Abstract
Contrast echocardiography is used for a variety of clinical applications including the delineation of endocardial borders to assess wall motion, enhancement of the blood pool to better appreciate intra-cavitary abnormalities such as thrombi and masses, and assessment of myocardial microvascular perfusion. Contrast echocardiography relies on the ultrasound detection of acoustically active nano- or microparticles that undergo cavitation in an ultrasound field. In their most common form, microbubble contrast agents are composed of a gas core containing a high molecular weight, low-solubility gases; and are encapsulated by lipid, albumin, or polymer shells which reduce gas diffusion and lower surface tension. This chapter describes key technical elements of clinical and research contrast echocardiography including: (1) microbubble composition and characteristics, (2) microbubble behavior in the vasculature and their interactions with native cells and plasma proteins, (3) unique acoustic properties of microbubbles and how they can be leveraged to specifically detect contrast signal during clinical ultrasound imaging, and (4) safety of microbubble contrast agents.
Keywords
Cavitation, Contrast echocardiography, Microbubbles, Microcirculation
Introduction
Contrast echocardiography is a broad term used to describe an array of approaches that can be used to improve and expand diagnostic capabilities by acoustic enhancement of the blood pool during cardiac ultrasound imaging. Ultrasound contrast agents are generally composed of gas-filled encapsulated microparticles, usually microbubbles that are 1–5 μm in diameter, or nanoparticles. The most common clinical application of contrast echocardiography has been to better delineate the endocardial contours of the left ventricular (LV) cavity, termed left ventricular opacification (LVO; Fig. 3.1 ). Although there are many reasons clinicians opt for performing LVO in a given patient, the most frequent indication is to better evaluate global or regional LV systolic function ( 
). Justification for this application of LVO is based on (1) the inability to fully examine LV myocardial thickening in 10%–20% of unselected patients; (2) the frequent use of echo to guide management in critically ill patients who have difficult acoustic windows due to positive pressure ventilation or inability to cooperate with the ultrasound examination; and (3) frequent use of echo to make critical decisions based on the presence of segmental wall motion, where every myocardial segment needs to be well seen with a high degree of reader confidence (e.g., stress echocardiography, point-of-care echo for detection of myocardial ischemia or evaluation of heart failure). There are many other clinical situations where LVO has had a positive impact in clinical echocardiography ( Box 3.1 ).
Detection of segmental wall motion abnormalities (rest or stress)
Quantification of LV ejection fraction
Quantification of LV volumes
Detection of LV or atrial thrombus
Detection and characterization of LV masses
Confirmation of the presence of apical hypertrophic cardiomyopathy
Evaluation of eosinophilic cardiomyopathy
Detection of ventricular pseudoaneurysms
Evaluation of ventricular noncompaction cardiomyopathy
Augmentation in Doppler signals
Detection of aortic thrombus or dissection
LV, Left ventricular.
Refinements in contrast ultrasound technology that improve the detection of microbubble signal in the coronary circulation relative to myocardial tissue signal have permitted the imaging of the myocardial microcirculation. These techniques are broadly referred to as myocardial contrast echocardiography (MCE). The most basic approach to MCE is to spatially evaluate the presence of an intact microcirculation. The presence of a functional microvascular bed can be used to assess myocardial viability, to characterize a cardiac mass as a tumor rather than thrombus based on the presence of functional microvessels, and to detect therapeutic or spontaneous reperfusion in acute myocardial infarction ( Fig. 3.2 ; 
). Quantitative or semiquantitative assessment of perfusion requires not only quantification of the intact microvasculature but also temporal information of microbubble transit through the microcirculation. This measurement generally requires destruction of microbubbles within the acoustic sector and evaluation of signal reappearance. This approach can be used to detect resting ischemia, flow heterogeneity during stress echocardiography, or microvascular dysfunction, or to assess the presence/adequacy of collateral blood flow.
In this chapter, the basic principles of contrast echocardiography will be described, including an overview of contrast agents and the specific imaging modalities that have been developed to improve microbubble signal-to-noise ratio during clinical imaging. Clinical applications of contrast echocardiography are detailed in Chapter 12 .
Microbubble Contrast Agents
Signal enhancement during contrast echocardiography relies on the dynamic interaction of ultrasound pressure waves, with a highly compressible and expandable particle that is smaller in scale than the wavelength of ultrasound applied. As will be described later, particle expansion and compression during ultrasound pressure peaks and nadirs, respectively, produces volumetric oscillations of these particles, which is the primary source of ultrasound signal generation. The rationale for using microbubbles, as ultrasound contrast agents is based on their compressibility/expandability, and on their in vivo stability. Air and high-molecular-weight gases that have been used in microbubble contrast agent preparations are several orders of magnitude more compressible than water or tissue. During most forms of clinical contrast echocardiography, contrast oscillation and the subsequent acoustic energy response occurs for particles that are resident within the vascular compartment of interest (e.g., the LV cavity or myocardial microcirculation).
