HDL cholesterol) with diabetes-specific factors (including age at diabetes diagnosis, HbA1c levels, and estimated glomerular filtration rate). This model is calibrated for four risk clusters (low, moderate, high, and very high CVD risk) based on the methodology used in the SCORE2 and SCORE2-OP algorithms and has been validated in a cohort of 217,036 individuals. A personalized approach with this algorithm offers a more accurate and practical risk estimation, supporting the effectiveness and long-term sustainability of CVD prevention strategies and guiding treatment decisions.

Compared to existing risk scores, SCORE2-Diabetes, offers several advantages that should improve the allocation of preventive interventions. First, SCORE2-Diabetes has been systematically recalibrated for four distinct European regions, each defined by different levels of CVD risk, using the most up-to-date and representative CVD rates available [25]; second, SCORE2-Diabetes also demonstrates strong discrimination and provides individual risk estimates for people with T2DM. It accounts for specific risk factors such as the age of diabetes onset, HbA1c levels, and kidney function.

Lifetime risk is defined as the probability that a specific event will occur at some point during an individual’s lifetime.

Risk estimation is essential for identifying patients who could benefit from adopting healthy behaviors and receiving pharmacological treatment to reduce the risk of adverse clinical outcomes and future CV events. The burden of risk factors, competing risks, and the duration of treatment determine the benefit derived from preventive therapy. The 2021 ESC Guidelines on CVD prevention [2] recommend utilizing lifetime risk prediction models to guide decisions on intensified preventive treatments in adults with T2DM. The recalibrated DIAbetes Lifetime perspective 2 (DIAL2) model, an updated version of the DIAL model, offers a valuable tool for predicting CVD-free life expectancy and lifetime CVD risk for individuals with T2DM but no prior CVD, in European low- and moderate-risk regions, aged 30–85 years. The DIAL2 model presents several advantages and increased clinical relevance compared to the earlier DIAL model. Ten-year predictions from SCORE2-Diabetes and lifetime predictions from DIAL2 can be used together, allowing for personalized patient management [26].

Despite extensive public health and clinical initiatives for primary CVD prevention, a significant proportion of individuals at ‘high’ and ‘very high’ risk remain unidentified and untreated.

In addition to the conventional CVD risk factors included in risk scores, other risk factors or types of individual information can also influence the calculated risk [27]. The assessment of potential risk modifiers is especially important when an individual’s risk is near a decision threshold. In low-risk or very-high-risk situations, additional information is less likely to influence management decisions. However, the number of individuals in this ‘grey zone’ is substantial [27]. Caution should be exercised to ensure that risk modifiers are not used solely to increase risk estimates when the modifier profile is unfavorable, but also to decrease risk estimates when the profile is favorable [27]. Risk modifiers are considered by ESC 2021 Guidelines:

• Psychosocial factors, a recent prospective cohort study with a median follow-up of 8.4 years reported positive effects of depression screening on reducing major ACAD events [28];

• Ethnicity, current risk scores may either under- or overestimate CVD risk in different ethnic minority groups;

• Frailty is a functional risk factor for both CV and non-CV morbidity and mortality. Frailty assessment is not used to determine eligibility for specific treatments, but rather to create an individualized care plan with established priorities.

  • Family history should be routinely assessed, and a positive family history of premature ASCVD should prompt a comprehensive CVD risk evaluation.

  • Socioeconomic determinants, the development and prognosis of ACAD are influenced by social gradients.

  • Environmental exposure, Air pollution has a strong association with ACAD.

• Body composition, the associations between BMI, waist circumference, and waist-to-hip ratio and CVD are maintained after adjustment for conventional risk factors,.

  • Arterial stiffness is generally measured using aortic pulse wave velocity or the arterial augmentation index. While studies suggest that arterial stiffness can predict future CVD risk and improve risk classification, difficulties in measurement and notable publication bias limit its widespread application [27].

  • Ankle Brachial Index

• Estimates suggest that 12–27% of middle-aged individuals have an ankle brachial index (ABI) < 0.9, with 50–89% of them not showing typical claudication symptoms. A meta-analysis of individual patient data found that the reclassification potential of ABI is limited, although it may hold more value for women at intermediate risk [27].

• Carotid ultrasound, the routine use of intima-media thickness (IMT) for risk assessment is not recommended due to the lack of standardized methodology and its limited additional value in predicting future CV events, even in individuals at intermediate risk. Plaque is defined as focal wall thickening greater than 50% of the surrounding vessel wall or a focal region with an IMT measurement exceeding 1.5 mm that protrudes into the lumen. Carotid artery plaque assessment via ultrasonography may still help reclassify CV disease risk and could serve as a risk modifier for intermediate-risk patients [27].

• To enable early intervention, a more precise identification of high-risk patients is essential.

• Chronic Inflammatory Diseases, such as rheumatoid arthritis, systemic lupus erythematosus, psoriasis, inflammatory bowel diseases, are increasingly recognized as important contributors to CV risk. In patients with diabetes, the coexistence of chronic inflammation can further accelerate atherosclerosis and increase the risk of ACAD. Epidemiological studies have shown that systemic inflammation is associated with endothelial dysfunction, plaque instability, and higher rates of CV events, independent of traditional risk factors [29, 30]. Including chronic inflammatory disorders as non-traditional risk enhancers may help refine risk stratification in diabetic patients and support individualized screening strategies. Recognizing these conditions as modifiers of CV risk underscores the need for tailored approaches, potentially justifying more intensive monitoring or earlier imaging in selected high-risk subgroups.

The value (if any) of serum biomarkers for CV risk prediction

Along with traditional risk factors, emerging non-traditional markers—such as apolipoproteins A and B, hs-CRP, brain natriuretic peptides, troponin I, homocysteine, interleukins 1 and 6, Lp(a), cholesterol remnants, LDL particle size and number, TNF-α, and uric acid—can inform personalized cardiovascular risk assessment. Among these, hs-CRP stands out as a key inflammatory biomarker, providing prognostic information beyond traditional factors. In type 2 diabetes, hs-CRP is particularly relevant due to the heightened systemic inflammation from chronic hyperglycemia, insulin resistance, and endothelial dysfunction. Elevated hs-CRP levels in diabetic patients correlate with the presence, severity, and progression of subclinical atherosclerosis, including coronary artery disease. [31, 32]. Elevated hs-CRP levels have been independently associated with CAC, impaired coronary flow reserve, and increased carotid IMT, suggesting a link between low-grade inflammation and early vascular remodeling [31]. Furthermore, prospective cohort studies have shown that hs-CRP levels above 3 mg/L are predictive of future cardiovascular events in diabetic patients, even when LDL cholesterol levels are well controlled [34].

