Indications

In 1999, the Committee on Coronary Angiography of the American College of Cardiology and American Heart Association Task Force established guidelines for coronary angiography (Box 4-1). The final decision must ultimately be based on an analysis of the expected benefits of the procedure balanced against the risks of the technique. Patients who undergo cardiac surgery are usually catheterized first, although with evolving noninvasive techniques, some aspects of the catheterization may be modified or omitted. The clinical variables that can be determined by cardiac catheterization include coronary angiograms, measurement of chamber pressures, detection of intracardiac shunts, characterization of myocardial performance, quantification of valvular stenoses and regurgitation, and imaging of cardiac anatomy. When such information is essential for a management decision, a catheterization is warranted.

For congenital heart diseases, coronary angiography is performed for three categorical indications: first, to assess the hemodynamic impact of congenital coronary lesions (congenital stenosis or atresia, coronary artery fistula); second, to assess the presence of coronary anomalies (anomalous left coronary artery arising from the pulmonary artery, anomalous left coronary artery arising from the right coronary artery or right sinus of Valsalva and passing between the aorta and the right ventricular outflow tract), or anomalies that may be innocent but whose presence may lead, during surgery, to coronary injury during the correction of other congenital heart lesions; and third, to perform interventional procedures. The severity of valvular disease and the pulmonary vascular resistance can be also assessed during catheterization of patients with congenital heart disease.

There are few relative contraindications to cardiac catheterization if the information to be obtained is critical to the patient’s care. If medical therapy can improve the patient’s hemodynamic status, some conditions warrant postponement of cardiac catheterization. Patients in congestive heart failure usually benefit from medical therapy to alleviate pulmonary edema before coronary angiography and left ventriculography are performed. Correction of abnormal bleeding and prothrombin times is essential to produce hemostasis; abnormal values frequently occur when anticoagulants or aspirin are not stopped long enough for the drug to be cleared. In a similar fashion, blood sugar and serum potassium levels should be corrected before catheterization. Other conditions such as digitalis intoxication, severe hypertension, poor renal function, and febrile illnesses, should be controlled before elective catheterization.

See Box 4-2 for relative contraindications to coronary angiography.

Catheterization Techniques

The technique of routine radiologic visualization by selective injection of the coronary arteries was first developed in 1959 by Mason Sones, a pediatric cardiologist who only the year before had successfully performed intracardiac angiography in children. In the Sones technique a single catheter is inserted through a brachial arteriotomy. The catheter can then be directed into each coronary artery.

Ricketts and Abrams extended the concept of selective coronary catheterization by devising two catheters, one for each coronary artery, that could be introduced percutaneously into the femoral artery by the Seldinger technique. The concept of a preshaped catheter for selective catheterization was refined by Amplatz and others; various shapes were devised to surmount the difficulties of different aortic sizes and ectopic locations of coronary arteries (Fig. 4-1). In 1967 radiologist Melvin Judkins designed preshaped catheters for both the right and left coronary arteries. The left coronary catheter, if properly aligned, needed only to be advanced around a normal aortic arch to fall into the ostium; whereas the right coronary catheter needed a 180-degree twist after it had passed around the arch (Fig. 4-2). Different catheter shapes and other refinements were introduced by Bourassa and associates, Schoonmaker, and others.

FIGURE 4-1 Coronary catheters. A and B, The Judkins and Amplatz series of catheters have a different shape for the right and left coronary arteries. C, The Sones and Schoonmaker catheters can be directed by the operator into either coronary artery.

FIGURE 4-2 Catheter positions. Left coronary (A) and right coronary (B) catheter positions for the Judkins, Amplatz, and Sones techniques. Because the second curve of the Judkins catheters touch the ascending aorta, they are affected by aortic dilatation and tortuosity. The Amplatz and Sones catheters rest on the aortic valve and are easier to control with a dilated root.

Cineangiography

Currently, every catheterization suite should be equipped with digital imaging in DICOM. Quantitative coronary angiography, allowing reduction of the wide variability in angiogram readings, has been validated by phantom studies with a high correlation (r = 0.95). Because cineangiography yields images with resolution up to 0.1 mm, this technique is the best way to visualize accurately coronary anatomy and stenoses. Only IVUS has an image resolution greater than that of coronary angiography. However, IVUS cannot visualize the entire coronary tree nor define the anatomic course of the vessels and has its own technical limitations (shadowing calcification and severe stenosis). Transthoracic and transesophageal echocardiography allow depiction of only the proximal 2 to 3 cm of the coronary arteries with sufficient resolution.