The initial description of contrast enhancement by microbubbles was made by Gramiak and Shah, when a cloud of echo signals was detected in the right heart, coming from the formation of microbubbles formed by fluid dynamic forces during rapid, high-pressure intravenous injection of a water-soluble fluorophore used at the time for measurement of cardiac output during heart catheterization. Over the ensuing years, several different forms of nonencapsulated microbubbles generated by hand agitation or low-frequency sonication were investigated, including for myocardial enhancement by MCE after intracoronary injection. These techniques were limited by the wide range of microbubble sizes produced, the inability of most of these microbubbles to pass to the left heart after intravenous injection, and the potential for large bubbles to become entrapped within the microcirculation when given as an intraarterial injection.
The safety, reproducibility, and widespread clinical feasibility of producing LV cavity and myocardial opacification with intravenous contrast administration microbubble contrast agents relied on the advent of small but stable and acoustically active microbubbles that are able to pass freely through pulmonary and systemic capillaries. Many of these microbubble agents also have a relatively narrow size distribution (relatively monodisperse). Those that do not, termed polydisperse agents , still contain relatively few microbubbles that are greater than the average functional capillary diameter of 5–7 μm when taking into account intraluminal projection of the glycocalyx. The creation of these stable size-controlled microbubbles that produce a strong acoustic signal relied on two major modifications: (1) a change in the gases used for the microbubble core material, and (2) microbubble encapsulation.
A partial list of some of the microbubble contrast agents that are currently commercially produced, marketed, and used in patients are shown in Fig. 3.3 . One of the common features of these agents is that the gas core is not composed of ambient atmospheric air components, which are for the most part nitrogen and oxygen. The reason for this compositional modification is based on mathematical models that have been used to predict the stability of a gas bubble. The rate of disappearance for a gas bubble in any given medium is dependent on the bubble size, the surface tension, and constants that describe the solubility and diffusion capacity of the gas in the bubble. Accordingly, the stability of microbubble contrast agents used in humans is improved when they contain gases with low diffusion coefficients and low solubility in water or blood, which is described by the ratio of the amount of gas dissolved in the surrounding liquid to that in the gas phase, or the Ostwald coefficient. These gases also must be inert, safe to use in humans, and cleared readily through respiration. These requirements are met in contemporary agents by using high-molecular-weight gases such as perfluorocarbons that remain in gas form at room and body temperature—for example, octafluoropropane (C 3 F 8 ), decafluorobutane (C 4 F 10 ), or sulfur hexafluoride (SF 6 ).
The encapsulation of the microbubbles represents a second common feature of contemporary contrast agents. On the most basic level, encapsulation with a “shell” composed of biocompatible materials such as protein (albumin) or lipid surfactants enhances in vivo stability by providing a barrier function that reduces outward diffusion of the gas core. Encapsulation also has an important role reducing surface tension of microbubbles, which allows tight control of microbubble size distribution, and both “on the shelf” and in vivo stability of small gas-filled particles. The use of air (mostly nitrogen) as the primary gas core component in microbubble contrast agents is still possible, provided encapsulation is performed with a relative thick and impermeable shell. However, first-generation contrast agents such as Albunex, which contained air and an albumin shell, still did not prove to be stable enough for reproducible LVO or for myocardial opacification, particularly in those with reduced cardiac output that result in long transit times from intravenous injection to systemic circulation, or in those receiving supplemental oxygen. The problem of air diffusion can be solved by the use of even more impermeable “air-tight” shells composed of thick layers of biopolymers. However, this compositional modification results in relatively inflexible microbubbles that can only be imaged well with very high acoustic powers.
The chemical nature and self-assembly of the components of the microbubble outer shell are heavily influenced by the gas composition and the process used to entrain gas into an encapsulated particle. For example, much of the albumin in the shell of microbubble agents is predicted to exist in a denatured form, owing to the temperature and pressure environment during manufacture. Because of the somewhat hydrophobic nature of perfluorocarbons, lipids in the microbubble shell arrange not as a bilayer configuration, as found in cell membranes, but rather as a monolayer configuration with inner orientation of the hydrocarbon residues. In research studies, nanoparticle ultrasound contrast agents with lipid bilayer configuration or multilamellar membranes based on a core that is mostly aqueous have also been produced. However, signal on conventional imaging has been low for these agents, due to the small amount of entrained gas.
