Carotid Ultrasonography in the Assessment of Cardiovascular Risk

Summary

Carotid ultrasound is one of the most accessible examinations in daily clinical practice for the evaluation of the arterial status. However, the clinical implications of the presence, the extension and the morphology of carotid damage are not entirely clear. Aim of this narrative review is to discuss the role of carotid ultrasound in the assessment of cardiovascular risk through the examination of the updated evidence in the literature. We describe the technical aspects of the procedure and the possible correlations between the imaging results and the assessment of the cardiovascular risk. Some insights about new, more sophisticated techniques for carotid evaluation, such as carotid three-dimensional and contrast-enhanced ultrasound, are also presented.

Carotid ultrasound evaluation

B-mode ultrasound carotid evaluation is a high-resolution, noninvasive and widely available technique for the direct visualisation of the arterial anatomy (Figure 1) [1]. The examination determines the carotid intima-media thickness (CIMT), defined as the thickness, measured in the far wall of the common carotid artery (CCA), from the media-adventitia interface to the intima-lumen interface (with a threshold value for subclinical vascular damage > 0.9 mm) (Figure 2), and the possible presence of atherosclerotic plaques, defined as CIMT > 1.5 mm or as focal lesions encroaching into the vascular lumen by at least 0.5 mm or 50% of the surrounding IMT value [2]. From a pathophysiological point of view it is important to stress that an increase in CIMT and a carotid plaque are different phenomena. The main predictors of increased CIMT, which is specifically an expression of hypertrophy of the media and adequately evaluated at the level of the CCA, are age and hypertension [3]. In contrast, plaque, which is an expression of endothelial dysfunction, recruitment of inflammatory cells, accumulation of cholesterol and cellular debris (i.e., of the atherosclerotic process), is located in the intima and is more strongly predicted by the traditional athero-sclerotic risk factors [4,5,6,7]. It occurs in sites with turbulent flow, such as the carotid bulb and the proximal internal carotid artery, with the CCA involved in very advanced damage only [8]. The carotid plaque can be characterised in terms of various features: degree of stenosis (mild < 50%, moderate 50–69%, severe 70–99%), surface regularity, and echogenicity. For the degree of stenosis, Schulte-Altedorneburg et al. showed a good correlation between in vivo ultrasound findings and evaluation of postmortem carotid specimens in the same critical patients [9]. The degree of carotid stenosis is weakly related to the traditional atherosclerotic risk factors [10]. In addition, irregularities of the carotid plaque surface may represent a good marker of its vulnerability, which translates into increased risk of thromboembolic cerebral events [11,12,13]. With regards to plaque echogenicity, several studies clearly showed that echolucent carotid plaques were more related to cardiovascular events (CVEs) than were echogenic ones [14,15,16]: in fact, echolucent plaques histologically demonstrate a higher lipid content and a reduced fibrous and calcium content, which represent important features of advanced and/or unstable atherosclerotic disease [17]. Different methods exist for the quantitative assessment of the ultrasonographic properties of the carotid plaques: the Gray Scale Median score, a measure of plaque brightness, and the integrated backscatter, which is able to differentiate lipids from other components [18].

Two-dimensional carotid ultrasound for the prediction of cardiovascular events

• Assessment of cardiovascular risk

Several scoring systems are used for cardiovascular risk (CVR) prediction in clinical practice: the American Framingham Risk Score (FRS) and the European Systematic Coronary Risk Evaluation (SCORE) charts have been proposed and validated [19,20], and recently the American College of Cardiology and the American Heart Association released a new 10-year cardiovascular disease (CVD) calculator [21]. The Reynolds Risk Score uses, in addition to traditional risk factors, information from high sensitivity C-reactive protein (hs-CRP) and parental history of premature CVD [22,23], which stresses the role of subclinical inflammation and of the genetic background in the pathogenesis of atherosclerosis; this score reclassified 30% of women estimated to be in the intermediate-risk group with the traditional FRS into a higher or lower risk category with improved accuracy [22].

These scores are simple to use in evaluating CVR; however, in some cases the estimate is suboptimal, particularly in subgroups of the general population (for example, those at low or intermediate risk), since they vary widely in terms of the background risk in study populations, predictors and outcomes considered [24,25,26,27,28,29]. Other limitations of these scores include a substantial underestimation of lifetime risk, with a potential misclassification of high-risk subjects at low or intermediate CVR [30]. Among healthy adults aged 20 to 79 years, risk scores classify 75% of the subjects as low or intermediate risk, but in these categories a substantial number of CVEs (about 60%) occurs [31]. The main goal for primary prevention seems to be the identification of low-risk or intermediate-risk subjects who have subclinical atherosclerosis, and therefore could be reclassified to a higher risk level [32]. To improve individual risk assessment, several studies investigated the role of CIMT and carotid plaque evaluation with a noninvasive and safe technique such as carotid ultrasound.

