INTRODUCTION — Significant progress has been made to develop noninvasive atherosclerosis imaging techniques that can serve as validated biomarkers (surrogate end points) for cardiovascular health status, aid in the identification of disease risk factors, and identify and evaluate measures to prevent disease. One of the rationales for atherosclerosis imaging originates from autopsy studies in adolescents and young adults, which revealed that atherosclerosis is present early in life and precedes clinically manifested cardiovascular events. Therefore, non-invasive arterial imaging has the potential to provide information on arterial wall structure in persons of all age groups as a continuous variable to describe all stages of atherosclerosis progression and regression [1,2]. Consequently, using imaging to detect early atherosclerosis has the potential to identify the status and predict the risk of future cardiovascular events in populations, as well as evaluate existing and novel drug efficacy to guide cardiovascular therapies prior to the occurrence of atherosclerotic cardiovascular disease.
One of the most widely used and best validated atherosclerosis imaging techniques is the ultrasound carotid intima-media thickness (CIMT) measurement. The techniques for imaging and measuring CIMT along with the potential clinical applications of CIMT will be discussed here. Carotid artery imaging for the detection of significant obstruction carotid disease is discussed separately. (See "Evaluation of carotid artery stenosis".)
CIMT MEASUREMENT BY ULTRASOUND — CIMT can be measured using either high resolution ultrasound or magnetic resonance imaging (MRI). At present, carotid ultrasound has provided the majority of imaging data; however, with its increasing availability, MRI has emerged as the technology with novel applications.
Image acquisition — The large superficial arteries, in particular the carotid arteries (figure 1), can be visualized at high resolution with B-mode (or "brightness" mode) ultrasound using linear array transducers for superficial and vascular structures (typically broadband transducers with frequency ranges 5 to 15 MHz and a high signal to noise ratio are used). The spatial resolution obtained with these transducers is on the order of 0.05 mm axially and 0.2 mm laterally. In most scan protocols, carotid ultrasound images are obtained from the near and far walls of the right and left distal common carotid arteries, the carotid bifurcation (image 1), and the proximal internal carotid arterial segments. CIMT in an individual patient is therefore often a composite of intima-media thickness measurements of various arterial segments, sometimes under multiple angles of insonation to decrease measurement variability in follow-up scans.
Measurement technique — For CIMT measurements, longitudinal images of the carotid arteries are obtained in which the leading edges of the lumen-intima and media-adventitia interfaces (the "double-line pattern") of the arterial wall represent intima-media complex (image 1) [1,2]. Typically, normal common carotid CIMT at age 10 is approximately 0.4 to 0.5 mm, while from the fifth decade of life onward this progresses to 0.7 to 0.8 mm or more [3].
Since the arterial walls are submillimeter structures, performing CIMT measurements requires experience and craftsmanship, and standardized image acquisition is paramount. Sonographers should be well trained, and standardized ultrasound equipment and protocols should be used. Furthermore, image analysis should preferably be done offline, with dedicated software using standardized protocols and performed by expert technicians. Moreover, quality control results of scans and feedback to sonographers must have a rapid turnaround. Under these conditions, ultrasound CIMT measurement has excellent interscan as well as inter- and intraobserver reproducibility [4].
Use of ultrasound measurements in single- and multicenter studies — In trial design and logistics, bringing study subjects (recruitment), research personnel (training, certification) and equipment (availability, standardization) together, are major challenges. Ultrasound imaging is non-invasive, subject-friendly, and due to standardization of instrument settings and scan protocols and relatively low equipment and infrastructure costs, study size is easily scalable between single- and multicenter settings. This flexibility of application of ultrasound technologies is one of great logistic benefit in atherosclerosis studies [1,4].
Applications for CIMT measurements — The suggested indications for CIMT measurement have included:
●Screening for atherosclerosis
●Risk stratification for future cardiovascular disease-related events
●Assessment of drug efficacy
The corresponding data for each of these indications along with recommendations regarding the use of CIMT measurement for each indication are discussed below.
CIMT for cardiovascular risk stratification — Cardiovascular risk prediction in asymptomatic individuals is based on risk scores. Although guidelines have based risk assessment on the use of multivariate risk models, their ability to predict future cardiovascular events has limitations [5-7]. In fact, a high percentage of cardiovascular events occur in persons at low and intermediate risk [8]. CIMT has been proposed as a potential tool to aid cardiovascular risk stratification as it comprises a direct measure of atherosclerosis, is associated with future cardiovascular events, and is a safe, inexpensive, and widely available technique [2,9,10]. (See "Atherosclerotic cardiovascular disease risk assessment for primary prevention in adults: Our approach".)
