Non-invasive imaging of flow and vascular function in disease of the aorta

Abstract

With advancements in technology and a better understanding of human cardiovascular physiology, research as well as clinical care can go beyond dimensional anatomy offered by traditional imaging and investigate aortic functional properties and the impact disease has on this function. Linking the knowledge of the histopathological changes with the alterations in aortic function observed on noninvasive imaging results in a better understanding of disease pathophysiology. Translating this to clinical medicine, these noninvasive imaging assessments of aortic function are proving to be able to diagnosis disease, better predict risk, and assess response to therapies. This review is designed to summarize the various hemodynamic measures that can characterize the aorta, the various non-invasive techniques, and applications for various disease states.

Introduction

Recognized as a conduit for distributing blood, abnormalities of the functional properties of the aorta are increasingly realized as important contributors to cardiovascular (CV) disease. This review is designed to summarize the various functional properties of the aorta, non-invasive imaging methodologies used to assess these properties, and abnormalities of these properties that contribute to CV disease.

The aorta performs several important functions. First, the aorta transmits and distributes blood from the heart to medium sized conduits that supply the periphery. Second, the aorta acts as an expandable reservoir that buffers the pulsatile force from left ventricular contraction (1). The Windkessel principle, first outlined in the late 1800’s, models the aorta as an elastic reservoir absorbing the systolic blood flow from left ventricular contraction and then releasing this blood by elastic recoil during diastole providing more constant (rather than pulsatile) blood flow to the periphery (2). Finally, a healthy aorta limits the augmentation phenomenon from the reflective pressure wavefronts that return toward the heart from the medium sized conduits that supply the periphery (3). As the pressure wavefront from ventricular contraction propagates down the vasculature tree, reflections of these wavefronts return from the periphery and amplify the systolic pressure within the aorta that thereby increases left ventricular afterload. This mechanical interplay between the left ventricle and the arterial tree is termed ventricular arterial coupling with the aorta playing a integral role in this relationship (4). In healthy normals, under a wide range of vascular loading conditions, the aorta allows for optimized work efficiency in conjunction with the left ventricle (5); however, alterations in the vessel wall and loss of these properties that occur with aging and disease processes can reduce circulatory efficiency (6).

Optimal function of the aorta is related to the structure and composition of three regions of the wall of the aorta (Figure 1): the intima, media and adventitia (7). The intima is comprised of the endothelium, the endothelial glycocalyx which is a network of glycosaminoglycans bordering the vessel lumen (8), and the endothelial space which contains collagen and sparse smooth muscle cells (7). With aging, flow mediated vaso-reactivity is diminished (9) and the intima exhibits increased collagen deposition and disoriented vascular smooth muscle cells (10).

Figure 1. Cross-sectional diagram of the aortic vessel with the components of the three layers in a healthy young individual (20 year old) and in an elderly individual (70 year old) demonstrating the effects of arterial aging.

With aging, the endothelium experiences thickening with hyperplasia and collagen deposition. The media has increased disorganized smooth muscle cells, increased collagen, and increased elastin fracturing and cross-linking. The adventitia demonstrates increased collagen deposition.

The media, the thickest layer of the aorta and separated from the intima by the internal elastic lamina contains layers of overlapping sheets of the extracellular matrix proteins elastin and collagen and vascular smooth muscle cells (11). Elastin is capable of stretching over 300% of its original length but fractures at relatively low levels of stress (12). Collagen is less pliable but is capable handling much higher stress loads and thus provides significant strength to the aorta (12). In the proximal thoracic aorta, an elevated elastin to collagen ratio heightens vascular compliance; whereas distally in the abdominal aorta, collagen and smooth muscle predominate leading to a stiffer conduit (13, 14). With aging, the percent of collagen increases in the media (15) and cross-linking and abnormal breakdown of the proteins due to mechanical fracturing as well as oxidation and nonenzymatic glycation occurs (6, 16).