In the diabetic population, the addition of hs-CRP to conventional risk models, has been shown to significantly improve risk discrimination and reclassification, especially in those with borderline or intermediate risk profiles [33].

Nevertheless, the use of hs-CRP in clinical decision-making is not without limitations. Its levels can be influenced by numerous non-cardiovascular conditions such as infections, trauma, or chronic inflammatory diseases. Hence, careful interpretation in the appropriate clinical context is essential. Despite these caveats, the overall evidence supports the role of hs-CRP as a valuable adjunct biomarker for cardiovascular risk stratification in asymptomatic diabetic patients, with the potential to guide individualized preventive strategies and improve long-term outcomes [33].

Epidemiological and genetic studies support a causal and continuous relationship between Lp(a) levels and CV outcomes across various ethnic groups. As recommended by the European Atherosclerosis Society consensus statement [35], Lp(a) should be measured at least once in adults, with the results interpreted in the context of the patient’s overall CV risk, and recommendations made for intensified early risk factor management through lifestyle changes.

In secondary analyses of the Women’s Health Study (WHS) [36] and the Aspirin in Reducing Events in the Elderly (ASPREE) trial [37], aspirin use was found to significantly reduce cardiovascular events in individuals with genetic polymorphisms associated with elevated Lp(a) levels.

In the Multi-Ethnic Study of Atherosclerosis (MESA) trial [38], participants were followed for a mean of 13.4 years and a total of 315 CHD events occurred; patients with high lpa (Lp(a) ≥ 50 mg/dL) levels exhibited significantly increased risk of coronary heart disease (CHD) events compared patients with low levels of lpa, regardless of baseline LDL-C. These findings suggest that more aggressive preventive measures may be appropriate for individuals with higher lpa levels.

Recently, a new risk calculator was introduced that incorporates Lp(a) values alongside traditional cardiovascular risk factors. However, further research is needed to fully assess the role of Lp(a) in improving cardiovascular risk prediction. The model is available online for both patients and clinicians ( and can be used to support clinical discussions and the implementation of preventive measures in everyday practice.

Platelets play a key role in atherothrombosis. Thromboxane (TX) A2, a transient lipid mediator released by activated platelets, facilitates platelet aggregation and vasoconstriction. It is involved in multiple pathophysiological processes, including primary hemostasis, the progression of atherosclerosis, thrombosis and inflammation [39].

The enzymatic metabolism of TXA₂ and its stable hydrolysis product, TXB₂, produces a stable end product, 11-dehydro-TXB₂ (TXM), which is measurable in urine and serves as an indicator of the overall rate of TXA₂ biosynthesis in the body [40]. Therefore, urinary TXM (U-TXM) excretion acts as a non-invasive indicator of in vivo platelet activation. Elevated U-TXM excretion has been associated with major CVR factors that accelerate atherogenesis, such as diabetes mellitus [41, 42]. TXM could thus be an important biomarker for assessing the impact of cardioprotective interventions on platelet activation, such as lipid-lowering drugs, improved glycemic control, weight loss, and antiplatelet therapies [43, 44]. TXM levels were significantly elevated in heart failure (HF) patients compared to those with ischemic heart disease (IHD) without HF [43]; the rate of TXM excretion was, at least in part, linked to disease severity, as indicated by New York Heart Association (NYHA) classifications [43]. Obese women show significantly increased excretion of TXM and higher circulating levels of C-reactive protein compared to healthy controls [45]; short term weight loss program resulted in a significant decrease in TXM [46].

In patients treated with aspirin, TXM predicts the future risk of MI or CV death [47], and is an externally valid and potentially modifiable determinant of stroke, MI, CV death in patients at risk for atherothrombotic events [48]. TXM generation is an independent predictor for all-cause and CV mortality, regardless of ASA use and its measurement could be useful for guiding therapy adjustments, particularly in individuals without CVD [49]. TXM is strongly linked to increased risk of developing HF, including both HF subtypes, even after adjusting for traditional risk factors [50]. Petrucci et al [51] showed a log-linearly association between TXM and risk of future serious vascular events or revascularizations in 5948 people with type 1 or 2 diabetes and no cardiovascular disease, in the ASCEND trial. The practical clinical application is the potential use of U-TXM as a biomarker to personalize and optimize treatment dosing. Recent work by Brambilla et al [52] has further emphasized the clinical relevance of soluble platelet activation biomarkers, underscoring their potential role in identifying high-risk individuals and guiding tailored preventive strategies. These findings suggest that soluble platelet-derived biomarkers could be integrated into risk stratification models and therapy monitoring. However, their application in routine clinical practice remains limited by issues of assay standardization, availability, and the need for validation in large-scale prospective studies. Further research is required before these markers can be widely implemented as part of personalized prevention strategies in asymptomatic diabetic patients.

Subclinical atherosclerosis as a tool to refine CV stratification

While risk scores and biomarkers reflect a probabilistic estimate of the likelihood of developing CVD, imaging tests for atherosclerosis seek to measure disease directly in its subclinical stage. As such, imaging tests for subclinical atherosclerosis can integrate all of a patient’s risk determinants—ranging from genetic to environmental to modifiable—into a single measure of disease and vulnerability (tissue impact) in the vascular bed of interest. Decades have studies have shown discordance between risk scores and subclinical disease burden, as some patients with no risk factors can have advanced disease, and some patients with multiple risk factors can live to old age without any detectable plaque. Across all risk scores ever tested, adding imaging tests for subclinical atherosclerosis substantial improve risk discrimination. Imaging tests for subclinical atherosclerosis tend to perform best in lower risk patients with risk modifiers or in intermediate risk patients where treatment decisions remain uncertain. The CAC score, which provides a measure of coronary plaque using non-contrast cardiac-gated CT, is the most researched tool for CV risk stratification and receives a Class IIA recommendation from many guidelines. Beyond its prognostic value, CAC scoring has several practical clinical applications. CAC = 0 is associated with very low short-to-intermediate term risk, and in such cases, it may be reasonable to defer pharmacologic interventions, focusing instead on lifestyle optimization, unless other high-risk features are present. A CAC score between 1 and 99 indicates subclinical atherosclerosis and may justify initiating or intensifying statin therapy in patients with borderline or intermediate estimated risk. Scores between 100 and 299 identify individuals at high risk, for whom guideline-directed therapies, including high-intensity statins, aggressive modifiable risk-factor control, and possibly aspirin in selected cases, should be strongly considered. CAC ≥ 300 reflects a very high-risk profile, akin to patients with established CAD, warranting comprehensive secondary-prevention-level therapy and closer clinical follow-up [53].