CIMT and future cardiovascular events — In patients without known cardiovascular disease (CVD), composite measures of CIMT of various segments, with or without the presence of atherosclerotic plaque, result in modest but statistically significant increases in net reclassification improvement when added to risk estimates based on traditional risk factors or risk models. Whether the improvement in risk stratification due to CIMT and plaque information translates into measurable health benefits for individuals or populations, or reductions in the cost of care, has not been established. Until additional data are available that directly correlate CIMT measurement with improvements in hard outcomes (ie, death, myocardial infarction, stroke), we do not recommend routine CIMT measurement.
Numerous prospective epidemiological studies have shown that CIMT is a predictor of future cardiovascular events, independent of traditional risk factors. In addition, three systematic reviews and meta-analyses have reviewed the one-time and serial use of CIMT for CVD risk prediction [11-13].
●One-time CIMT measurement – In a systematic review and meta-analysis on the individual data of 14 studies of 45,828 asymptomatic individuals who underwent one-time CIMT measurement at baseline (median follow-up 10.8 years), the hazard ratio (per 0.10 mm adjusted mean common CIMT difference) for first myocardial infarction and stroke was 1.09 (95% CI, 1.07-1.12), for first myocardial infarction 1.08 (95% CI, 1.05-1.10), and for first stroke 1.12 (95% CI, 1.10-1.15) [11]. CIMT measures of other carotid artery segments and the presence of plaque showed similar relationships with future cardiovascular events [10,12,14-16].
While a one-time CIMT measurement can provide some predictive value for CVD risk assessment, we do not routinely perform CIMT measurement in most patients, preferring to use a standard cardiovascular risk assessment model based on traditional cardiac risk factors.
●Serial CIMT measurements – In a systematic review and meta-analysis on the individual data of 16 studies of 36,984 patients without known CVD who underwent serial CIMT measurement (mean follow-up seven years), in which the yearly progression rates were calculated for various CIMT measures (mean and maximum CIMT values of the common, bifurcation, and internal carotid arteries), there was no association between progression of CIMT and future events [12]. These findings are corroborated by a more recent meta-analysis including 31 studies of 89,070 patients showing a consistent association between CIMT and the combined endpoint of myocardial infarction, stroke, and vascular death, while there was no association between CIMT change and vascular event risk [13]. While there are various potential methodological as well as biological explanations, there is no proof of a relation between CIMT progression and future cardiovascular events.
Because of this, we do not recommend the use of serial CIMT measurements for the purpose of CVD risk assessment.
Although CIMT independently predicts future cardiovascular events, some issues need to be addressed before CIMT can be considered for cardiovascular risk stratification. First, the addition of CIMT needs to show improvement in cardiovascular risk stratification when added to traditional risk scores, as expressed by improvement in the net reclassification index [10,17]. Second, an improvement by CIMT in cardiovascular risk stratification should translate into a measurable health benefit and should be cost effective [18].
Regarding the added value of mean common CIMT to the Framingham 10-year risk prediction of myocardial infarction and stroke, the previously mentioned meta-analysis showed a net reclassification improvement in all subjects of 0.8 percent, and in subjects at intermediate risk of 3.6 percent [11]. This minimal increase in net reclassification following the addition of mean common CIMT to the Framingham risk predictor was not felt to be useful for cardiovascular risk stratification.
Net reclassification improvement has also been investigated for CIMT measurement of other segments of the carotid arteries with mixed results.
●In the ARIC study of 13,145 participants without known CVD, mean CIMT of all segments did show an improvement in reclassification of 7.1 percent in all subjects and 16.7 percent in subjects at intermediate risk [14]. Similar results were found in the IMPROVE study of 3703 asymptomatic subjects with at least three CVD risk factors, in which the net reclassification improvement in all subjects was 11.3 percent for mean CIMT and 10.5 percent for mean maximum CIMT of all segments [16].
●By contrast, in the CAPS study in 4904 participants without known CVD, CIMT of each of the different segments (common, bifurcation, and internal carotid) did not result in an increase in the net reclassification improvement when added to traditional risk factors [19].
