Abstract
Since the end of the 20th century, echocardiography has been a first-line diagnostic method in patients in the pediatric cardiac intensive care unit PCICU). Echocardiography is used routinely and extensively in the PCICU for diagnosis, to guide clinical management, and to assist in procedures.
Key Words
Echocardiography, Noninvasive Imaging, Gradients, Doppler, ECMO, VAD
Since the end of the 20th century, noninvasive imaging techniques have assumed the primary role in the initial diagnosis of congenital and acquired heart disease in children. The majority of children with congenital heart disease safely proceed to surgical or transcatheter treatment on the basis of noninvasive testing alone ( Tables 34.1 and 34.2 ). Echocardiography remains the primary noninvasive tool because of its excellent resolution, ability to assess both anatomy and function, and lack of ionizing radiation. The examination can be performed at the patient’s bedside with relative ease and expediency. When transthoracic acoustic windows are compromised by surgical dressings or chest tubes in the perioperative period, transesophageal echocardiography is very useful. Other noninvasive techniques, including magnetic resonance imaging and computed tomography, are increasingly used and will be discussed elsewhere. This chapter reviews the basic principles of echocardiography and highlights applications and challenges of echocardiography in specific clinical settings.
| Lesion | Features to Be Identified With Noninvasive Imaging | Associated Lesions to Identify/Exclude | Indications for Cardiac Catheterization |
|---|---|---|---|
| Shunts | |||
| Atrial septal defect | Type of atrial defect, size, quantification of shunt (degree of right ventricular volume overload), pulmonary artery pressure, suitability for transcatheter closure. | Anomalous pulmonary venous return, pulmonary stenosis, pulmonary hypertension | Transcatheter closure commonly performed for secundum atrial septal defects. Catheterization usually unnecessary unless pulmonary hypertension or pulmonary vascular disease suspected and management depends on measure of pulmonary resistance. |
| Ventricular septal defect | Type, anatomic size, quantification of shunt (determined in part by left ventricular size), pulmonary artery pressure. | Multiple ventricular defects, coarctation | Catheterization usually unnecessary unless pulmonary vascular obstructive disease suggests inoperability. Some defects may allow transcatheter closure. |
| Patent ductus arteriosus | Anatomic size, pulmonary artery pressure, quantification of shunt (determined in part by left ventricular size), arch side and branching. | Coarctation of the aorta | Transcatheter closure. |
| Complete atrioventricular canal | Size of atrial and ventricular septal defects, atrioventricular valve morphology, balance, and function. | Outflow obstruction, patent ductus arteriosus | Only in rare patient for whom quantification of pulmonary resistance desired. |
| Cyanotic Lesions | |||
| Transposition of the great arteries | Coronary artery origins and branching, intracardiac shunts, pulmonary artery pressure. Magnetic resonance imaging useful postoperatively to assess branch pulmonary arteries, ventricular function. | Outflow obstruction, arch obstruction, septal defects | Balloon atrial septostomy when needed, usually performed in intensive care unit. |
| Tetralogy of Fallot | Mechanism and severity of pulmonary stenosis, main pulmonary artery and branch pulmonary artery sizes, coronary artery origin and branching. Magnetic resonance imaging very useful postoperatively to image branch pulmonary arteries, quantify severity of pulmonary regurgitation, assess right ventricular function. | Multiple ventricular septal defects, aortic arch sidedness and branching | Uncommonly needed in older childhood left unrepaired and echocardiography unable to establish pertinent features. In rare cases, pulmonary balloon valvuloplasty performed to palliate and delay surgery. |
| Tricuspid atresia with normally related great vessels | Size of ventricular septal defect and severity of pulmonary stenosis, adequacy of foramen ovale, identification of other sources of pulmonary blood flow, branch pulmonary artery size and continuity, arch side and branching. | Pulmonary stenosis, left juxtaposition of the atrial appendages | Very rare need to perform balloon atrial septostomy to enlarge foramen ovale. Minority of surgeons prefer angiogram of branch pulmonary arteries before shunt. Catheterization commonly performed before potential Glenn or Fontan operation. |
| Ebstein anomaly | Morphology of tricuspid valve, severity of tricuspid regurgitation, right ventricular size and function. | Pulmonary valve or branch pulmonary artery stenosis, patent ductus arteriosus, pulmonary artery pressure | Usually not necessary. |