In the 2025 ESC/EAS focused update on dyslipidaemias [54], an increased CAC score is explicitly recognised as a risk modifier in individuals at moderate cardiovascular risk or those near treatment decision thresholds, helping to improve risk stratification (Class IIa, Level of evidence B).

However, other imaging tools are also effect. Carotid ultrasound, particularly when interpreted as a carotid plaque score, is a strong predictor of risk. Total plaque burden by coronary CTA is also an emerging tool in asymptomatic patients, particularly with new protocols that decrease ionizing radiation and AI-tools that automate volumetric quantification of non-calcified and calcified total coronary plaque burden.

Primary vs secondary prevention as opposed to the concept of cardiovascular risk continuum

The traditional approach of categorizing patients with ACAD as uniformly high risk for future events is evolving. While these patients have been historically placed in a “clinical high-risk” category, it overlooks the fact that the process of atherosclerosis itself is not binary but rather a continuous spectrum. Some individuals may have very early-stage disease, while others may be dealing with advanced stages of the disease when they first experience a clinical event, such as a heart attack or stroke.

Historically, patients with T2DM have been categorized based on CV risk into those needing primary prevention (those without ACAD) and those requiring secondary prevention (those with ACAD affecting any vascular district). However, this classification is both misleading and arbitrary, as not all patients needing secondary prevention have the same risk for such events, and the same applies to those needing primary prevention (i.e., CV risk can vary from moderate to very high). Furthermore, in some cases, the distinction between primary and secondary prevention is not based on symptoms but rather on the results of additional diagnostic tests most notably imaging [55] (Figure 1).

We should reconsider the long-standing belief that there is a distinct and clear-cut separation between primary and secondary prevention.

The overlap in risk between primary and secondary prevention has also been highlighted. Peng et al. [56] compared the annualized composite event rate, including cardiovascular death, MI, and stroke, in the placebo group of the FOURIER trial with that of 6,814 participants from the MESA. They found that individuals with very high CAC scores (≥1000) form a distinct group at markedly higher risk for cardiovascular events, non-cardiovascular outcomes, and mortality compared to those with lower CAC scores. Their 3-point major adverse cardiovascular event rates are comparable to those observed in a stable, treated secondary prevention cohort. Recent data suggest that a CAC score >300 is equivalent to having a severe stenosis detected by coronary CTA.

More than fifty percent of patients suffer on ischemic event with a high plaque burden but not stenosis [57]. The recognition of the broad spectrum of ACAD has led to the understanding that the concepts of “primary” and “secondary” prevention are outdated [58]. The concept of “advanced subclinical atherosclerosis” is emerging as a new and distinct clinical group, positioned between the traditional categories of primary and secondary prevention [57, 59].

Patients receiving treatment have a lower risk of a subsequent event compared to those with a high plaque burden but no prior events (i.e., they may be at lower risk than those with advanced subclinical atherosclerosis). However, some patients may still progress to a “very high risk” status [57].

Despite the seemingly logical nature of this concept, we have continued to depend on older classifications, mainly due to the challenge of accurately assessing ACAD risk across its full spectrum.

In essence, we need a more nuanced understanding of atherosclerosis, one that recognizes the disease as a continuum, helping clinicians provide more individualized and effective care for patients with ACAD. This may also involve continuous follow-up to monitor disease progression and tailor interventions accordingly [57].

The growing use of new glucose-lowering medications, such as sodium/glucose co-transporter-2 (SGLT2) inhibitors and glucagon-like peptide-1 (GLP-1) receptor agonists, has further sparked interest in diagnosing CVD in diabetic patients to prevent complications and adverse events as early as possible [60].

Symptomatic vs asymptomatic patients: the issue of silent ischemia, or unrecognized CVD

The ischemic cascade starts with changes in the coronary artery (noncalcified and calcified plaque within the vessel wall). As long as the luminal narrowing of the artery does not exceed 50%, no limitation of flow is expected. When the narrowing of the artery exceeds 50%, reduced myocardial blood flow with decreased myocardial perfusion can occur. The decrease of myocardial perfusion results in diastolic and systolic dysfunction of the left ventricle. After this stage, ECG changes and chest pain can often be observed [61].

Asymptomatic ACAD includes: atherosclerotic lesions that do not cause luminal narrowing to and those obstructive lesions that can produce ischemia during stress testing, but may not be appreciated clinically by the patient. Unrecognized and recognized myocardial infarctions have similar prognosis [62]. Patients with unrecognized myocardial infarction compared with recognized myocardial infarction patients were less likely to receive guideline-directed medical therapies [62].

Diabetic heart disease is characterized by a rapid progression, a multivessel involvement, more vulnerable features, 86% higher rates of in-stent restenosis and cardiac autonomic neuropathy [63]. In patient with previous PCI or CABG and in patient with any previous coronary revascularization 16% were asymptomatic [64]. The prevalence of asymptomatic CAD or silent MI depends on several factors: 1) the sensitivity and specificity of the test performed, with the performance to detect ACAD being better for stress Single Photon Emission Computed Tomography (SPECT) and stress echocardiography than for ECG stress tests. Also, the positive predictive value for ACAD detection is higher when two tests are concordant; 2) the number of CV risk factors, which changes the pretest probability and the positive and negative predictive values; 3) the presence of target organ damage, such as retinopathy, nephropathy, cardiac autonomic neuropathy and PAD are associated with higher prevalence of silent MI and ACAD; 4) the influence of care management with a large proportion of patients with silent MI demonstrating resolution of ischaemia 5) upon repeat stress imaging in relation with more intensive control on CV risk factors.

Prevalence of silent CAD vary from 7 to 46.3% if assessed with exercise test or with significant CAC score [65].

