Imaging subclinical atherosclerosis in cardiovascular risk stratification

Atherothrombotic diseases represent the most common cause of death worldwide. The atherosclerotic process begins early in the life, has a prolonged subclinical phase and may remain unrecognized until it triggers clinical events. Indeed, more than a half of the men and women who die suddenly of atherothrombotic cardiovascular events have no previous symptoms. These data emphasize the need to identify ‘vulnerable patients’ for atherothrombotic events already at a silent subclinical stage in order to apply adequate lifestyle and therapeutic interventions aimed to attenuate progression or even to promote regression of existing lesions. Current guidelines on primary prevention of atherothrombotic cardiovascular disease1 recommend to estimate total cardiovascular risk by means of risk prediction algorithms and consider additional evaluation by imaging techniques in subjects with calculated risk near the decisional thresholds for preventive therapy or in those with another strong risk factor, such as family history of premature events, metabolic syndrome or chronic inflammation. Prediction algorithms are based on standard cardiovascular risk factors and do not reflect the influence of genetic, epigenetic and lifestyle variables. In addition, the two non-modifiable risk factors, age and sex, are the most heavily weighted variable in the majority of risk models and, accordingly, risk charts are poorly adapted in women and in younger age groups. The sensitivity of the Framingham risk chart and SCORE chart was tested in more than 25,000 participants of the FINRISK Study.2 While the sensitivity of the two charts for cardiovascular disease mortality was 93% and 64% in men, it was only 72% and 24% in women. Furthermore, the sensitivity of SCORE charts for cardiovascular mortality was much higher in men aged 55 to 64 years than in men below 55 years of age, being 91% and 33%, respectively. Imaging techniques can personalize the risk assessment by integrating the cumulative effect of genetic and epigenetic risk determinants and of lifetime exposure to established and emerging risk factors on the arterial tree in each individual patient. The two vascular beds that are the basis for most of the cardiovascular events are coronaries and carotid arteries and, consequently, coronary artery calcium (CAC) scoring and extracranial carotid plaque detection are recommended for improving risk stratification (class IIb-IIa of recommendation). The greatest contribution of imaging biomarkers is believed to be their negative predictive value, that is, the capability to identify the subjects free of subclinical atherosclerosis who will not benefit from primary prevention interventions that are not free of side effects and cost. In the Multi-Ethnic Study of Atherosclerosis (MESA), 31% of subjects in whom lipid-lowering therapy was recommended upfront, according to the 2016 European Society of Cardiology recommendations, had CAC score = 0.3 However, given the high sensitivity of atherosclerotic imaging techniques, they also identify the presence of atherosclerotic lesions in a certain portion of subjects considered to be at low risk. In the Progression Early Subclinical Atherosclerosis study,4 58% of the middle-aged asymptomatic participants at low cardiovascular risk as estimated by SCORE chart had subclinical atherosclerosis in at least one vascular district and 9% had generalized atherosclerosis (in at least four different districts). In the Coronary Artery Risk Development in Young Adults study,5 44% of asymptomatic early middle-aged subjects with Framingham risk score <10% had CAC score >0, and 17.2% had CAC score ≥100. Similar results were obtained in women of the MESA Study,6 where 90% were classified by Framingham risk score as being at low risk, and within them 32% had CAC score >0 and 4% had CAC score ≥300. The important topic regarding the presence of atherosclerotic burden in the general population according to risk groups is also addressed in the study by Östgren et al., published in this issue of the European Journal of Preventive Cardiology.7 Authors have investigated the presence of CAC and carotid artery plaques in 13,114 participants of Swedish Cardiopulmonary BioImage Study, aged 50–64 years and free of prevalent cardiovascular disease or diabetes mellitus. In such a cohort, the prevalence of subclinical coronary and carotid atherosclerosis was 39.5% and 56.0%, respectively, and the presence and extent of atherosclerosis was associated with SCORE estimated risk. This observation is not surprising as both SCORE algorithm and atherosclerotic process are age-driven. The results also confirmed that the greater usefulness of vascular biomarkers is their negative predictive value. In fact, a SCORE < 1% indicated a low risk of significant coronary (CAC ≥100) or carotid artery (more than one plaque on both sides) disease, with a negative predictive