Despite years of development and improved acquisition techniques (gating with or without breath hold) magnetic resonance imaging (MRI) still cannot depict coronary arteries with results equivalent to angiography, even with blood pool contrast agent. Furthermore, MRI has limitations when cardiac patients, such as those with pacemakers, are imaged and this technique requires still, cooperative subjects. Recent developments with electrocardiography gated 64-detector row computed tomography (CT) units (Fig. 4-3) have provided higher resolution than MRI, but CT has its own limitations including the need for iodinated contrast, a high radiation dose, and a heart rate below 65 beats per minute.

FIGURE 4-3 Coronary computed tomography angiography. An isotropic resolution of 0.5 mm is obtained at a temporal resolution of 210 msec, allowing imaging of the coronary arteries during diastole. A, Axial section at the level of the left main and left anterior descending artery. B, Three-dimensional surface rendering reconstruction of the epicardial coronary arteries. Ao, aorta; D, diagonal artery; LA, left atrium; LAD, left anterior descending artery; LCX, left circumflex artery; LM, left main artery; RA, right atrium; RCA, right coronary artery; RV, right ventricle.

Because coronary angiograms are obtained with digital imaging, immediate reviewing is possible during the procedure, which allows the operator to be confident or prompts him or her to obtain additional views.

Digital subtraction angiography is useful in imaging the aortic arch, where the background frame has little motion.

Projection Positions

Most cardiac angiography is performed with projections that align the x-ray beam with the axis of the heart. The right anterior oblique (RAO) view profiles the long axis of the right and left ventricles parallel to the interventricular septum. The mitral and tricuspid valves are in tangent so that regurgitation is projected in the plane of the image intensifier. In the left anterior oblique (LAO) view the interventricular and interatrial septa are aligned perpendicular to the image intensifier plane. In this view, the tricuspid and mitral valves are seen in their frontal projections. Ventricular septal defects, aortic root to right heart fistulas, and systolic anterior motion of the mitral valve project in the plane of the image intensifier.

Compound angulation is frequently used to align the x-ray beam orthogonal to the heart. In addition to RAO and LAO projections, the image intensifier is tilted toward the head or the feet of the patient. In a cranial projection the image intensifier angles toward the head of the patient, whereas in a caudal projection the image intensifier angles toward the patient’s feet. A cranial projection can be identified by an unusually high diaphragm. In a caudal angulation the diaphragm is generally not visible. Cranial angulation also projects the heart over the abdomen. Compound angulation is part of the standard evaluation of the coronary arteries to project each branch and its bifurcation so that no overlap is present.

Because an angiogram is a projection image rather than a tomogram, the picture of the coronary arteries, cardiac chambers, and valves is a summation of many overlapping structures. Biplane orthogonal projections are required to evaluate cardiac structures completely.

The catheter techniques, injection rates, and other technical factors are summarized in Tables 4-1 and 4-2. Power injection techniques are used for all types of cardiac angiography except for coronary arteriography, in which injection is carried out by hand. When high flow exists in the coronary arteries, as in arteriovenous fistulas and in aortic regurgitation, a power injection can be used safely with the same techniques used to inject small arteries in other parts of the body.

TABLE 4-2 Technical factors for thoracic angiography in children

Volume is adjusted to the size of the vascular bed and flow through it.

Delivery time is roughly that of two heartbeats (e.g., at heart rate of 120 beats per sec, inject contrast over 1 sec).

Absolute maximum contrast dose is 5 milliliters per kilogram per day adjusted downward if renal function is decreased and if baby is less than 1 month old.

Radiation Exposure and Protection

Radiation exposure to both the patient and the operating personnel has been the focus of numerous studies. The amount of radiation received varies widely among laboratories and depends on the type of equipment employed, the operating techniques, the length of the procedure, the administrative procedures concerning the placement of personnel, and the shielding of the x-ray beam.