The adventitia is the outermost layer and is composed mainly of collagen, fibroblasts and fibrocytes (7). The adventitia reinforces aortic wall strength (to limit excessive aortic distension) and also anchors the vessel to the surrounding tissues. The adventitia also contains the vaso vasorum, the supporting blood supply to the cells of the aorta. With aging, increased collagen accumulation from fibroblasts have been seen in animal studies contributing to stiffening (17). At a macroscopic anatomic level, the aorta diameter grows in relationship to increasing body surface area during the time from infant to young adulthood (18). With further aging, the arterial wall thickens, predominantly in the intima and the aorta dilates, greater in the ascending aorta as compared to the abdominal aorta (19).

Vascular Biomechanics

Both simple and complex measures quantify the biomechanical functional properties of the aorta (Table 1). As shown in Figure 2, the simplest measures including blood velocity and flow (which couples velocity with vessel area to measure the volume of fluid traversing a defined point per unit time).

Table 1.

Noninvasive imaging modalities useful in the aorta and subsequent measurements of aortic flow and function capable by each modality

Modality	Measurement
Echocardiography	Velocity Flow Tissue strain Pulse wave velocity Distensibility Pressure gradient
Cardiovascular CT	Distensibility Tissue Strain
Cardiovascular MR	Velocity Flow Flow visualization and pattern Flow displacement Pulse wave velocity Distensibility Pressure Gradient and Maps Wall shear stress

Figure 2. Cross sectional diagram of an aorta demonstrating the biomechanical measures of velocity, flow, and stress.

The directionality of the different types of stress within the vessel wall are also shown.

Additional simple measures include stress, strain and Young’s modulus which relates the two. Stress is defined as the force per unit area applied to an object and within a blood vessel, specific types of stress are defined by the direction of force include radial, circumferential and longitudinal stress (Figure 2). For example, arterial pressure exerts outward force on the vessel wall resulting circumferential stress. Strain, ɛ, quantifies the deformational response of the tissue to a force:

$$\epsilon =\frac{L-L_0}{L_0}$$ Equation 1

where L is the final dimension after applied stress and L₀ is the starting dimension. The relationship between stress and strain in a material is defined by the constant, Young’s modulus which is the ratio of stress to strain. A very rigid material has a very high Young’s modulus (ie. with stress, the aortic tissue deforms very little).

When fluid such as blood, travels within a structure, there is also wall shear stress which represents frictional force of the fluid parallel to the flow of the fluid. In other words, blood flowing along the vessel wall creates a “drag” force in the direction of blood flow on the endothelial cells (Figure 2).

Building upon these basic equations, more complex measures of aortic function have been derived. Arterial compliance represents the change in the diameter of a vessel divided by the change in the distending pressure within the vessel. This essentially the reciprocal of Young’s modulus in a cylinder. Arterial distensibility is similar to compliance but is normalized for the size of the vessel. Aortic distensibility is the degree to which the aorta can expand such that a stiff vessel has low distensibility. Using 2D imaging techniques, distensibility can be expressed follows (20):

$$\mathit{\text{Distensibility}}\left(\frac{1}{\mathit{mmHg}\times {10}^3}\right)=\frac{(\mathit{\text{SA}}-\mathit{\text{DA}})}{\mathit{\text{DA}}\times (\mathit{\text{SBP}}-\mathit{\text{DBP}})}$$ Equation 2

where SA is the maximal systolic area, DA is the minimal diastolic area, SBP is systolic blood pressure millimeters of mercury (mmHg) and DBP is diastolic blood pressure. Due to the peripheral amplication phenomenon, brachial blood pressures do not always accurately reflect central blood pressures, particularly in younger subjects and thus, should not be used as a surrogate (21). Noninvasively, several methods to estimate central blood pressure have been developed using pressure waveforms taken from carotid, brachial or radial arteries and calibrated to cuff pressures. The strengths and limitations of these techniques are reviewed elsewhere (22). Pulse wave velocity (PWV) is another measure of aortic stiffness and is related to the Young’s modulus by the Moens-Korteweg equation (1):

$$\mathit{\text{PWV}}=\sqrt{\frac{E\times h}{2\times r\times \rho }}$$ Equation 3

where E is circumferential Young’s modulus, h is vessel wall thickness, r is luminal radius and ρ is density of blood. Recall earlier that Young’s modulus is the ratio of stress to strain in a material; thus, a rigid material (high Young’s modulus) will result in a high PWV, all else equal. Descriptively, PWV represents the measure of how fast the systolic pulse from a volume of blood being ejected from the left ventricle travels down the aorta. In general, a stiffer more rigid vessel transmits the pulse faster (higher PWV). PWV and distensibility are inversely related to one another the Bramwell-Hill equation (23):