• Role of the carotid intima-media thickness

Some years ago, Lorenz et al. published a systematic review and meta-analysis involving 37,197 subjects from the general population and showed that for an absolute CIMT difference of 0.1 mm the future risk of CVEs increased by more than 10%: this relationship was stronger for stroke than for myocardial infarction and was independent of traditional risk factors [33]. However, the incremental value of CIMT measurement for risk stratification in addition to those risk factors was less clear: a subsequent meta-analysis by Den Ruijter et al. involving 45,828 subjects showed that the addition of CIMT to FRS determination modestly improved 10-year risk prediction (net reclassification improvement of 0.8%), although a more evident influence could be seen in subjects at intermediate risk, with a net reclassification improvement of 3.6% [34]. Very recently, Bots et al. confirmed these conclusions in 17,254 subjects with elevated blood pressure, demonstrating the limited but significant utility of CIMT measurement in CVR prediction for individuals at intermediate risk [35].

• Role of carotid intima-media thickness progression or regression

Several studies evaluated the progression of CIMT as a predictor of CVD: although the MESA trial showed a significant association between CIMT progression and CVEs [36], a subsequent large meta-analysis showed that the progression of CIMT (detected through the changes between two ultrasound examinations 2–7 years apart) did not improve the CVR prediction [37]. In addition, the evaluation of treatment-induced regression of CIMT failed to demonstrate a predictive value for CVEs [38]. For these reasons, in recent European Society of Hypertension/European Society of Cardiology guidelines for management of arterial hypertension, CIMT evaluation was considered of limited value for the follow-up of hypertensive patients [39].

• Recent statements of the guidelines

In 2013, a guideline from the American College of Cardiology and the American Heart Association reconsidered the role of CIMT, in conjunction with other biomarkers, in the assessment of CVR [40]: the conclusions of the panel’s experts were that, aher quantitative risk assessment with the usual tools, in the presence of uncertainties about treatment decisions family history of premature CVD, hs-CRP serum levels, coronary calcium score (CCS) or ankle-brachial index, but not CIMT, could be evaluated to reclassify the risk. On the other hand, a previous guideline from the American College of Cardiology Foundation and the American Heart Association for the assessment of the CVR in asymptomatic adults evaluated as “reasonable” the measurement of CIMT in subjects at intermediate risk (Class IIa of recommendation, level of evidence B) [41].

• Role of carotid plaque in cardiovascular risk prediction: comparison with CIMT

The definition of CIMT is straightforward and allows adequate comparisons between studies. On the contrary, criteria for defining and assessing carotid plaque and, more generally, the global carotid damage, defined as carotid burden, can vary [42]. Some authors initially tried to evaluate the cardiovascular predictive value through a carotid plaque score, calculated as the sum of the maximum thickness of the plaques in the various segments of both carotid arteries. Several studies showed that, compared with CIMT, plaque score was a stronger predictor of CVEs and correlated with the complexity of coronary lesions [43,44,45]. However, Spence et al. suggested that the best method for assessing the carotid burden was to calculate the carotid plaque area (CPA), defined as the sum of the areas of all the plaques in the carotid tree evaluated in a longitudinal view: subjects with high values of CPA at baseline (patients in the highest quartile for CPA) presented a significant increase in CVEs (Figure 3 and Figure 4) [46]. With consecutive measure-ments of CPA, the progression of the atherosclerotic damage was linked to increased CVR, something not demonstrated with CIMT (see above) [42]. During recent years it has emerged that the assessment of carotid plaque was a better predictor of cardiovascular risk, compared with CIMT. In fact, although the Rotterdam Study did not show differences between CIMT and a carotid plaque score as predictors of myocardial infarction [47], several subsequent studies stressed the importance of carotid plaque evaluation as compared with CIMT in predicting coronary artery disease. In the initial presentation of the results of the Tromsø Study, CPA, evaluated in more than 6,000 subjects of a general population, was a stronger predictor of myocardial infarction than was CIMT [48]; aher 10 years of follow-up CPA was highly predictive for stroke also, whereas CIMT was still not predictive [49]. Subsequently, the Strong Heart Study demonstrated that the presence of carotid plaque was an important predictor of coronary artery disease, whereas CIMT was not an independent risk factor for CVD [50]. More recently, Polak et al., in a cohort of the Framingham Offspring Study, and Plichart et al., in the French Three-City Study, demonstrated that carotid plaque, but not CIMT, improved cardiovascular risk prediction [51,52]. A large meta-analysis, which evaluated most of these studies, confirmed these conclusions and stressed again the differences between the measurements of the vascular wall in the CCA (deliberately not containing plaque) and in the carotid bulb (possibly comprising a plaque) [53]. Furthermore, in the large cohort of the REACH registry (more than 10,000 subjects), Sirimarco et al. showed that carotid atherosclerosis was an independent predictor of coronary events in the recruited subjects, with different types of symptomatic vascular disease and presence of risk factors at baseline [54]. In comparison with other markers of atherosclerotic damage, such as serum CRP and CCS, Brook et al. demonstrated that, although each test may improve CVR stratification, a negative CPA determination is superior to CIMT, CCS, and CRP in its ability to reduce the likelihood of concomitant significant coronary atherosclerosis [55]. The main results of the cited trials are presented in