Depending on the radiation exposure time, the effective radiation doses can reach 2.1 to 5.6 mSv. If more angiographic views are obtained and if longer cine angiograms are performed, the dose may be higher. The dose-area product (DAP) is the most reliable measurement technique for dynamic examinations such as fluoroscopy in which the projection direction and technique parameters are continually varying. For dose evaluation, the contribution of cine fluorography to the total DAP is higher than that of fluoroscopy. For a coronary angiography, the DAP can be between 14 and 66.5 Gy/cm^-2, depending on the fluoroscopy and cine fluorography times. The effective radiation doses can be up to 20% higher in women.

The hazards of ionizing radiation are well known to radiologists and should be understood by all physicians who use x-ray equipment. Guidelines for maximum yearly occupational exposure have been established by the National Council on Radiation Protection (NCRP; Table 4-3). The NCRP is considering lowering the total body limit to 20 mSv annually. Lead eyeglasses and thyroid shielding help to diminish the dose received by the angiographer. The effective dose for a physician wearing a lead apron and thyroid shield is about 1.7 mSv/year, rising to 3.5 mSv/year without the thyroid shield. The average effective dose per application, for all types of procedures, is around 1 to 2 μSv for the physician when fully protected.

TABLE 4-3 Summary of recommendations for maximum x-ray dose

Note: Average annual background radiation from natural sources: 3 mSv (300 mrem).

ICRP, International Council on Radiation Protection; NCRP, National Council on Radiation Protection and Measurements.

Adapted with permission from National Council on Radiation Protection and Measurements, Limitation of exposure to ionizing radiation, Report No 116, Bethesda, MD, 1993, National Council on Radiation Protection and Measurements.

Operator radiation exposure during interventional procedures is much higher than for diagnostic angiography. Mikalason and colleagues surveyed interventional angiographers from 17 institutions and calculated a mean annual effective dose to the angiographer of 0.3 to 1 rem (3 to 10 mSv). Dash and colleagues found that during percutaneous angioplasty, operator radiation exposure is nearly doubled compared to routine coronary angiography.

Radiation exposure to the patient undergoing cardiac catheterization is higher than for any other type of radiologic examination but has been justified because the information gained is considered to be necessary for clinical management (Table 4-4). Repeated catheterization, particularly in critically ill children, may give a large radiation dose over a relatively short time span.

TABLE 4-4 Summary of mean radiation exposure during cardiac catheterization to patient and physician

Reprinted with permission of American Journal of Cardiology from Miller SW, Castronovo FP Jr: Radiation exposure and protection in cardiac catheterization laboratories, Am J Cardiol 55:171–176, 1985. Copyright © 1985 by Excerpta Medica, Inc.

Radiation reduction involves the three classic parameters of time, shielding, and distance and can be individually adapted by each laboratory to make the x-ray exposure to both the patient and the physicians as low as possible. The time an operator is exposed can be reduced in several ways. About half of the cardiac catheterization exposure occurs during fluoroscopy and the remainder during cineangiography. The fluoroscopy time can be considerably reduced by using short bursts of the fluoroscope rather than a prolonged, continuous exposure. In addition, by prolonging the time between catheterizations, the operator can lessen the amount of radiation received per unit of time. For example, Dash and colleagues have calculated that it is necessary to limit the number of coronary angioplasty procedures for a cardiologist to five cases per week to meet the occupational exposure guidelines of 100 mrad/week. Nurses and technicians can be rotated to other duties so that they are not continuously present in the angiography room.

Additional local lead shielding should be considered to help limit scattered radiation. The smallest x-ray beam possible will help to reduce the exposure of both the patient and the operator. Movable shields or drapes are available for most of the current angiographic units. Side drapes between the patient and the operator reduce scatter passing through the patient that would ordinarily be received by the operator. Cranial and caudal angulations considerably increase the x-ray tube output, the radiation received by the patient, and the secondary scatter received by the angiographer.

Because radiation decreases as the square of the distance, all personnel who are not needed in the room should be located elsewhere. For instance, the electrophysiologic data collection can be performed from a remote location rather than from beside the fluoroscopy table. Those nurses and technicians remaining in the room should stay as far as practical from the x-ray tube. The radiation physicist can monitor the radiation burden, evaluate the fluoroscopy techniques for each angiographer, and periodically measure the radiation output at various places in the room.