$$\mathit{\text{PWV}}=\frac{3.57}{\sqrt{\mathit{\text{Distensibility}}}}$$ Equation 4

when units of mmHg and meters per second are used. In clinical practice, PWV is calculated by recording (commonly using transducers locating at the carotid artery and femoral artery, hence cfPWV) two locations separated by a known distance. By recording at the same time or by triggering relative to the electrocardiogram, the transducers determine the difference in the arrival time of the systolic wavefront at each location. The distance, in meters, along the aorta between the two transducers is divided by the difference in arrival time, in seconds, yielding a velocity in meters per second (m/s).

$$\mathit{\text{PWV}}(m/s)=\frac{\mathit{\text{Distance}}}{\mathit{\text{difference in pulse arrival times}}}$$ Equation 5

Accurate PWV determinations rely on several factors. First, one must accurately measure the distance along the blood vessel that the pulse wave has traversed. Body surface estimations can be affected by body habitus such as obesity or by tortuosity of the blood vessels that is unknown when simply estimating the path by looking at the body surface. Second, the fact that blood is flowing in opposite directions in the carotid and femoral arteries needs to be taken into account either by measurement technique or converted via formula (24). Finally, cfPWV is considered a global measure of aortic stiffness; however, the ascending aorta from the root to the takeoff of the right brachiocephalic is not included due to measuring from the peripheral sides of the carotid and femoral arteries. This area may represent a clinically important source of aortic stiffening in certain disease states.

Pressure gradients within the aorta can be estimated from Doppler echocardiography and phase contrast CMR by utilizing Bernoulli’s principle. With several assumptions, the equation relating velocity at two locations (v_b, v_a) to pressure is the modified Bernoulli’s equation (25):

$$\mathit{\text{Pressure Difference}}=4(v_b^2-v_a^2)$$ Equation 6

when using the units of mmHg and meters/second.

Newer imaging technologies have provided additional visualization capabilities including three dimensional flow changes with time. This has led to additional measures including blood flow eccentricity or displacement (26), pressure gradients (27), and blood turbulence quantification (28). These are further discussed in the CMR section.

Imaging Modalities

Echocardiography

Echocardiography uses ultrasound waves and their differential reflection from body tissues to create 2D and 3D images of cardiovascular structures (29) as well as quantifying velocity of structures by employing the Doppler effect (30). With known blood velocity over a period of time (velocity time integral or VTI) and vessel area, the volume of blood per heart beat traversing the vessel can be quantified (30). Additional computation allows for the assessment of strain within the wall of the aorta using tissue Doppler (31) or speckle tracking technology (32). Focusing on the individual components of the arterial wall, the longitudinal strain and wall shear stress within the 3 layers of the aortic wall can also be evaluated with echo (33). Additionally, Doppler derived velocity measurements allow one to calculate pressure differences within the aorta using modified Bernoulli’s equation (Equation 6) (34).

Distensibility assessed by transthoracic echocardiography (TTE) is typically measured in the ascending aorta just above the aortic valve using the M-mode aortic diameter at end-systole and end-diastole. This technique has shown excellent correlation with invasive measurements (35) and can be used to measure other aortic regions including the abdominal aorta (36). Transesophageal echo (TEE) can be used in a similar fashion to TTE using M-mode tracings (37) (Central Illustration, A) or 2D cross sectional cine images (38) with the advantage of measuring multiple levels of the descending aorta as to reduce the risk of sample bias at a single location.

Central Illustration.