Investigational Ultrasound Contrast Agents
There have been many investigational contrast agents that are not currently in routine clinical use that have been specially formulated for a unique diagnostic or therapeutic purpose ( Box 3.2 ). One general direction has been to create nanoparticle contrast agents that are approximately an order of magnitude smaller than most conventionally produced microbubbles. There are several reasons for wanting to use submicron acoustically active nanoagents. Based on experience with liposomes, smaller lipid-based agents could have longer intravascular circulation times before removal by the reticulo-endothelial phagocytic organs. Because of the inverse relationship between bubble size and its ideal resonant frequency (discussed later), smaller microbubbles may also be advantageous with high-frequency ultrasound applications, such as intracardiac ultrasound, intravascular ultrasound, or small animal imaging. With regard to using microbubbles as therapeutic delivery vehicles, very small contrast agents may also potentially extravasate, especially in inflamed or ischemic tissues where endothelial permeability is increased. The major limitation of nanobubbles is their instability. Accordingly, most of the experience with these agents has been with agents such as the multilamellar nanoparticles or perfluorocarbon emulsion (mostly liquid phase) nanodroplets, discussed previously, that do not produce strong acoustic signals with conventional contrast imaging protocols and frequencies.
Molecular imaging (e.g., atherosclerosis, ischemic memory)
Acceleration of clot lysis (sonothrombolysis)
Acoustically targeted gene or drug delivery
Augmentation of tissue ablation with HIFU
Augmentation of tissue perfusion through shear pathways
HIFU, High-intensity focused ultrasound.
Some of the limitations of nanoparticle ultrasound contrast agents could potentially be solved by the creation of phase-shifting nanodroplets. These encapsulated particles are an order of magnitude smaller in diameter than microbubbles, possess a condensed liquid-phase perfluorocarbon core, and require activation, whereby the diameter increases upon vaporization of the core to gas phase during the negative pressure phase of an acoustic field. These agents have recently been shown to produce myocardial microvascular opacification. Their use could potentially create novel ways of performing quantitative flow imaging or for reducing far-field attenuation caused by high gas-phase concentration within the LV cavity. Barriers to their use include the need to reproducibly convert from liquid to gas phase and to avoid large particles that lodge in the microcirculation.
There has been extensive investigation into the use of targeted microbubble contrast agents that can be employed for ultrasound-based molecular imaging. The most common approach has been to conjugate binding ligands to the surface of lipid-shelled perfluorocarbon microbubbles using a molecular spacer that projects the ligand away from the bubble surface and that theoretically produces a lever arm that reduces the force necessary to achieve bubble ligation. Tens of thousands of ligands can be conjugated to the surface of each microbubble. The most common approach to imaging these agents has been to perform ultrasound imaging 5–10 minutes after intravenous injection to detect retained microbubbles with minimal background signal from freely circulating microbubbles. Since microbubble-based ultrasound contrast agents are confined within the vascular space, the biologic processes that have been targeted have generally involved events that occur at the blood pool-endothelial interface. Clinical areas of greatest interest in cardiovascular medicine include molecular imaging of tissue ischemia, vascular inflammation or atherosclerosis, thrombus formation, and angiogenesis. Active research is also investigating how to improve microbubble adhesion through ligand, spacer arm, or microbubble shape modification, and novel ultrasound algorithms to detect the signal from retained microbubbles.
Novel microbubble agents have also been formulated for therapeutic purposes. Although a discussion of the full range of these agents is beyond the purview of this chapter, it is worth noting the ultrasound-facilitated delivery of genes has been augmented using cationic microbubbles to which cDNA can be charge-coupled, and by targeting these agents to the vascular endothelium. Similarly, ultrasound-facilitated drug delivery has been augmented by microbubbles that are specifically designed to carry either lipophilic drugs that can be loaded into an oil rim in the microbubbles or in the lipid shell, or hydrophilic drugs placed directly or indirectly on the microbubble surface.
Microvascular Behavior of Microbubbles
The rheology or vascular kinetics of conventional microbubble contrast agents in the microcirculation has been of great interest for safety consideration, understanding tissue-specific differences in the temporal changes in video intensity, and tracer kinetic modeling for perfusion imaging. There are several approaches that can be used to assess rheology. One approach to microbubble rheology has been to compare the transfer functions of microbubbles to that of technetium-labeled red blood cells (RBCs) through the myocardial microcirculation. To avoid confounding effects of recirculation and differential clearance from the blood pool, these studies were performed with intraarterial injections. These studies demonstrated similar first pass tracer kinetics for albumin microbubbles and RBCs through the myocardial circulation. A different approach is to use intravital microscopy to directly visualize microbubbles within an intact microvascular network. This technique has demonstrated that microbubbles transit the microcirculation of normal muscle beds unimpeded, do not coalesce or aggregate, and have a similar velocity profile as RBCs in arterioles, venules, and capillaries.