Following coronary artery occlusion, mechanical and electrocardiographic abnormalities in the left ventricle often occur prior to the emergence of symptoms [66, 67]. Therefore, it is not surprising that patients may demonstrate signs of myocardial ischemia without presenting symptoms. Silent ischemia is generally defined as the presence of objective evidence of myocardial ischemia in individuals who do not experience symptoms associated with that ischemia. Silent ischemia can be detected in patients who remain asymptomatic during exercise or pharmacological stress testing but show transient ST-segment deviations, perfusion abnormalities, or reversible changes in regional wall motion [68].

Diabetic neuropathy, which leads to abnormalities in pain perception, is generally regarded as the main factor contributing to asymptomatic ACAD in diabetic patients. However, the precise mechanisms behind this condition remain incompletely understood [68].

Proposed pathophysiological mechanisms include diminished sensory inputs from ischemic myocardium, altered central processing of cardiac pain through sympathetic afferent fibers [69], and sympathetic denervation [70]. Furthermore, abnormalities in cytokine production are thought to play a role in impaired nociception [71]. A robust and significant correlation between cardiac autonomic neuropathy and silent myocardial ischemia has long been established [72].

Changes in myocardial oxygen demand or supply have been proposed as possible mechanisms underlying silent ischemia. Several studies have indicated a demand-related mechanism of ischemia by observing an increase in heart rate before the onset of silent ischemic events [73].

Furthermore, blood pressure has been shown to increase in the minutes preceding silent ischemic episodes [74]. Other studies have suggested that demand alone does not fully explain silent ischemia. For example, the heart rate at which asymptomatic ST-segment depression occurs during ambulatory ECG (AECG) monitoring is significantly lower than the heart rate at which ST-segment depression begins during exercise testing in the same patient [75]. The interplay between increased demand and impaired supply, caused by abnormal microvascular and endothelial responses, offers a possible explanation for the mechanism of silent ischemia. Asymptomatic myocardial ischemia is observed more commonly than symptomatic ischemia in patients with ACAD [76].

Through AECG monitoring, nearly half of patients with stable ACAD are found to exhibit transient ST-segment depressions, which are likely indicative of silent ischemic events [77].

In a study, 12% of noninsulin-dependent subjects with diabetes with no symptoms suggestive of ACAD showed abnormal exercise ECGs [72], although only half of these patients exhibited perfusion defects during thallium scintigraphy. Another study found that 33% of diabetic patients with at least one additional CV risk factor experienced silent ischemia [78].

In patients with mild-to-moderate ACAD, silent ischemia provides prognostic information for adverse outcomes similar to that of symptomatic ischemia. In a study of patients with medically managed ACAD, the risk of death or myocardial infarction over 7 years of follow-up was comparable between those with asymptomatic ST-segment depression during exercise and those with symptomatic ST-segment depression. However, in patients with extensive ACAD receiving medical treatment, silent ischemia is associated with a worse prognosis than symptomatic ischemia [79].

“Unrecognised” diabetic cardiac impairment is characterized by atypical symptoms like “silent” ischemia, “silent” myocardial dysfunction, cardiac arrhythmia and alternatively presenting as diastolic dyfunction [80].

Given the wide range of terms used to describe unrecognized CVD in T2D, it is not surprising that this phenomenon is not widely recognized or fully acknowledged. Establishing a consensus on the terminology would help ensure that no important data related to this condition are missed. It is suggested that ‘silent’ or ‘asymptomatic’ CVD in individuals with T2D be renamed ‘unrecognized diabetic cardiac impairment’ (UDCI). This term underscores the hidden nature of the condition and emphasizes the relationship between atypical symptoms and diabetes [80].

In the multicenter SPINS (stress cardiac magnetic resonance CMR Perfusion Imaging in the United States) study [62], 2,349 consecutive patients (mean age 63 ± 11 years, 53% male) with suspected ACAD underwent stress CMR imaging and were followed for a median of 5.4 years. Unrecognized myocardial infarction (UMI) was defined as the presence of late gadolinium enhancement consistent with MI in the absence of any previous medical history of MI. In comparison to patients with recognized myocardial infarction (RMI), those with UMI had a comparable burden of CV risk factors but exhibited significantly lower rates of guideline-directed medical therapies, including aspirin, statins, and beta-blockers [62].

Early screening for silent ACAD could be a promising strategy to reduce CV complications and mortality in diabetic patients. Silent ACAD, naturally, goes unrecognized unless detected through diagnostic tools, leading to suboptimal use of guideline-directed medical therapies. However, whether screening diabetic patients for silent ACAD is truly beneficial remains a significant and unresolved question.

Main non-invasive test for the screening of silent ACAD

Functional, non-invasive tests for screening silent ACAD include the ECG exercise tolerance test, exercise/pharmacological stress echocardiography, and exercise/pharmacological myocardial perfusion imaging, while anatomical tests involve coronary artery calcium scoring, CT angiography, cardiac magnetic resonance imaging, and positron emission tomography [81]. Choosing the appropriate screening test can improve the accuracy of disease prediction. The characteristics of the main non-invasive test, along with the prevalence of silent ACAD and associated prognosis, are shown in Table 1.

The concept of plaque burden as opposed to coronary ischemia

Over the past two decades, substantial evidence has underscored the prognostic significance of coronary atherosclerotic plaque independent of stenosis severity [89]. Total plaque burden, the presence of high-risk morphological features, and the tendency toward plaque progression have consistently emerged as robust predictors of future CV events [89]. Importantly, non-obstructive plaques are markedly more prevalent than hemodynamically significant stenoses and are associated with a greater overall population-attributable risk [89].

The Lancet Commission on Rethinking Coronary Artery Disease [90] introduces a timely and forward-thinking shift in clinical perspective—moving the emphasis from ischaemia to atheroma. By reclassifying the condition as ACAD, it highlights the need to address the underlying pathology—atherosclerosis—rather than focusing solely on flow-limiting lesions, which have traditionally dominated clinical management [90]. This redefinition encourages earlier identification and intervention, aiming to prevent disease progression and reduce long-term morbidity and mortality. By reframing coronary artery disease in this way, the Commission advocates for a preventive, population-level approach with the potential to reshape clinical practice and improve CV outcomes on a broader scale [91].

Contemporary diagnostic approaches for suspected ACAD largely centre on the detection of flow-limiting lesions, employing either functional testing for ischaemia or anatomical imaging to assess stenosis. A paradigm shift toward earlier detection would reorient the primary objective of testing—from the identification of obstructive disease alone to a more comprehensive assessment of an individual’s future CV risk [91].