| Valvular pulmonary stenosis | Mechanism and severity of stenosis, tricuspid valve and right ventricular size and function, patent ductus arteriosus. | Can occur with more complex heart disease including heterotaxy syndrome | Balloon dilation is now treatment of choice. |
| Total anomalous pulmonary venous connection | Identification of individual pulmonary veins and their size and site of return, exclusion of pulmonary venous obstruction. | Can occur with other complex congenital heart disease, left heart hypoplasia | Generally not necessary and of significant risk when required to establish pulmonary venous return (potential for pulmonary hypertensive crisis). |
| Admixture Lesions | |||
| Truncus arteriosus common | Morphology and function of truncal valve, branch pulmonary size and distortion. | Exclusion of multiple ventricular septal defects, ventricular hypoplasia (particularly right), aortic arch size | Generally unnecessary, rarely if pulmonary vascular obstructive disease is considered in an older infant or child. |
| Left Heart Obstructive Lesions | |||
| Valvular aortic stenosis | Size of aortic annulus and morphology, severity of aortic obstruction, left ventricular size and function. | Mitral stenosis, coarctation, left heart hypoplasia | In most cases, used as initial choice for relief of obstruction, although there is controversy concerning superiority over surgical approach. |
| Aortic coarctation | Aortic arch anatomy and size, mechanism of aortic coarctation, aortic arch branching, left ventricular size and function. Magnetic resonance imaging or computed tomography excellent when echocardiographic visualization inadequate. | Mitral and/or aortic stenosis | Generally unnecessary before surgical repair. Some favor balloon dilation versus surgical treatment. |
| Hypoplastic left heart syndrome | Size of atrial septal defect, tricuspid valve function, right ventricular function, nature of mitral and aortic obstruction, left ventricular size, arch morphology. Postoperatively, magnetic resonance imaging useful in examining arch and branch pulmonary arteries. | Patent ductus arteriosus, partial anomalous pulmonary venous return | Not generally needed before Norwood procedure, but performed before Glenn and before Fontan for hemodynamics and visualization of branch pulmonary arteries. |
| Lesion | Features to Be Identified With Noninvasive Imaging | Associated Lesions to Identify/Exclude | Indications for Cardiac Catheterization/CT |
|---|---|---|---|
| Pulmonary atresia with intact ventricular septum | Tricuspid valve size and function, right ventricular size and function, right ventricular pressure, presence of coronary sinusoids or fistula. | Coronary artery stenoses, right ventricular dependent coronary artery anatomy | Coronary artery anatomy, stenoses, presence of right ventricular dependent coronary artery distribution |
| Tetralogy of Fallot with pulmonary atresia | Nature of pulmonary atresia (absent main pulmonary artery versus short segment valvular pulmonary atresia), branch pulmonary artery size, coronary artery anatomy. Magnetic resonance imaging may obviate need for catheterization especially when branch pulmonary arteries good size and collaterals absent. | Multiple ventricular septal defects, aortopulmonary artery collaterals, coronary artery abnormalities, branch pulmonary artery distortion, hypoplasia, or discontinuity | Establishment of pulmonary artery supply and branch pulmonary continuity, determination of overlap between antegrade pulmonary flow and collaterals, in some cases to allow balloon dilation of stenoses, in some cases to allow embolization of aortopulmonary collaterals |
| Glenn or Fontan shunt | Ventricular function, valvular function. | Very dependent on anatomy requiring Glenn shunt | Pulmonary artery anatomy and hemodynamics, exclusion of decompressing vertical vein from innominate vein |
Physical Principles
Pulse Transmission and Reflection
Before discussing details of image acquisition, it is useful to understand how echocardiographic images are generated. Piezoelectric crystals are arranged at the tip of the imaging transducer. Piezoelectric crystals have an amazing property. When physically deformed, the crystals produce small amounts of electrical energy. Conversely, when electrical energy is applied to piezoelectric crystals, physical deformation occurs, and ultrasound waves are generated. Ultrasound is sound in frequencies ranging from 1 to 12 MHz (million cycles per second). Exciting the crystals with electrical energy, thereby producing ultrasound, is analogous to striking a gong with a mallet and producing audible sound.