Examples of different imaging modalities used to assess aorta flow and function: A.) Transesophageal echo measuring aortic distensibility using a M-mode tracing of the descending aorta. The systolic and diastolic diameters are measured from the tracing and used with the central blood pressure to calculate distensibility B.) Gated cardiac computed tomography assessing aortic distensibility. A sagittal image of the aorta arch displays the location of the axial plane gated at systole and diastole. The systolic and diastolic areas of the ascending aorta are used with the central blood pressure to calculate distensibility C.) Two dimensional phase contrast cine CMR measuring pulse wave velocity. A sagittal image of the aorta arch displays the location of the axial plane and is used to calculate the vessel centerline distance between the ascending and descending aorta. Through plane phase contrast and magnitude images in the axial plane are segmented for the ascending and descending aorta and velocity versus time flow curves are generated (red is from the ascending aorta, blue is from the descending). The time difference between the arrival of two waveforms is determined. Pulse wave velocity is calculated by dividing the distance by the arrival time difference. Distensibility of the ascending and descending aorta can be calculated from the change in diastolic and systolic areas from the magnitude images and the central blood pressure. D.) Cardiovascular magnetic resonance using 4D flow assess multi-plane pulse wave velocity 1.Ten analysis planes placed equidistantly along the 3 dimensional segmented aorta containing 3 direction velocity information 2. Ten through-plane velocity versus time waveforms derived from the ten analysis planes. The time difference in arrival for each waveform as compared to the first waveform is calculated. 3. Plot of arrival time difference versus distance along the aorta. The inverse of the linear regression line is the multiplane averaged PWV for the entire aorta. E.) Aortic flow visualization by streamlines in 4D flow CMR in a 67 year old female without known aortic disease. Streamlines connect all the vector field points into a line and represent the instantaneous flow field within the aorta. Three different phases of peak systole, late systole and diastole are shown and the velocity the streamline is represented by color (red is faster, blue is slower).

Three dimensional (3D) TEE (39) and TTE (40) can also be used to measure distensibility of the aorta. The advantage of these techniques relates to the acquisition of a large 3D volume data set of the aorta from which post-exam processing allows for measuring multiple regions of interest. Pulse wave velocity can be measured using ultrasound Doppler in similar fashion to the pressure transducer technique described above. By placing the ultrasound probe at the carotid artery and the femoral artery, the Doppler derived arrival time difference of the velocity wave forms can be calculated. In comparison with applanation tonometry derived cfPWV, correlation is excellent (41, 42).

Cardiac Computed Tomography (CT)

Gated to the cardiac cycle, cardiovascular CT is able to produce tomographic images of the aorta at different phases in the cardiac cycle (43). Using this technology, distensibility of the aorta can be assessed (Central Illustration B). In a similar fashion to echocardiography, the maximum and minimum cross-sectional areas of the aorta at a region of interest are measured and the distensibility is calculated with the pulse pressure (Equation 2) (20). Benefits to this technique include the excellent spatial resolution of CT as well as the used of area measurements instead of simple diameters. One disadvantage is the decreased temporal resolution which may lead to underestimation of distensibility if the time point at which maximal distention is not captured. The major drawback to this technique relates to the ionizing radiation exposure associated with CT imaging as it requires continuous CT imaging during the entire cardiac cycle (20).

Cardiovascular Magnetic Resonance (CMR)

Cardiovascular magnetic resonance utilizes strong magnetic fields and non-ionizing radiation to form images from the target molecules’ response to radiowave energy input (44). By adjusting the input signal, various tissue properties can be highlighted and quantified and images formed (44). The measure of velocity can be quantified using phase-contrast techniques which is based upon blood flowing in a magnetic field results in a phase shift which is proportional to the velocity (45). By acquiring multiple velocity maps over the entire heart cycle, a cine can be produced showing the velocity changes over time and subsequently, with vessel area, the blood flow can be calculated (45). Similar to echocardiography, the measures of velocity and flow are applied to the blood traveling in the aorta and are useful in disease states such as coarctation (46) with the advantages of complete thorax coverage for collateral localization and extremely accurately flow quantification.

Using steady state free precession (SSFP) cine imaging, CMR can also measure aortic distensibility by calculating the systolic and diastolic areas of any region with the aorta and applying equation 2. Studies have also demonstrated the ability to substitute gradient echo PC imaging for SSFP which has the advantage of decreasing the imaging time if PWV from PC imaging is also being calculated (47).

Pulse wave velocity can be measured using 2D phase contrast cine velocity or flow curves to calculate pulse wave arrival time at various locations within the aorta. Central Illustration C demonstrates a common application of calculating the aortic arch PWV (48) and compared with invasive pressure catheters, CMR has demonstrated good agreement (49). Due to the widespread availability of this CMR technique, several publications of age stratified normals have been produced (50–53). Benefits of the 2D technique are widespread familiarity with 2D PC imaging, regional assessment capabilities, relatively short scan times, and good temporal resolution (10 to 20 milliseconds).