Rationale for ACAD Screening in asymptomatic patients with diabetes

ACAD screening in asymptomatic patients with diabetes should be indicated when a number of assumptions is satisfied:

1. 1.

   Silent, unrecognized ACAD is prevalent among subjects with diabetes.

2. 2.

   Unrecognized ACAD or non-obstructive atherosclerosis is associated with poor prognosis

3. 3.

   Imaging techniques are able to accurately rule-in or rule-out ACAD in the asymptomatic patient with diabetes

4. 4.

   Coronary revascularization and/or intensification of medical therapy based on routine ACAD screening can improve outcomes of asymptomatic diabetic patients

Is unrecognized ACAD prevalent among subjects with diabetes?

As mentioned above, the prevalence of silent myocardial ischemia (SMI) in individuals with diabetes has been reported to vary between 6 [72] and 35% [92] (Table). Among those with SMI, the proportion of patients with significant ACAD identified via angiography ranges from 35 [93] to 90% [94]. Ischemia can be caused not only by ACAD but also by functional abnormalities such as endothelial dysfunction [95], disturbed microcirculation, and impaired coronary reserve [81] The phenomenon of less symptomatic myocardial ischemia, often observed in diabetic patients, is linked to a poorer prognosis[96, 97].

Several factors influence the prevalence of SMI [98]:

1. 1.

   The sensitivity and specificity of the diagnostic test used. Techniques like stress SPECT and stress echocardiography have higher accuracy in detecting ACAD compared to ECG stress tests [99]. Furthermore, the positive predictive value for ACAD detection improves when two diagnostic tests yield consistent results [100].

2. 2.

   The number of CV risk factors, which affects the pretest probability and, in turn, impacts both the positive and negative predictive values of the test [101].

3. 3.

   The presence of target organ damage is a key factor in the prevalence of SMI. Conditions like retinopathy [101], nephropathy [102], cardiac autonomic neuropathy (CAN)[103], and peripheral artery disease [101] are all strongly associated with an increased likelihood of both SMI and ACAD. In addition, elevated CAC scores and hypokinetic changes on echocardiography have also been linked to a higher prevalence of SMI.

4. 4.

   Care management strategies also play a significant role in the prevalence of SMI. Many patients with SMI experience a resolution of ischemia on follow-up stress imaging, particularly when there is more intensive control of cardiovascular risk factors [87]. Over longer periods, the prevalence of SMI has shown significant decline, with research demonstrating a notable reduction over a span of 10 years [104].

Is unrecognized ACAD or non-obstructive atherosclerosis associated with poor prognosis?

Exercise stress test has a 7-8% prevalence of silent ischemia and abnormal heart rate recovery, but it does not provide additional prognostic information [81].

In a study involving 161 asymptomatic patients with T2DM and no previous history of ACAD, abnormalities in dobutamine stress echocardiography, observed in 28% of the cases, were predictive of future adverse CV events during a median follow-up of 5 years, providing additional prognostic information beyond clinical variables [105]. Similarly, SMI detected through exercise or dobutamine stress echocardiography was predictive of future major adverse cardiac and cerebrovascular events in another study of asymptomatic patients with T2DM [106].

Recent studies have investigated the role of cardiac MRI in identifying SMI. In a study involving 327 individuals at elevated risk for CV events but without symptoms or clinical ACAD, dobutamine stress-induced myocardial ischemia detected via cardiac MRI was found to be predictive of future CV events and survival, particularly in men [107].

Beyond inducible ischemia on stress echocardiography, coronary microvascular dysfunction, indicated by abnormal coronary flow reserve, may also serve as a marker for increased CV risk in asymptomatic, high-risk patients with diabetes [108]. It is important to highlight that, according to a study of patients with similar levels of ischemia on dobutamine stress echocardiography (DSE), those with SMI, as opposed to symptomatic ACAD, had a higher risk of death or MI, potentially due to being undertreated [109].

In a large-scale study involving 3,664 patients undergoing exercise Tc-99m sestamibi SPECT, the prevalence of SMI was around 21%, with 6% of patients showing high-risk ischemia, defined as ischemia affecting ≥7.5% of the myocardium [110]. In terms of prognostic value, a summed stress score of ≥9 was identified as an independent predictor of MACE in asymptomatic patients with T2DM [111].

In asymptomatic diabetic patients, anatomical plaque analysis using CCTA can help identify vulnerable plaques at risk of causing future acute events [112]. In the Scottish Computed Tomography of the Heart (SCOT-HEART) trial, which included 4,146 patients evaluated for chest pain, participants were randomized to either the usual care pathway or additional CCTA [83]. Among the CCTA group, 2.3% experienced a MI or died from coronary heart disease, compared to 3.9% in the usual care group. A more significant reduction in MI and mortality was observed in patients with diabetes. The improved outcomes were largely attributed to the initiation of appropriate medical treatment following the identification of obstructive or non-obstructive coronary lesions through CCTA.

In a risk-adjusted hazard analysis, both obstructive and non-obstructive ACAD in patients with diabetes were associated with a higher risk of all-cause mortality compared to those without atherosclerosis on coronary CTA [113]. Non-obstructive disease had a hazard ratio (HR) of 2.07 (95% CI: 1.33 to 3.24; p = 0.001), while obstructive disease showed a HR of 2.22 (95% CI: 1.47 to 3.36; p < 0.001)[113]. The presence of non-obstructive disease was linked to a particularly poor prognosis in diabetic patients. No significant survival differences were found between diabetic patients with non-obstructive ACAD and those with 1-vessel ACAD, 2-vessel ACAD, or 3-vessel ACAD [113]. These data strongly indicate that non-obstructive disease has substantial long-term prognostic significance in patients with diabetes. Both non-obstructive and obstructive ACAD, as identified by coronary CTA, are associated with an increased long-term risk of MACE and mortality in patients with diabetes [113].

Are imaging techniques able to accurately rule-in or rule-out ACAD in the asymptomatic patient with diabetes?