Table 1. Main results of the cited clinical trials evaluating the predictive role of the carotid ultrasound features for cardiovascular risk.

.

Three-dimensional ultrasound and contrast enhanced ultrasound

• An improvement in carotid assessment and cardiovascular risk prediction?

Only a few studies analysed the role of carotid three-dimensional ultrasound (3dUS) in the assessment of CVR, in which the plaque is evaluated in both a longitudinal and a perpendicular cross-sectional axis [56,57]. This technique seemed to improve the carotid burden evaluation through the estimation of the total plaque volume (TPV). A recent trial showed that progression of TPV better predicted CVEs compared with total plaque area and CIMT [58]. Subsequently, Kuk et al. showed that also the volume of carotid ulceration, evaluated with the same technique, significantly correlated with CVEs [59]. The strong correlation between 3dUS and coronary artery disease was confirmed by Johri et al. in subjects undergoing carotid ultrasound evaluation and concomitant coronary angiography [60].

Another important emerging test for the early detection of plaque instability and prediction of CVEs is contrast-enhanced ultrasound (CEUS) of the carotid artery. This technique is able to identify intraplaque neovascularisation and adventitia vasa vasorum neoproliferation through the injection of microspheres via a peripheral vein [61]. Staub et al., evaluating 147 subjects retrospectively, demonstrated the association of these features with CVD and previous CVEs [62]. For further confirmation, Deyama et al. showed that, in subjects with known carotid plaque undergoing coronary angiography, CEUS evaluation of the carotid plaque neovascularisation is strongly correlated with the severity of coronary artery disease [63]. In addition, van den Oord et al. demonstrated that CEUS significantly improved the accuracy of plaque assessment by ultrasound [64]. See Table 1 for a summary of the cited articles.

Further studies are needed to confirm the role of 3dUS and CEUS as new tools for early identification of individuals at risk of new CVEs [65].

Conclusions

Carotid ultrasound is a useful, extremely safe technique for the evaluation of vascular anatomy and may allow a net reclassification improvement of the CVR. Several reports indicate that CIMT is clinically useful at least in subjects at intermediate risk [33,34]. The carotid plaque has a greater predictive value: the evaluation of the CPA seems to give the best information for the initial assessment of the atherosclerotic damage and the response to appropriate therapies. In recent years, new sophisticated ultrasound-based techniques have emerged, such as carotid 3dUS and carotid CEUS. Large ongoing observational studies already underway include 2d and 3d carotid ultrasound for the screening of subclinical atherosclerosis and the evaluation of its progression [66]. Another im-portant clinical effort, within the High-Risk Plaque Initiative, is the BioImage Study in which enrolled subjects were subdivided in a survey-only group, a group undergoing traditional risk factor scoring, and a third group with the risk factor scoring and the additional evaluation of subclinical atherosclerosis (in conjunction with the determination of various new biomarkers): follow-up will be terminated by the occurrence of an adequate number of atherothrombotic events [67].

Although it is without doubt that the development of carotid and coronary plaques is closely related [68], the transition from a population level to an individual patient management in terms of cardiovascular prevention remains not unanimously stated, since in the individual subject the presence of a carotid plaque does not automatically imply the presence ofan underlying coronary artery disease and hence the need to perform a coronary angiography [69].

In addition, some studies suggest that carotid evaluation does not seem to motivate patients to follow a healthy lifestyle or to improve treatment adherence [70,71]. Nevertheless CIMT and/or carotid plaque detection in subjects at low or intermediate risk can certainly represent a useful biomarker to start, or to change with a more aggressive strategy, drug treatment aimed to correct the traditional risk factors.