Additional improvements to measuring PWV by CMR include the use of multiple planes which improves the accuracy of calculating pulse wave velocity (54). With this technique, multiple points of analysis along the length of the aorta are utilized and linear regression is employed to determine PWV in the region of interest (see Figure 4A as applied to 4D flow imaging). This technique has been further developed and enchanced with the addition of 2 directional in-plane velocity encoding which improves the accuracy of this measurement (55).

Recently, improvements in the acquisition of multi-directional phase contrast imaging have led to significantly more research utilizing this technique to investigate aortic pathophysiology. Three dimensional three directionally encoded cine velocity imaging (also known 3D cine phase contrast or used in this paper, 4D flow) acquires velocity information in all three directions at all locations within a defined volume. Utilizing post-processing software to calculate PWV, the velocity waveforms can be acquired post-imaging for any regional location and the technique of multiple plane and linear regression can be employed to diminish the impact of potential erroneous values (Figure 4A) (56, 57). The acquisition time (5 to 15 minutes) and complexity of post-processing software has thus far limited the clinical utility of multi-planar techniques. 4D flow technology also allows for unique visualization of the blood flow within the aorta. Several different techniques for visualizing this flow have been developed. These include vector fields, streamlines, and pathlines (58). Choice of visualization depends upon the application of 4D flow, though often they are used in combination as the different techniques offer unique perspectives of blood flow in the aorta (Figure 4B).

With additional post-processing, 4D flow technology also provides the ability to calculate more complex hemodynamic measures including wall shear stress, flow displacement, helical flow, turbulence and pressure gradients. Wall shear stress is the frictional force of blood on the endothelium and with the assistance of computational software, can be calculated from 4D flow datasets (59). Wall shear stress has been implicated in vascular remodeling, endothelial dysfunction, and development of atherosclerosis (59–62). Flow displacement is a measure of how “off-center” the blood flow is from the center of the vessel lumen and semi-quantifies abnormal flow within the aorta (26). Visualization and semi-quantification of abnormal helical flow patterns has also been developed using 4DFlow (63, 64). Turbulence defined as non-laminar and chaotic flow, is a contributing factor to aortic disease processes and its intensity quantified (28). An extension of this quantification of turbulent kinetic energy, the energy within the areas of chaotic flow, which has applications to post stenotic flow and pressure gradients (65). Pressure gradients are typically used to estimated the severity of an obstruction such as coarctation and can now be calculated using complex computational fluid dynamics (27).

Application to Aorta in Disease States

Flow and functional measurements can be applied to aortic diseases be used in the investigation of pathologic mechanisms. Additionally, these measures may potentially aid in diagnoses, prognosis and assessing disease severity.

Arteriosclerosis and Aging

Defined as the thickening and stiffening of blood vessels, arteriosclerosis has been extensively studied with non-invasive imaging to determine its impact on function (24, 53, 54). Intertwined with vascular aging and hypertension, the changes associated with arteriosclerosis are described in Figure 2. For example, the PWV increases from approximately 7 m/s at age 30 and increases to 12 at age 70 (24). This almost doubling of PWV implies an almost four-fold change in the elastic modulus (Equation 3).

Previously thought to be simply a part of the normal aging process, abnormal arterial stiffening has been demonstrated as a predictor of adverse outcomes. Arterial stiffening leads to increased systolic and widened pulse pressures due to loss of both the ventricular contraction cushioning as well as loss of the reflective wave dampening (6). These compound effects on afterload are thought to contribute to left ventricular hypertrophy and ultimately, predispose one to left ventricular failure (66, 67). Furthermore, it has been hypothesized that the loss of dampening also results in increased pulsations being transmitted to the vulnerable tissues such as the kidney and brain leading to microvascular and end-organ damage (68).