A contemporary fair answer to whether imaging techniques can accurately rule-in or rule-out ACAD in asymptomatic diabetic patients is complex and depends on the purpose of the imaging, the type of disease being assessed, and the stage of the disease. Coronary evaluation in the asymptomatic patient with diabetes begins with a redefined clinical goal: not the detection of inducible ischaemia, but the early identification of high-risk atherosclerotic disease that would otherwise remain silent [83]. In this population, atheroma develops earlier, distributes more diffusely, and progresses with fewer symptoms than in non-diabetic individuals. Most events are triggered by non-obstructive plaques, and classic angina is often absent. The findings of the ISCHEMIA trials further support this view: among diabetic participants with moderate-to-severe ischaemia, an initial invasive strategy offered no mortality benefit over optimal medical therapy [114]. These data do not negate the value of imaging, but rather clarify its purpose—to guide preventive pharmacology and long-term risk stratification rather than to routinely justify revascularisation. As a result, while imaging techniques can provide valuable insight into the extent of disease, they are best utilized not only to detect ACAD but to identify high-risk patients early, guide therapeutic decision-making and detect residual risk. Yet, the question remains as to how we can provide innovative care models aimed at personalized risk stratification.

Imaging plays a key role in the early identification of atherosclerotic burden, which is the key driver of cardiovascular events in diabetes. In asymptomatic patients, CAC scoring and coronary CT angiography (CCTA) are instrumental in revealing subclinical disease. A CAC score of zero has strong negative predictive value and can be used to effectively de-risk a patient, indicating a low short- to medium-term risk of events. Conversely, a CAC score above 300 signifies a risk profile comparable to that of a patient with prior myocardial infarction or arterial revascularization, and should prompt treatment escalation to secondary prevention targets counteracting the therapeutic inertia that commonly affect this population[116]. CCTA adds further granularity, allowing direct visualization of total plaque burden and detailed assessment of plaque morphology. It is highly effective in excluding obstructive ACAD, but also reveals high-risk phenotypes—such as low-attenuation plaques, positive remodeling, or napkin-ring signs—that carry prognostic significance even in the absence of significant stenosis. In contrast, functional imaging techniques can assess inducible regional wall motion abnormalities, myocardial perfusion and microvascular dysfunction, a particularly relevant consideration in diabetes, where the coronary microcirculation may be compromised [115,116,117,118]. Particularly, positron emission tomography (PET) and stress CMR perfusion imaging are important not only in identifying flow-limiting lesions but also quantifying myocardial blood flow [119, 120] and revealing clinically unrecognized silent myocardial infarction [18] (Figure 2).

While these imaging modalities offer significant diagnostic value, their primary role is not simply to “rule in” or “rule out” ACAD in a binary manner. Rather, they are tools for risk stratification, guiding decisions about treatment intensity and the need for more aggressive monitoring. Accordingly, the 2024 ESC guidelines recommend imaging as part of a structured diagnostic pathway, which begins with estimating the likelihood of ACAD based on clinical risk factors [121]. For individuals with intermediate probability, imaging such as CCTA or CAC scoring is used to confirm the presence or absence of disease and to assess plaque burden. For higher probabilities, functional imaging may be used to further interrogate lesions or assess coronary microvascular dysfunction.

Crucially, imaging is not intended to drive an increase in revascularisation rates for asymptomatic individuals, but rather to uncover disease at an earlier stage when preventive pharmacology can make the most significant impact [122]. This highlights the importance of early detection and aggressive prevention, rather than waiting for symptomatic disease or attempting to fix every lesion seen on imaging. Ultimately, the role of imaging extends beyond diagnosis. It serves as a means of redefining cardiovascular risk by visualising the burden and biological behaviour of coronary atherosclerosis—often diffuse, silent, and prognostically significant in this population. By identifying subclinical plaque, characterising high-risk morphologies, and detecting coronary microvascular dysfunction, imaging informs the intensity and timing of preventive strategies. It also helps refine decisions about when revascularisation may be warranted, aligning therapeutic choices with the underlying pathophysiology rather than symptoms alone. Within this framework, imaging is positioned not as a diagnostic endpoint, but as a strategic tool for preventive care, aiming to intercept disease trajectories before clinical events occur.

Furthermore, the role of imaging must be considered in terms of sustainability and the balance between the benefits of early detection and the risks of unnecessary repeat imaging. While coronary CT angiography and stress imaging provide critical information, excessive use of these tests—especially when prior studies have already revealed significant plaque burden—should be avoided. In younger patients, the cumulative radiation exposure from repeated imaging is a valid concern [123], and therefore a careful, evidence-based approach is necessary to prevent unnecessary follow-up imaging unless new symptoms develop or risk profiles change.

Despite technological advances and increasing endorsement in contemporary guidelines, the evidence base for imaging-guided prevention in asymptomatic diabetic patients remains incomplete. Much of current practice is inferred from broader cohorts or trials in which imaging was used for diagnostic rather than therapeutic guidance. What is lacking are prospective studies that directly test whether treatment intensification based on specific imaging findings—such as total plaque burden, high-risk plaque features on coronary CT angiography, or impaired myocardial flow reserve on PET or stress CMR—leads to improved outcomes beyond what is achievable with standard risk stratification alone[124, 125].

While coronary CT angiography has been proposed as a screening tool for ACAD in asymptomatic diabetic patients, several studies have raised concerns regarding its cost-effectiveness and the associated radiation exposure. The cost-effectiveness of routine ACAD screening using CT angiography in patients with T2DM remains a subject of debate. A study by Hayashino et al. reported an incremental cost-effectiveness ratio (ICER) of $31,400 per quality-adjusted life-year (QALY) for exercise electrocardiography compared to no screening in asymptomatic patients aged 60 years. This suggests that while screening may be beneficial, the costs associated with CT could be higher [126].

Furthermore, the SCADIAB study [127], which is currently emulating a randomized controlled trial using electronic health records, aims to evaluate the cost-effectiveness of routine ACAD screening in T2DM patients at very high cardiovascular risk. Preliminary findings indicate that widespread ACAD screening leads to significant healthcare costs due to invasive testing, revascularization procedures, and intensified pharmacological therapies. These costs may outweigh the clinical benefits, particularly in populations where the prevalence of ACAD is low.

CT involves exposure to ionizing radiation, which poses potential long-term health risks, including an increased risk of cancer. The radiation dose associated with CT can vary depending on factors such as equipment type and scanning protocols. For instance, a study by Kim et al. [128] reported that CAC scoring, resulted in radiation exposure ranging from 0.8 to 5 millisieverts (mSv). This exposure is comparable to the radiation dose received from several chest X-rays.