Table 4-5 compares the radiation dose of coronary angiography with other radiology examinations.

TABLE 4-5 Typical effective dose values for radiology examinations

CT, computed tomography.

Modified and reprinted with permission from Bauhs JA, Vrieze TJ, Primak AN, et al.: CT dosimetry: comparison of measurement techniques and devices, Radiographics 28:245-253, 2008.

Right Coronary Artery

The right coronary artery originates from the right sinus of Valsalva and continues in the right atrioventricular groove to the crux of the heart. Its ostium is usually in the upper two thirds of the sinus but may be ectopically located from slightly below the aortic valve leaflets in the left ventricle to a few centimeters above in the ascending aorta. In bicuspid aortic valves in which the two sinuses of Valsalva are placed anteriorly and posteriorly, the right coronary artery may go anterior from the sinus as compared to the more common angulation of 30 degrees to the right of the sternum in patients with tricuspid aortic valves. The right coronary artery in the LAO view is C-shaped and is conveniently divided into proximal, middle, and distal segments (Fig. 4-5). The proximal right coronary segment lies beneath the right atrial appendage.

FIGURE 4-5 The right coronary artery. A, Cranial left anterior oblique projection. B, Lateral projection, same patient. CB, conus branch; PD, posterior descending artery; PLV, posterior left ventricular artery; RVM, right ventricular marginal artery; SAN, sinoatrial nodal artery.

Sinoatrial Nodal Artery

The sinoatrial nodal artery branches from the proximal right coronary artery in roughly 50% of patients (see Figure 4-5). When it does not come from the right coronary artery, a longer atrial artery originates from the left circumflex artery and terminates in the same location by sending an anterior and posterior branch around the superior vena cava at the right atrial junction.

Conus Artery

The conus artery is the first ventricular branch of the proximal right coronary artery. This artery has a separate origin from the sinus of Valsalva in about half of all angiograms (Fig. 4-6). The conus artery goes anteriorly around the right ventricular conus or infundibulum and frequently ends in three short branches that resemble a pitchfork. This artery is a common collateral channel to the left anterior descending artery and is then called the circle of Vieussens. The conus artery may also go inferiorly over the right ventricular free wall to form a distal marginal branch. Other right ventricular branches from the middle and distal segments of the right coronary artery are called marginal branches. The acute marginal artery is a large right ventricular marginal artery near the bend in the artery between the middle and distal segments (the acute margin of the heart). Any right ventricular marginal branch may continue along the diaphragmatic wall of the right ventricle to become a short posterior descending artery (Fig. 4-7).

FIGURE 4-6 Separate origin of the conus artery (arrow) from the right sinus of Valsalva.

FIGURE 4-7 The right coronary artery. In the right oblique projection the inferior septum is supplied by both the posterior descending artery (solid arrow) and the acute marginal branch (open arrow).

In about 3% of patients a small artery originates near the conus artery that supplies the superior portion of the interventricular septum. The right superior septal perforator artery goes deeper into the myocardium than is usually the case with the conus artery and ends in several straight branches parallel to the interventricular septum (Fig. 4-8).

FIGURE 4-8 Right superior septal perforator artery (arrows). A, Left anterior oblique projection. B, Right anterior oblique projection.

Distal Right Coronary Artery

The distal right coronary artery has so many variations that the naming of small branches is at times difficult. The distal right coronary artery typically ends by dividing into a posterior descending artery and a posterolateral left ventricular artery (Fig. 4-9). In about 90% of people the right coronary artery ends in the posterior descending artery and is therefore called the dominant artery. When the left circumflex artery ends in the posterior descending artery, the right coronary artery is nondominant, and the left circumflex artery is then the dominant blood supply inferiorly (Fig. 4-10).

FIGURE 4-9 Distal right coronary artery. The cranial left oblique projection spreads the posterior descending artery (PD) away from the posterior left ventricular artery, which is a continuation of the posterolateral left ventricular artery (PLV). RCA, right coronary artery.

FIGURE 4-10 Nondominant right coronary artery. Several vestigial right ventricular marginal branches supply the anterior wall of the right ventricle in this right anterior oblique projection.