Using the Framingham cohort, carotid-femoral PWV (cfPWV) was demonstrated to be an independent risk factor for first CV event and improved risk prediction when added to the Framingham risk model (69). A recent meta-analysis of over 15,000 participants showed cfPWV to be an independent predictor of cardiovascular events and overall mortality in the general population (70). Furthermore, it was estimated that an increase of 1 m/s increased the risk of mortality increases 15% even after adjusting for known cardiovascular risk factors. The strength of findings like this have led to the recommendation by the European Society of Cardiology to include PWV assessment in the risk stratification of hypertensive individuals as a means of determining end-organ damage (71).

Similar predictive value of CMR-derived aortic distensibility for all cause mortality was shown in a population free of known CV disease from the Multi-ethnic Study of Atherosclerosis (72). In those with low and intermediate risk for CV disease, aortic distensibility also independently predicted CV events. In the Dallas Heart study, elevated pulse wave velocity of the aortic arch by CMR was independently associated with nonfatal extra-cardiac vascular events with a modest improvement to the Framingham risk score (73).

Atherosclerosis

A distinct concept to arteriosclerosis but closely related by common, overlapping pathologic processes, atherosclerosis includes lipid deposition, inflammation, fibrosis and calcification within the vessel wall (74). Though several studies have demonstrated aortic stiffness is associated with atherosclerotic risk factors such as hyperlipidemia, smoking, obesity, and diabetes mellitus, an interesting meta-analysis suggests that these associations are rather weak and that age and hypertension have the strongest degree of association with cfPWV (75). Similar to vascular aging, atherosclerosis adversely remodels the arterial wall leading to a loss of normal function. Atherosclerosis and calcification of the coronary arteries as well as the thoracic aorta was shown to be strongly associated with decreased gated CT derived distensibility in the aorta (76). This was further demonstrated in longitudinal evaluation of aortic distensibility by gated CT and coronary artery atherosclerosis which revealed parallel progression of coronary artery atherosclerosis and worsening aortic distensibility even after accounting for known risk factors including age (77).

Furthermore, cfPWV has is independently associated with other noninvasive markers of atherosclerosis and end-organ damage including ankle-brachial index and the presence of brain white matter hyperintensity (78). Flow visualization has also been used to assess atherosclerotic disease. In a novel investigation utilizing 4D flow CMR imaging, it was demonstrated that the descending aorta was a potential source for cryptogenic embolic stroke. Cine visualization of the particle trace lines within the descending aorta demonstrated retrograde aortic flow distal to the ostium of the left common carotid at the location of atheromatous plaques in the aorta (79).

Abdominal Aneurysm

Thought to be due to a complex interaction of genetic pre-disposition and exposure to atherosclerotic risk factors, abdominal aneurysms are defined by degradation of the elastic media and atheromatous changes in the intima (80). Using gated CT to link abnormal vessel mechanics to rupture, the aneurysmal portion of the abdominal aorta was found to be markedly less distensible than the more normal regions; however, there was no association between distensibility and aneurysm diameter, the traditional validated risk factor for aneurysm rupture (81). Additionally, 4D flow imaging has been demonstrated as a technique to visualize post-endovascular repair complications such as endoleaks, with the benefit of no intravenous contrast and quantification of direction and magnitude of the leak (82) though distortion artifacts particularly with stainless steel stents may limit the applicability of this technique.

Bicuspid Aortic Valve

The pathogenesis of the aortic abnormalities associated with bicuspid aortic valve as assessed by echo may be due to a primary aortopathy or a consequence of abnormal hemodynamics or a combination of the two (83). Cardiac MR reinforces these findings demonstrating that PWV in both the arch and descending aorta were increased in individuals with nonstenotic bicuspid valves and left ventricular mass and degree of regurgitant dysfunction were correlated to PWV (84).

In CMR, 4D flow has been used to help find potential hemodynamic markers that pre-dispose individuals to root dilation and aneurysm formation. Jet direction and abnormal helical flow were two qualitative measures associated with bicuspid aortic valve (64). Other studies have demonstrated increased flow abnormalities associated with cusp-fusion subtype. These abnormal flow patterns lead to increased wall shear stress in those with bicuspid aortic valves (60–62). In a longitudinal study, flow displacement as measured by 4D flow sequences was found to be most correlated with growth of the aorta in bicuspid valve patients (85). A transesophageal echo study of the descending aorta also demonstrated that in bicuspid aortic valve with severe aortic stenosis had altered distensibility as compared to those with tricuspid valve aortic stenosis suggesting intrinsic alteration of aortic properties in bicuspid extending to the descending aorta (32).