While CT offers high diagnostic accuracy in detecting ACAD, its routine use as a screening tool in asymptomatic diabetic patients should be carefully evaluated. The potential benefits must be weighed against the economic costs and the risks associated with radiation exposure.

There is an urgent need for trials designed to close this gap: to evaluate whether imaging-directed modification of lipid-lowering, anti-inflammatory, antithrombotic, or glucose-lowering therapies can meaningfully reduce cardiovascular events in this high-risk population. Such trials should also advance the recognition of specific imaging biomarkers not merely as prognostic tools, but as modifiable and targetable risk factors. If validated, parameters such as coronary plaque volume, vulnerability indices, or myocardial flow reserve could serve as surrogate endpoints in prevention trials, offering a dynamic measure of therapeutic efficacy. This would enable a shift from static risk prediction to biologically grounded, image-guided risk modification. At the same time, it would help define actionable thresholds, optimise resource allocation, and clarify the effectiveness of serial imaging in monitoring disease progression. Without this level of validation, imaging remains a powerful yet underutilised asset—one whose promise will only be fulfilled when its findings translate into measurable clinical benefit.

Can coronary revascularization and/or intensification of medical therapy based on routine ACAD screening improve outcomes of asymptomatic diabetic patients?

The effectiveness of pre-emptive coronary revascularization and intensified medical therapy, guided by routine ACAD screening, in improving outcomes for asymptomatic diabetic patients remains a topic of debate [129].

Furthermore, initial exploratory randomized controlled trials (RCTs) did not demonstrate a prognostic benefit from ACAD screening. However, these trials may have been limited by under-sampling and potential Type II errors, as the event rates were lower than anticipated [130, 131].

The prospective multicenter BARDOT trial [132] that evaluated prevalence, progression, treatment and outcome of silent ACAD in asymptomatic patients with DM at high coronary risk, showed that patients with normal myocardial perfusion single-photon emission computed tomography (MPS) (78%) had a low incidence of first manifestation of CAD, and patients with abnormal MPS at baseline (22%) experienced a seven-fold higher rate of progression to either “overt or silent ACAD”, even with treatment [132]. Outcomes from randomized trials indicate that a combined invasive and medical approach for silent ACAD may reduce scintigraphic progression, but does not impact the progression of symptomatic ACAD when compared to medical therapy alone [132].

A meta-analysis by Clerc et al [133], which included five RCTs and 3,299 asymptomatic diabetic patients, compared the outcomes of routine ACAD screening versus standard care in relation to cardiac events over a weighted average follow-up period of 4.1 years.

The analysis demonstrated a significant 27% reduction in the primary composite endpoint of any cardiac event. This reduction was primarily attributed to lower rates of non-fatal MI (− 35%) and HF hospitalizations (− 39%), although neither of these outcomes reached statistical significance when assessed individually. To prevent a single cardiac event over a 4-year follow-up, 56 asymptomatic Type 2 diabetic patients would need to undergo CAD screening [133]. The reduction in the primary outcome was fairly consistent across the five trials included in the meta-analysis, but it did not achieve statistical significance when analyzed separately in each individual trial.

In the DIAD and FACTOR-64 trials, the two largest RCTs included in this meta-analysis, the average annual major cardiac event rate (cardiac death or MI) was 0.6% and 0.8%, respectively—3 to 4 times lower than initially anticipated. Unlike the other trials, which had higher event rates, DIAD [130] and FACTOR-64 [131] did not require additional risk factors for inclusion. Consequently, both trials were likely underpowered to detect small risk differences with adequate statistical power (80–90%). Several studies also faced recruitment challenges: DYNAMIT [134] was prematurely stopped due to low patient enrollment, while FACTOR-64 and DADDY-D [135] extended follow-up periods to compensate for poor recruitment. The FACTOR-64 follow-up study showed that among asymptomatic patients with type 1 or T2DM followed for over twelve years, the use of CCTA to screen for ACAD did not affect the rates of all-cause mortality or nonfatal MI [136]. All patients enrolled in this trial [136] had much better medical management of blood sugar, cholesterol levels and blood pressure.

Moreover, this meta-analysis included a range of screening modalities: FACTOR-64 [131] used CACS and CTCA, DIAD [130] employed myocardial perfusion imaging (MPI), DYNAMIT [134] and DADDY-D [135] relied on exercise electrocardiogram Test (EET), while Faglia et al. [137] combined EET and stress echocardiography (SE). Heterogeneity among trials also existed as to the treatment strategy chosen by the practicing physician, and in the number of patients with abnormal screening undergoing coronary angiography [138]. Again, the FACTOR-64 study, which assigned asymptomatic diabetic patients to CAD screening by CCTA or standard diabetes care, did not show a significant decrease in adverse events in the CCTA group [131]. The patient cohort in this study was not considered very high risk (only 40.7% had a CAC score >100), and severe proximal stenosis was detected in only 6% of patients in the CCTA group. Furthermore, the effect of revascularization—which was performed in only a small proportion of patients (6% in the CCTA group, 1.8% in the non-CCTA group)—did not substantially impact overall mortality.

These methodological heterogeneities represent a limitation of the meta-analysis. However, the meta-analysis addresses some of the statistical limitations of the individual trials and confirms a modest but significant trend toward better outcomes with ACAD screening.

In a population-based, parallel-group, randomized controlled trial involving men aged 65 to 74 years from 15 Danish municipalities [139], participants were randomly assigned in a 1:2 ratio to either undergo screening (the invited group) or not undergo screening (the control group) for subclinical CV disease. The screening involved non-contrast electrocardiography-gated computed tomography to determine the CAC score. The primary outcome was death from any cause. In intention-to-treat analyses, after a median follow-up of 5.6 years, 2,106 men (12.6%) in the invited group and 3,915 men (13.1%) in the control group had died (hazard ratio, 0.95; 95% confidence interval [CI], 0.90 to 1.00; P = 0.06). The hazard ratio for stroke in the invited group compared to the control group was 0.93 (95% CI, 0.86 to 0.99); for myocardial infarction, 0.91 (95% CI, 0.81 to 1.03). No significant differences were found between groups for safety outcomes. The results of this trial [139] showed that after more than 5 years, offering comprehensive CV screening did not lead to a significant reduction in the incidence of death from any cause, although the CI and p-value show a trend for a benefit.