There are many variations in the vascular pattern of the inferior part of the ventricular septum. A codominant pattern is when either a posterior descending artery originates from both the distal right coronary and the left circumflex arteries or the posterior descending artery comes from the right coronary artery and a long left circumflex artery supplies the posterior left ventricular wall. There may be between one and five posterior descending arteries, some of which begin as either right ventricular marginal arteries or left circumflex marginal arteries. The posterolateral left ventricular artery lies in the distal atrioventricular groove and supplies several posterior left ventricular arteries over the posterolateral wall of the left ventricle. This segment frequently has a middle part that is shaped like an inverted U at the crux of the heart (Fig. 4-11). The atrioventricular nodal artery originates near this U-bend and goes superiorly for about 1 cm to the region of the atrioventricular node (Fig. 4-12).

FIGURE 4-11 Two variations of the U-bend (arrow) in the posterior left ventricular branch of the right coronary artery.

FIGURE 4-12 Atrioventricular nodal artery. The U-bend of the posterior left ventricular artery is the origin of the atrioventricular nodal artery (arrow) as seen on this left anterior oblique view.

Left Coronary Artery

The left sinus of Valsalva is the origin of the left coronary artery. The left main artery does not taper and typically is about 1 cm long (Fig. 4-13). However, there may be no left main artery, with the left anterior descending artery and left circumflex arteries originating separately from the left sinus. The left main coronary artery may trifurcate into a left anterior descending artery, an intermediate artery or ramus medianus (Fig. 4-14), and the left circumflex artery. The caudal LAO (also called the “spider” view because of the arachnoid shape of the proximal left coronary artery) puts the left main and proximal circumflex arteries in the plane of the film but foreshortens the left anterior descending artery (Fig. 4-15).

FIGURE 4-13 Left main coronary artery. A, Caudal right anterior oblique projection. B, Lateral projection. D, diagonal branch; LAD, left anterior descending artery; LCX, left circumflex artery; LM, left main artery; M, circumflex marginal branch.

FIGURE 4-14 Intermediate artery. Intermediate artery (arrow) between the left anterior descending (LAD) and the left circumflex (LCX) arteries. LM, left main artery; M, circumflex marginal branch.

FIGURE 4-15 “Spider” projection. A, The caudal left anterior oblique projection separates the left anterior descending artery (LAD) from the left circumflex (LCX) and intermediate (arrow) arteries. B, Two ramus branches are originating from the left main coronary artery (LM).

Left Circumflex Artery

The left circumflex artery lies in the left atrioventricular groove and may exist only as a vestigial twig or may be so long that it ends by becoming the left posterior descending artery (Fig. 4-16). Its major branches are called left circumflex marginal arteries and are numbered first, second, and so on. Because the inferior and left side of the heart is the obtuse border, marginal branches in this location may be called obtuse marginal arteries (Fig. 4-17). Late filming of a left coronary artery injection shows the coronary veins. The great cardiac vein, which becomes the coronary sinus, is in the left atrioventricular groove and serves as a landmark for the left circumflex artery.

FIGURE 4-16 Dominant left coronary artery. The posterior descending artery (arrows) originates from the left circumflex artery in the right (A) and left (B) oblique projections. The right coronary artery (C) does not extend to the inferior septum.

FIGURE 4-17 The left circumflex artery gives rise to obtuse marginal branches. A, Lateral projection. B, Spider view. Orthogonal view of part A. LAD, left anterior descending artery; LCX, left circumflex artery; M, marginal branch.

Left Anterior Descending Artery

The left anterior descending artery lies in the interventricular groove and supplies two distinctive types of branches, septal and epicardial (Fig. 4-18). Septal branches go to the interventricular septum, usually along the right ventricular side of the septum, and originate from the left anterior descending artery in a nearly perpendicular direction. The septal branches may themselves have branches, and commonly the first septal branch may have a broomlike appearance. Epicardial branches over the anterolateral wall are called diagonal arteries and number from one to many.

FIGURE 4-18 Left anterior descending artery. Cranial left anterior oblique view projects the left circumflex artery (LCX) lateral to the left anterior descending artery (LAD). This compound view separates the diagonal arteries (D) from the septal branches. LM, left main coronary artery; M, marginal branch.