Chronic Kidney Disease

Cardiovascular disease is a major cause of morbidity and mortality in individuals with chronic kidney disease (CKD) and investigators have utilized measures of aortic function to further delineate this relationship (86–88). As assessed by CMR and echo, patients with early stage CKD (86) as well as end stage CKD (87) have reduced aortic distensibility. This increased aortic stiffness by means of cfPWV measurement have been shown to be an independent risk factor allcause mortality and CVD events in advanced CKD patients and improved risk prediction above traditional risk factors (88).

Vasculitis

Kawasaki disease and Behcet’s disease have been demonstrated to have detrimental effects on the function of the aorta, even in the absence of active aortic involvement of the disease (89–91). Using echo, individuals with Behcet’s disease were demonstrated to have both reduced aortic strain and distensibility as compared to age-matched controls (89, 90). In a cohort of Kawasaki’s disease, pulse wave velocity in the aortic arch was elevated as assessed by echo as compared to controls despite apparent resolution of the vasculitis (91).

Toxin Exposure

Anthracycline based chemotherapy, used most frequently in breast cancer, lymphomas and leukemias has been associated with stiffening of the proximal portion of the aorta. Hypothesized to be due to oxidative stress vascular matrix and endothelial dysfunction, aortic distensibility was adversely affected at 4 months post anthracycline as compared to controls (92). A similar study demonstrated that the onset of adverse aortic stiffening was evident at 1 month after cancer treatment initiation and at 6 months. Furthermore, the aortic stiffening occurred concurrently with reductions in left ventricular ejection fraction and circumferential strain indicating potential aorto-ventricular interaction which may be responsible for the longterm cardiotoxic complications of cancer therapy including congestive heart failure (93).

Hypertrophic Cardiomyopathy

An inherited disorder associated with mutations in multiple genes encoding proteins of the cardiac sarcomere, abnormalities in aortic function have also been found. Aortic stiffness as assessed by CMR derived PWV in the arch was found to be elevated in individuals with hypertrophic cardiomyopathy. Additionally, those with extensive myocardial fibrosis tended to be even stiffer (94). Echo derived aortic distensibility and vessel wall strain also have been shown to be reduced further supporting the notion that altered aortic elastic properties are present in patients with hypertrophic cardiomyopathy (95).

Marfan’s Syndrome

An autosomal dominant genetic disorder of the fibrillin protein which is present in the vascular wall, aortic pathology is common in Marfan’s patients (96). In a case-control study using CMR, the Marfan’s patients exhibited lower distensibility and increased PWV in the ascending aorta as compared to healthy subjects (p<0.01). With beta blocker therapy, the Marfan’s patients experienced a decrease in PWV and an improvement in ascending aortic distensibility and the control patients had no such changes (97). Furthermore, using CMR derived PWV in Marfan’s patients, the baseline aortic stiffness measurements were able to predict the absence of luminal dilation at 2 years of followup offering another potential marker of risk stratification (98). 4D flow has shown potential for monitoring for complications post valve sparing surgical replacement (99). Transesophageal echo also demonstrated that distensibility and tissue doppler strain also could discriminate Marfan’s from normals and predict dilation and dissection (100).

Coarctation of the aorta

The congenital condition of narrowing at the ductus arteriosus was previously diagnosed with imaging using conventional angiography but now CMR and echocardiography can both visualize the stenosis as well as assessing the hemodynamic significance of the obstruction without the risks of ionizing radiation or nephrotoxic contrast agents. Using Doppler waveforms to measure the maximum velocity in the descending aorta, the peak pressure gradient at the stenosis was similar to cardiac catheterization (34).

Cardiac MR can demonstrate the existence of the narrowing and with PC imaging, assess both the pressure gradient across the coarctation and quantify the collateral flow within the thorax (46) and thereby select those in need of surgical repair. 4D flow imaging has also been demonstrated as a method of quantification of collateral flow in patients with coarctation (101). With advanced post-processing techniques, 4D flow has also been demonstrated to accurately determine the peak to peak pressure gradients as compared to invasive catheterization (27). Additionally, non-invasive imaging can monitor postoperatively the adequacy of the correction as well as monitor serially for complications such as restenosis or aneurysm formation (46) or adverse remodeling of the aorta due to arch shape (102).