The safety issues related to rheology are based on the need to ensure that microbubbles do not lodge in the systemic microcirculation. After intravenous injection, the pulmonary circulation, which possesses capillaries of a similar diameter as the heart, acts as a filter for microbubbles that are larger than average capillary diameter (∼5 μm), thereby preventing systemic transit ( Fig. 3.4 ). There are potential exceptions to this process. One possibility is that the presence of any occult right-to-left shunting could result in systemic lodging. The extent of lodging is expected to be proportional to the number of larger microbubbles, and, accordingly, is of greater concern for agents with more polydisperse size ranges. Even for these agents, <1%–2% of microbubbles are of sufficient size to lodge, and microscopy has indicated that retention is often a transient event due to gradual deflation. Nonetheless, the presence of a “significant” right-to-left shunt remains a contraindication to microbubble contrast agents, despite the recognition that approximately 20% of nonselected subjects referred for echocardiography have the presence of a patent foramen ovale, which has not resulted in any major events on large safety studies. In small animal models where pulmonary arteriovenous shunting is proportionally high (3%–8% of transpulmonary flow), passage of microbubbles sufficiently large to be trapped by the systemic circulation can occur and can be used to assess regional blood flow after clearance of microbubbles from the blood pool.
Systemic lodging can also occur if microbubbles were to coalesce, aggregate, or enlarge after injection. The latter is possible from thermal increase in gas volume when bubbles are introduced to body temperature. This process has been noted for some experimental nonencapsulated microbubble contrast agents composed of mostly liquid emulsions with a low boiling point that can undergo uncontrolled increase in dimension after intravenous injection. For agents that are approved for use in humans, this problem has been resolved through the use of smaller molecular weight perfluorocarbon gases and bubble encapsulation. Another theoretical issue is rectified diffusion, where dissolved air in blood enters into a relatively nitrogen- or oxygen-free microbubble according concentration gradient, particularly during the negative pressure phase of ultrasound, which produces bubble expansion. This phenomenon has not been detected to any degree during in vivo imaging. Fortunately, aggregation and coalescence in vivo also do not occur, due to the relatively low concentration of microbubbles diluted in blood.
Although microbubbles that reach the systemic circulation have a similar rheology as RBCs in the normal coronary microcirculation, there are some exceptions that occur after microvascular injury. Microbubble clearance from the blood pool generally relies on the functions of the monocytic/phagocytic cells of reticuloendothelial organs. For many of the contrast agents, this process relies on opsonization, whereby serum complement mediates receptor-mediated uptake in cells, such as Kupffer cells of the liver. In areas of vascular inflammation or postischemic injury, lipid microbubbles can adhere directly to the endothelium or to adherent leukocytes, which results in a microvascular retention. Opsonization is amplified by the use of lipids in the microbubbles’ shell that possess a strong negative charge which tends to amplify opsonization, and also with the absence of shell components such as polyethylene glycol, which provide steric hindrance to bubble-cell interaction. With strong enough negative charge or with certain lipid components in the shell, microbubbles can even be retained within the normal microcirculation. This phenomenon has not resulted in any safety concerns, but rather has been leveraged as a technique for detecting recent but resolved myocardial ischemia, which results in an increase in microbubble retention.
Detection of Microbubble Ultrasound Signal
Microbubbles and other ultrasound contrast agents are effective because their size and deformability allow them to compress and expand in the sinusoidal pressure environment of an ultrasound field ( Fig. 3.5 ). The degree of signal enhancement is related to the magnitude and type of microbubble oscillation about its equilibrium radius. There have been many different models that have been used to describe the determinants of bubble volumetric resonance. The parent equation is the Rayleigh-Plesset equation, which describes the bubble dynamics according to pressures applied, surface tension, and boundary conditions. One of the earliest applications of this model was to understand how bubble collapse contributes to the physical degradation of ship propellers. Ensuing descriptions of bubble oscillation in an acoustic field based on the parent equation were used to describe energy losses that occur from thermal damping (heat loss into the medium), viscous damping (work to produce resonance in a viscous medium), and radiation damping, part of which is manifest as sound emission. These models were useful for defining how the magnitude of sound-producing oscillation is dependent on the compressibility and density of the gas core, the viscosity and density of the surrounding medium, the frequency and power of ultrasound applied, and microbubble radius. For encapsulated microbubbles, which are traditionally used as ultrasound contrast agents, viscoelastic damping from the shell is another important factor that influences bubble oscillation. In practical terms, for any given ultrasound power and frequency, less oscillation and acoustic signal is generated from microbubbles that have thick, stiff shells. This concept has been confirmed by comparing the acoustic signal from microbubbles with different shell thickness in vitro and in vivo. However, even microbubbles with extremely rigid shells can produce strong enhancement when imaging at sufficiently high power to release free gas bubbles through defects produced in the shell.