The effect of the presence of plaque (defined as maximum CIMT >1.5 mm) on risk classification has also been investigated. In the ARIC study mentioned above, the addition of presence of plaque to traditional risk factors increased the net reclassification improvement by 7.7 percent in all subjects and 17.7 percent in subjects at intermediate risk [14]. Similar results have been reported from the Framingham Offspring Study cohort [15].
Recommendations of others — The 2013 American College of Cardiology Foundation/American Heart Association guideline for the assessment of CVD risk does not recommend the routine use of CIMT in clinical practice for CVD risk assessment [20].
CIMT and CVD risk in specific populations — An appealing aspect of CIMT measurement is that it assesses the atherosclerotic disease process itself, which reflects the net effect of known and unknown hereditary and environmental factors. CIMT can therefore be used to assess cardiovascular risk in populations where risk estimation with traditional risk factors doesn't apply, such as populations with specific genetic mutations in lipid metabolism. Ideally, prospective observational studies are needed to determine CVD risk in such populations, and randomized controlled trials should be performed to establish if this risk can be ameliorated. However, considering the rareness of such mutations, these types of studies are unlikely to become available.
CIMT can aid in assessing whether such patients have accelerated atherosclerosis development. An example is provided by studies in asymptomatic patients with genetic mutations in HDL metabolism, such as patients with LCAT and ABCA1 deficiency [21,22]. CIMT studies have indicated accelerated atherosclerosis development in these populations, which can be informative and of important guidance for clinicians treating these patients. Children with familial hypercholesterolemia (FH) represent another population in which CIMT has proven to be a valuable tool. Guidelines advise starting treatment with cholesterol lowering medications in children with FH from age 10 years. These recommendations are for an important part based on data from CIMT studies [23,24]. These examples illustrate that CIMT can be an informative and clinically relevant tool to estimate CVD risk in specific populations.
CIMT for the assessment of drug efficacy — CIMT has frequently been utilized in clinical trials to test the efficacy of cardiovascular drugs and results are directly associated with other vascular beds, like the coronary arteries [25]. However, while CIMT is an efficient tool to monitor the effects of therapy in clinical trials, modification of therapy based on CIMT results has not been shown to alter hard end points (eg, death, myocardial infarction, stroke). As such, CIMT is not routinely used in clinical practice to monitor the effects of medical therapy and is not used as an instrument to assess therapeutic efficacy in the individual patient.
Lipid-lowering therapy — With regard to lipid-lowering medications, various placebo-controlled CIMT studies in the 1990s showed that atherosclerosis progression was retarded by statin use in patients with cardiovascular disease [25-28]. Subsequently, aggressive lipid lowering with statins also showed markedly reduced CIMT progression when compared with moderate lipid lowering with statins, which was consistent with subsequent observations in clinical end point studies [29,30]. Furthermore, several CIMT trials revealed a beneficial effect of statins on atherosclerosis progression in asymptomatic subjects [31-34] and proved safe and sensitive enough to pick up efficacious benefits of statin treatment, even at a very young age [24,35,36]. In addition to statins, the efficacy of numerous non-statin lipid-modifying medications on atherosclerosis progression, as measured by CIMT, has been evaluated (eg, niacin, pactimibe, torcetrapib, ezetimibe) [4,37-43].
Anti-hypertensive therapy — The effects of various antihypertensive drugs on atherosclerosis, as measured by CIMT, have also been reported in numerous trials. As examples:
●Metoprolol decreased CIMT progression compared with placebo in subjects without cardiovascular disease (CVD) [44,45].
●Amlodipine decreased CIMT progression compared with placebo in patients with coronary heart disease [46].
●Conversely, ramipril did not show an effect on CIMT progression compared with placebo in patients with CVD or at high risk for CVD [47,48].
Other studies have compared the effect of different classes of blood pressure lowering drugs on CIMT progression [49-55].
Other medications — In addition to lipid-lowering medications and anti-hypertensive medications, antioxidants (eg, vitamin E, folic acid, vitamins B12 and B6), hormone replacement therapies (eg, estradiol, tibolone, raloxifene), glucose-lowering therapies (eg, rosiglitazone, pioglitazone, glimepiride), and anti-obesity medication (rimonabant) have also been investigated in CIMT studies [56-68].