Future Investigations

Future areas of research key to improving this multi-modality approach to aortic disease include standardization of methodology in acquiring the measurements and creation of larger normative databases that have been normalized for age and gender. Subsequent development of normal and abnormal thresholds for each modality would allow for comparison across modalities and is key for integration into clinical practice.

Ongoing areas of research include the investigation of regional assessments of aortic stiffness and how certain disease processes may affect the proximal aorta, high in elastin content, as compared with the abdominal aorta which has an increased propensity for atherosclerotic processes. Response to pharmacologic and non-pharmacologic interventions in various disease states is another important area of ongoing research that will aid with clinical decision making.

One area of rapidly expanding research is 4D flow imaging in CMR. With improvement in sequences, better post-processing software and validation of techniques, the applications to various aortic disease states is anticipated to grow rapidly. With these improvements, larger longitudinal clinical trials are necessary to demonstrate the clinical utility of the novel hemodynamic measures and visualization capable with the technology. Another example of an innovative application of this technology is combining the cutting edge technologies of rapid prototyping with 3D printing and 4D flow CMR imaging to construct hemodynamically accurate models of the aorta (103). With these models, patient specific modeling and testing of procedural interventions and their impact on the aortic function is then possible.

Summary

The aorta is a complex structure, efficiently designed to transmit blood, buffer the vascular system from the pulsative ventricular contraction, and minimize reflective waves. The composition of the aorta is optimized to achieve these functions. In a complex interplay between changes in the composition of the aorta and pathologic hemodynamics, aging and disease cause perturbations in both the composition and hemodynamics. Non-invasive imaging techniques using echocardiography, cardiovascular CT and CMR have been developed and can enable quantification of the changes in hemodynamics and improve diagnosis and risk prediction. Furthermore, these imaging techniques can assess the response of the aorta to both pharmacologic as well as surgical procedures and further predict prognosis or complications. With further development and research including of longitudinal studies, non-invasive imaging assessment of aortic function and flow will likely become an integral part of the evaluation and diagnosis of aortic disease.

Table 2.

Aortic diseases and potential applications of hemodynamic measurements of potential value

Disease	Non-invasive imaging assessment of aortic function or flow	Potential Value
Arteriosclerosis/hypertension	○PWV ○Distensibility	○Assess risk of cardiovascular event and overall mortality
Abdominal Aneurysm	○Flow	○Visualize and quantify complications (eg. endoleaks) post endovascular repair
Chronic kidney disease	○PWV ○Distensibility	○Assess risk of cardiovascular events and overall mortality
Vasculitis	○Strain ○Distensibility ○PWV	○Assess extent of detrimental hemodynamic effects of inflammation
Cancer therapy	○PWV ○Distensibility	○Investigate as a mechanism for left ventricular dysfunction and increased CV events associated with cardiotoxic cancer therapy
Bicuspid aortic valve	○PWV ○Distensibility ○Flow pattern and flow displacement ○Wall shear stress	○Assess risk of aortic root dilation
Hypertrophic cardiomyopathy	○PWV	○Investigate the association of aortic stiffness with increased myocardial fibrosis
Marfan’s syndrome	○Aortic wall strain ○Distensibility of the aortic root ○Regional PWV ○Flow pattern	○Assess risk of aortic root dilation and dissection. ○Quantify response to pharmacology therapy. ○Post-operatively, visualize and quantify risk of adverse remodeling.
Coarctation of the aorta	○Pressure gradient ○Aortic and collateral flow	○Assessment of the severity of stenosis and quantify collaterals. ○Assess adequacy of stenosis correction and post-operatively, predict risk of restenosis or aneurysm formation

Abbreviations

CMR

cardiovascular magnetic resonance

CV

cardiovascular

PWV

pulse wave velocity

CT

computed tomography

cfPWV

carotid-femoral pulse wave velocity

PC

phase contrast

2D

two dimensional

4D

four dimensional

mmHg

millimeters of mercury

m/s

meters per second

SSFP

steady state free precision

mm

millimeters

TTE

transthoracic echocardiography

TEE

transesophageal echocardiography