Imaging systems do not send a single pulse of electrical energy to the crystals; rather, multiple, very short pulses are sent to the crystals. With each electrical pulse, a pulse of acoustic energy is produced. By pulsing the transducer repeatedly at very rapid rates, multiple acoustic pulses result.
If the transducer were simply placed on the patient’s chest, most of the ultrasound energy would reflect back off the skin. Like light reflecting off a mirror, the ultrasound energy would never enter the patient. This is due to marked differences in impedance between the transducer and the skin. This impedance mismatch is markedly decreased by applying acoustic gel to the transducer head, thereby allowing acoustic energy to enter the patient.
Tissue properties determine how much of the acoustic energy penetrates into the tissues and how much reflects back to the transducer. Fat, for example, reflects little and transmits most energy. In contrast, bone and air reflect nearly all and transmit very little energy. This explains why it is so difficult to image the heart when there is air (e.g., pneumothorax or hyperinflated lungs) between the transducer and the heart. Air-containing structures prevent acoustic pulses from reaching the heart, like a curtain blocking an audience’s view of actors on a stage. This also explains why the heart can be imaged from only a few select places on the body where the heart is near or in direct contact with the chest wall or diaphragm without intervening lung tissues. Often, repositioning the patient is required to achieve optimal echocardiographic images.
For an image of the heart to be displayed, the transmitted ultrasound pulse must reach the heart and reflect back to the transducer. As the received sound deforms the piezoelectric crystals in the transducer head, energy is converted back from acoustic to electrical energy. The imaging system can then display the returning signal as an illuminated pixel on a video monitor.
Signal Processing and Image Generation
Once the transducer converts the acoustic energy back into electrical energy, the imaging system performs complicated amplification and signal processing to optimize the signal-to-noise ratio. The returning signals are much weaker than the transmitted signal. Each reflector is assigned a velocity, a depth, and a density, which are detailed in the following paragraphs.
The system can accurately display a reflector’s location on the monitor because the speed of sound is nearly constant through different tissue types. The distance of a reflector from the transducer is directly related to the time for sound to travel to, and from, the reflector. The imaging system calculates the distance of each reflector from the transducer using the time-distance formula:
D = c × t / 2
Once signal amplification occurs and the distance is calculated for each reflector, other postprocessing must take place before an image is displayed on the monitor. In brief, this postprocessing involves the conversion of an electrical voltage into a gray scale value. The brightness of an illuminated pixel on the monitor correlates directly with the strength of received energy. More powerful reflectors (e.g., fibrous tissue) are represented as a brighter gray scale value (i.e., closer to white), whereas weaker reflectors (e.g., pericardial fluid) are represented as darker values (i.e., closer to black).
This entire process occurs at rates between 50 and 150 times per second. The system has a processing capacity, and the system operator has considerable control over how that processing power is used. Though this control is very helpful and allows image optimization, there is always the possibility that incorrect system adjustment will diminish image quality and limit diagnostic information.
Determinants of Image Resolution
Image quality is quantified by spatial and temporal resolution. Spatial resolution refers to two-point spatial discrimination: how closely two adjacent structures can be located and correctly resolved as two separate structures rather than incorrectly as a single reflector. Imaging systems should be able to achieve spatial resolution along the plane of the imaging beam to 0.8 mm. This means that two structures as close as 1 mm or slightly less from one another can be resolved from one another. Spatial resolution is superior along the axial plane (i.e., the plane parallel to the imaging beam) compared to the lateral plane (i.e., the plane orthogonal to the imaging probe), where spatial resolution falls to between 1 and 2 mm.
Temporal resolution refers to the ability of the system to resolve accurately the position of a moving structure in time. Though temporal resolution is unimportant in imaging a static structure like the brain (because the structure being imaged does not change position from one moment to the next), it is obviously critically important when the goal is to accurately demonstrate cardiac motion. If, for example, a rapidly moving structure (e.g., a valve leaflet) moves from point A to point B and back to point A in 10 ms and the temporal resolution of the imaging system is 10 ms, the motion of the valve will not be displayed accurately. In this example the structure will not appear to move but will appear to remain at point A. Temporal resolution is particularly important when heart rates are faster, as in the fetus or neonate.