A limitation of the trial is that it included only men; in addition, 1.7% of participants were found to have previously undiagnosed diabetes at baseline, which may limit the generalizability of the findings to population with a higher prevalence of diabetes. Furthermore, the current follow-up had a median of 5.6 years, while the trial was initially powered for a planned 10-year follow-up. Longer-term follow-up is still in progress.

Guidelines Recommendations for screening CAD in asymptomatic patients

The 2024 American Diabetes Association Guidelines section of Cardiovascular disease and risk management suggest that routine screening for ACAD is not recommended in asymptomatic individuals, as it does not improve outcomes when ACAD risk factors are properly managed, and consider investigating ACAD if any of the following are present: atypical cardiac symptoms; signs or symptoms of associated vascular disease, including carotid bruits, transient ischemic attack, stroke, claudication, PAD, or electrocardiogram abnormalities (e.g., Q waves) [140] Table 2.

According to these guidelines, screening for ACAD in asymptomatic individuals at high risk is not recommended [3], partly because these individuals should already be receiving intensive medical therapy, which provides benefits similar to those of invasive revascularization [141, 142]. There is also evidence suggesting that silent ischemia may resolve over time, which adds to the controversy surrounding aggressive screening strategies [87]. In prospective studies, coronary artery calcium has been identified as an independent predictor of future ACAD events in individuals with diabetes, and it consistently outperforms both the UK Prospective Diabetes Study (UKPDS) risk engine and the Framingham Risk Score in predicting risk in this population [143].

The 2023 European Society of Cardiology Guidelines for the management of CV disease in patients with diabetes [5] suggest that in diabetic patients with asymptomatic ASCVD (including documented ACAD confirmed by imaging) and elevated CV risk, the potential benefit of platelet inhibition with aspirin (ASA) may be greater, making it important to tailor therapy to the individual.

The 2024 European Society of Cardiology Guidelines for Chronic Coronary Syndromes [121] suggest about screening for ACAD in asymptomatic individuals when coronary artery calcification findings from prior chest CT scans are available. It may be beneficial to use these findings to improve risk stratification and guide the management of modifiable risk factors, (level of evidence IIa class C); CACS could be considered to enhance risk classification at treatment decision thresholds; Total risk estimation using a risk-estimation system such as SCORE is recommended for asymptomatic adults >40 years of age without evidence of CVD (level of evidence I class C). CACS may be considered to improve risk classification around treatment decision thresholds (level of evidence IIb, class C) [121].

A retrospective study was performed in 385 asymptomatic patients with diabetes and no history of ACAD, but with target organ damage or at least three additional risk factors besides diabetes, to test the performance of the ESC-EASD proposal for screening for silent myocardial ischemia asymptomatic patients [144]. CAC score was assessed using CT and stress myocardial scintigraphy was performed to identify silent MI, with coronary angiography conducted for those with SMI. They showed that the most effective strategy for selecting patients for SMI screening consisted in performing myocardial scintigraphy in the 146 patients with severe TOD and, in the 239 remaining patients without severe TOD, in those with a CAC score of ≥ 100 AU. This approach achieved 82% sensitivity for SMI detection and successfully identified all patients with coronary stenoses; these results emphasized also the role of CAC as a useful marker of silent MI in patient with diabetes but no evidence of TOD.

The French Society of Cardiology in collaboration with the French-speaking Society of Diabetology [145] proposed an algorithm for stratifying and managing the risk of ACAD in Asymptomatic Diabetic Patients, involving CAC in the decision making:

Very High Risk Patients includes:

Patients with target organ damage are classified as very high risk. If any factor (previous CV disease including atrial fibrillation/heart failure; LDL cholesterol >190 mg/dL despite treatment, albuminuria >300 mg/24 hours or 200 mg/L or equivalent; eGFR < 30 mL/min/1.73 m², ECG with abnormal Q waves, Echocardiography with abnormal LV function/hypertrophy, peripheral atheromatous stenosis >50%) is present, the patient is categorized as very high risk.

High risk include patients with T2DM lasting 10 years or more, who represent the majority of the diabetic population, premature ACAD in a first-degree relative, persistently uncontrolled risk factors, confirmed albuminuria, severe retinopathy or autonomic neuropathy or erectile dysfunction, low physical activity (cannot climb more than 2 stairs).

Among the above-mentioned items, if no or only one item is ticked, the patient is considered at moderate risk.

If two or more items are ticked, a CAC score may be recommended to further assess the risk.

CAC 0–10 AU: Moderate risk; 11–100 AU: Moderate risk if age is ≥50 years; high risk if age is <50 years; 101–400 AU: high risk if age is ≥60 years; very high risk if age is <60 years

>400 AU: Very high risk

For patients identified as very high risk, screening for CAD should be carried out according to local protocols, such as ESC Guidelines [5]. If the patients are at very high risk with ischemia or suspected significant CAD they may have a cardiological advice or coronary angiography. In particular:

Patients at very high risk

• Regular screening: It is recommended to regularly perform diagnostic tests for coronary artery disease, such as coronary angiography, to identify potential issues early.

• Intensive therapeutic interventions: These include lifestyle changes, strict control of blood glucose levels, blood pressure, and lipids, along with the use of antiplatelet therapies and, if indicated, revascularization interventions.

Patients at moderate risk

• Individual assessment: It is important to carefully assess the individual risk, considering factors such as age, duration of diabetes, the presence of microangiopathies, and other cardiovascular risk factors.

• Personalized screening: Screening should be conducted based on individual risk assessment, adopting targeted preventive strategies that include lifestyle changes and optimal control of risk factors.

• Cosson et al. [146] investigated the extent and direction in which CAC score may reclassify coronary risk in asymptomatic diabetic patients with high a priori coronary risk, and whether screening for asymptomatic MI or coronary stenosis in patients at very high coronary risk—either a priori or based on reclassification with CAC score—offers good sensitivity for detecting these conditions,. They retrospectively selected 377 asymptomatic diabetic patients at high or very high a priori coronary risk. All patients had their CAC score measured and underwent stress myocardial scintigraphy to detect MI. Those found to have ischemia then underwent coronary angiography to identify coronary stenoses. Based on French guidelines, 66% of asymptomatic diabetic persons at high a priori coronary risk were reclassified into the moderate risk category, with consequent less tight prevention goals applied. At variance, 18% percent were reclassified into the very high-risk category.