CIMT as a surrogate marker to establish efficacy prior to large hard end point trials — CIMT measurement may be a useful surrogate marker of efficacy that can be used in small trials of new therapies prior to the initiation of a large-scale trial. In one study that investigated the agreement between the outcome of CIMT drug efficacy trials and morbidity and mortality studies investigating the same drug, the positive and negative predictive values of a CIMT trial to predict the outcome of a morbidity and mortality trial were 96 and 83 percent, respectively [69]. A meta-analysis of 119 randomized clinical trials involving 100.667 patients with an average follow-up of 3.7 years showed that interventions reducing CIMT progression were also likely to reduce CVD event rates. Each 10 micrometers per year reduction of CIMT progression, induced either by lipid-lowering, antidiabetic, antihypertensive, dietary, or other interventions, resulted in a relative risk for CVD of 0.91 (95% CI 0.87-0.94) [10].
This shows that in drug development, a CIMT trial can be used as a validated benchmark tool to guide the decision of whether or not to embark on a large-scale morbidity and mortality trial. The research, business, regulatory, and clinical impacts of these measurements are therefore of considerable importance in the decision making on acceptance and reimbursement of novel cardiovascular therapies.
CIMT MEASUREMENT BY MRI — CIMT can be measured using magnetic resonance imaging (MRI). However, given the additional complexity and cost of MRI when compared with ultrasound, CIMT measurement should not primarily be performed with MRI in most patients.
MRI does possess a number of important characteristics that make it well suited for the visualization of atherosclerosis. First, MRI is noninvasive, safe, and no ionizing radiation is involved. Second, MRI enables imaging of all stages of atherosclerosis (image 2), from healthy vessels to advanced disease in submillimeter resolution [70-73]. Furthermore, disparities in the spin-lattice relaxation time (T1), the spin-spin relaxation time (T2), and the proton density (PD) enable discrimination of plaque components such as intraplaque hemorrhage, lipid-rich necrotic core, and calcification (image 2) [70]. MRI of vessel wall thickness as well as imaging of plaque size and composition is highly reproducible [71,73].
The predictive value of carotid vessel wall thickness measurement by MRI for future cardiovascular events is currently unknown. However, MRI of vessel wall thickness is highly correlated with ultrasound carotid intima-media thickness (CIMT), so it is reasonable to expect that the predictive value of both modalities will be in the same order of magnitude [71]. Several smaller studies have been conducted to investigate the relation between plaque size and composition and future cardiovascular events [74-77]. Various observational studies have investigated the relationship between traditional cardiovascular risk factors and carotid vessel wall thickness and plaque size and composition measured with MRI [78,79]. Furthermore, atherosclerosis imaging with MRI has shown to be a valuable modality for the assessment of atherosclerosis progression in specific patient populations, as well as for the assessment of cardiovascular drug efficacy [21,22,80-82].
As such, CIMT has proven a strong research tool if well monitored and quality controlled [4,7,9,11-13].
SUMMARY AND RECOMMENDATIONS
●Background – Ultrasound carotid intima-media thickness (CIMT) is a widely available, safe, and reproducible measure when performed by trained and certified sonographers with standardized equipment. (See 'Introduction' above and 'Image acquisition' above.)
●Use in cardiovascular disease prediction – In patients without known cardiovascular disease (CVD), composite measures of CIMT of various segments, with or without the presence of atherosclerotic plaque, result in modest but statistically significant increases in net reclassification improvement when added to risk estimates based on traditional risk factors or risk models. Whether the improvement in risk stratification due to CIMT and plaque information translates into measurable health benefits for individuals or into reductions in the cost of care has not been established. We do not recommend routine CIMT measurement to assess cardiovascular disease risk in the individual patient. (See 'CIMT and future cardiovascular events' above.)
●Use in research – CIMT is a well-validated research tool to investigate atherosclerosis development in epidemiological studies, provide insight in cardiovascular disease risk in specific populations, and to evaluate efficacy of preventive cardiovascular therapies in clinical trials. (See 'CIMT for the assessment of drug efficacy' above.)
●No role in monitoring drug efficacy – CIMT is not recommended to be routinely used in clinical practice to monitor the effects of medical therapy in the individual patient. (See 'CIMT for the assessment of drug efficacy' above.)
●MRI-based measurement – CIMT can be measured using magnetic resonance imaging (MRI). However, given the additional complexity and cost of MRI when compared with ultrasound, CIMT measurement should not primarily be performed with MRI in most patients. (See 'CIMT measurement by MRI' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff thank Dr. John J. P. Kastelein for his past contributions as an author to prior versions of this topic review.