Transducer Selection: Tradeoff Between Resolution and Penetration
The sonographer chooses the optimal imaging probe, weighing the requirements for spatial and temporal resolution against the requirements for tissue penetration ( Figs. 34.1 and 34.2 ).
Tissue penetration refers to the distance an acoustic pulse can reach. Penetration is indirectly related to probe frequency and directly related to beam transmission power. Higher-frequency transducers, covering frequencies between 8 and 12 MHz, provide superior spatial resolution but at the expense of lower penetration. Structures within 8 cm are typically imaged with high-frequency probes. Lower-frequency transducers covering frequencies between 1 and 3 MHz penetrate farther into tissue (up to 15 to 20 cm) but sacrifice spatial resolution. The sonographer tends to use the highest-frequency transducer that provides adequate tissue penetration for the distance required, just as the golfer chooses the putter for short shots requiring more accuracy and the driver when shots require more power. The sonographer trades off the needs of spatial resolution against the requirements of tissue penetration.
Penetration is also directly related to transmission power. Just as a 100-watt lightbulb illuminates farther than a 15-watt bulb, increased beam power allows the ultrasound beam to penetrate farther. However, acoustic power is constrained by safety limitations. Tissue warming can occur if excessively high power is used. In addition, image resolution can decrease with higher power, another tradeoff for which the sonographer must accommodate.
Transducer “footprint,” which is how the size of the transducer head is commonly described, is another practical consideration when imaging children. Higher-frequency transducers have a smaller footprint and therefore are ideal in imaging between rib spaces of neonates and small infants, compared with the lower-frequency transducers, which have larger footprints and may not physically fit in the intercostal spaces of a smaller pediatric patient. On occasion the sonographer will have to choose a higher-frequency transducer just to image in an intercostal space, compromising penetration for the footprint.
Transducers vary in other important ways, such as location of use. Transthoracic probes acquire images from the chest wall, whereas transesophageal probes are passed from the mouth and pharynx into the esophagus or stomach. Transesophageal probes must be smaller and more flexible. Transesophageal echocardiography should be considered when transthoracic images are poor or when transthoracic imaging is not possible due to logistics, such as a procedure being performed on the chest. Transesophageal imaging is contraindicated in the setting of esophageal varices, gastric bleeding, recent upper gastrointestinal surgery, or elevated intracranial pressure. Intravascular probes are mounted on catheters to allow intracardiac or intravascular imaging. Again, tradeoffs are present; miniaturization is achieved at the expense of some other imaging parameter, usually tissue penetration or resolution.
Doppler Analysis
Doppler analysis is a powerful complementary modality to two-dimensional imaging. It is used primarily to quantify flow velocity and pressure gradients within the vascular system. Similar to the principles discussed concerning image generation, small packets of ultrasound energy are transmitted, but in this case to the blood pool. When ultrasound energy strikes a moving target such as the red cells, the returning signal frequency is shifted compared with the transmitted frequency. Blood flow toward the transducer shifts the returning frequency higher, whereas flow away shifts the frequency lower. This velocity causing the frequency change (referred to as the Doppler frequency shift) is characterized by the Doppler equation:
V = f d × 2 c / f o × 1 / cos θ
Directionality of blood flow is determined by whether the frequency shift is higher (i.e., flow toward the probe) or lower (flow away from the probe). The speed of the red cells is related to the magnitude of the frequency shift. Fast-moving targets such as flow across a stenotic valve produce a larger frequency shift. Slower moving targets such as venous flow cause less of a frequency shift. The imaging systems routinely calculate the velocity from the measured frequency shift and then display that velocity for the sonographer.
The result of Doppler interrogation is a spectral display with velocity on the y-axis and time along the x-axis. The convention is to display flow away from the transducer as a positive value (i.e., above the zero baseline) and flow away from the transducer as a negative valve (i.e., below the zero baseline).
A very important relationship exists between the velocity and the pressure gradient producing this velocity. This is described by the Bernoulli equation:
Δ P = 4 ( V 2 2 − V 1 2 )
Δ P = 4 ( 4.5 2 − 1.5 2 )
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