value of 92% and 98%, whereas a SCORE ≥5 indicated a high risk, with a positive predictive value of 32% and 45%. However, looking at the results in detail, we can see that within subjects at low risk, which represented 25% of population, 17% and 8% have already some degree of coronary and carotid atherosclerosis, respectively. These data once again raise the questions of if, when and how to screen for subclinical atherosclerosis in asymptomatic subjects. Current guidelines indicate the use of imaging techniques in intermediate-risk patients or in those with another strong indication. Yet, as demonstrated by the above quoted studies,4,5,6–7 there is some heterogeneity between risk estimated by prediction algorithms and actual atherosclerotic burden; a certain percentage of low-risk asymptomatic subjects has already atherosclerotic changes in carotid and/or coronary arteries. The importance of subclinical atherosclerosis detection in asymptomatic subjects has been highlighted in the BioImage study,8 where both carotid plaque burden and CAC significantly improved risk predictions for major adverse cardiac events as compared with conventional risk factors, with comparable results for either arterial district. A beneficial effect of lifestyle interventions and statin therapy on atherosclerotic burden has been also demonstrated. In a small study,9 patients with moderate to severe coronary heart disease at coronary angiography were randomized to an intensive lifestyle-change group or to usual-care control group. Coronary angiography repeated after five years showed 7.5% reduction in coronary stenosis in the lifestyle-change group and 27.7% worsening in coronary stenosis in the control group. The incidence of cardiac events per patient was more than double in the control group. Statins seem to influence not only the extent of coronary atherosclerotic lesions but also their composition; in a large multinational study on patients without history of coronary artery disease undergoing coronary computed tomography angiography at an interscan interval of at least two years,10 statins slowed the rate of overall atheroma volume progression, increased plaque calcification and reduced high-risk plaque features. Considering that the events rate is predominantly driven by burden of atherosclerosis and that preventive interventions may reduce atherosclerotic burden, low-risk asymptomatic patients with atherosclerosis might benefit from early prevention. There is an ongoing discussion whether or not to use statins in primary prevention in subjects at lower cardiovascular risk, balancing the benefits and harms. It could be assumed that the detection of subclinical atherosclerosis can identify the subjects in whom the benefit overwhelms the harm, even though the coronary events may occur in the absence of significant stenosis.11 The next question regards the selection of optimal imaging method. Different studies comparing the effectiveness of CAC and carotid plaques in risk prediction obtained conflicting results, probably due to the differences in study populations and in methods used for quantification of carotid atherosclerosis. Recent data from the MESA study12 indicate that CAC score is a stronger predictor of cardiovascular disease and coronary events (hazard ratio 1.73 and 2.09) than carotid plaque score (hazard ratio 1.27 and 1.35), while both scores have similar mild incremental predictive value for stroke/transient ischaemic attack (hazard ratio 1.22 for CAC and 1.15 for carotid plaques). In fact, the presence of CAC is a surrogate for the extent of coronary atherosclerotic burden; however, concerns for cost and radiation exposure may limit the use of CAC as a population-wide screening tool. On the other side, the potential of carotid ultrasound for risk prediction is probably not fully exploited. Carotid plaque volume quantification by three-dimensional approach and identification of vulnerable plaque by tissue characterization could provide additional information, both for risk stratification and assessment of preventive interventions. Finally, recent clinical evidence suggests that cardiac calcifications, as detected by echocardiography at mitral apparatus, aortic valve or ascending aorta, may be also predictive for future cardiovascular events.13 Therefore, existing data sustain the use of imaging biomarkers, which are proxies of atherosclerotic disease, for cardiovascular risk prediction even in asymptomatic low-risk subjects. The awareness of having atherosclerosis may motivate the patient to adhere to primary prevention recommendations, as is also underlined in the paper of Östgren et al. Yet, the optimal method(s) is/are still waiting for validation and cost-effective analysis. The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The author(s) received no financial support for the research, authorship, and/or publication of this article. The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.