Several characteristics of the left anterior descending artery are unique and help to identify this artery on a coronary angiogram. The anterior descending artery is usually the longest branch of the left coronary artery and ends at the cardiac apex or occasionally continues to supply most of the inferior septum. The termination characteristically looks like an inverted Y (Fig. 4-19). Unlike other branches of the left coronary artery, the left anterior descending artery has numerous septal branches throughout its length (Fig. 4-20). A rare septal branch may come from the left main, a diagonal, or a circumflex marginal artery proximally. There is little motion of the left anterior descending artery in contrast to the 1-cm excursion of the left circumflex artery adjacent to the left atrium.

FIGURE 4-19 Typical Y-shaped termination of the left anterior descending artery (arrow). D, diagonal branch; LCX, left circumflex artery; LM, left main coronary artery; M, marginal branch; S, septal branch.

FIGURE 4-20 Left coronary artery variations. A, Left anterior oblique view shows trifurcation of the left main artery into the left anterior descending (LAD), intermediate, and circumflex arteries. There are two septal branches arising from the intermediate artery (arrow). B, The left anterior descending artery (arrow) in the right anterior oblique view goes around the apex to supply the entire inferior septum. C, The left circumflex artery gives off two marginal branches, then continues as a vestigial vessel (arrow) in the atrioventricular groove.

Cardiac Veins

Identification of the cardiac veins is useful in angiography because they mark the atrioventricular and interventricular boundaries of the chambers. Occasionally they demonstrate anomalies such as persistent left superior vena cava terminating in the coronary sinus, absence of the coronary sinus, or anomalous pulmonary venous connection to the coronary sinus. Veins are distinguished from coronary arteries because the veins opacify several seconds after arterial injection, have less opacification than the corresponding adjacent arteries, are generally larger than the adjacent arteries, and drain into the coronary sinus or a cardiac chamber.

The coronary sinus begins at its opening into the right atrium with its thebesian valve and extends along the left atrioventricular sulcus to the bifurcation with the oblique vein of Marshall. In a normal heart, this latter vein is obliterated but when it remains patent it continues as a left superior vena cava. The continuation of the coronary sinus beyond this vein of Marshall is the great cardiac vein. This vein extends beneath the left atrial appendage and becomes the anterior interventricular vein beside the left anterior descending artery.

The left ventricle has veins that lie roughly beside the major arteries (Figure 4-21). The anterior interventricular vein lies adjacent to the left anterior descending artery; it extends superiorly from the apex to pass beneath the left atrial appendage and joins the great cardiac vein beside the left circumflex artery.

FIGURE 4-21 Cardiac veins connecting to the coronary sinus.

The posterior interventricular vein, also called the middle cardiac vein, runs beside the posterior descending artery from the apex to the crux and either drains into the coronary sinus or separately enters the right atrium.

The small cardiac vein lies in the right atrioventricular sulcus beside the distal right coronary artery and drains into the posterior interventricular vein, the coronary sinus, or directly into the right atrium. Right atrial veins are rarely seen during injection into the right coronary artery but they may be visible if the catheter is obturated and the injection is made under pressure.

Right Atrium

The right atrium has an appendage that is larger than the left atrial appendage and has a broad-based connection to the main chamber (Fig. 4-22). The inflow structures of the right atrium are the inferior vena cava, the superior vena cava, and the coronary sinus. The internal structures are difficult to identify angiographically but MRI usually shows the fossa ovalis, the crista terminalis, and the pectinate muscles.

FIGURE 4-22 The right atrium. The wide atrial appendage (A) projects over the right pulmonary artery. The tricuspid valve (arrow) is closed.

Right Ventricle

The right ventricle has a complex shape consisting of a triangular body and a cylindrical outflow tract. Its three parts are the inflow segment, the body, and the outflow segment (Fig. 4-23). The inflow region is the tricuspid valve and its apparatus, including the papillary muscles. The anterior, posterior, and septal leaflets of the tricuspid valve are easily identified on an echocardiogram and can frequently be separately distinguished during angiography and MRI. Compared to the left ventricle, the right ventricle has larger trabeculations that extend on to the septum. The moderator band is usually the largest trabeculation near the septum. The walls of the right ventricle are named according to their location: anteroseptal, apical, and diaphragmatic in the RAO view and septal and anterior in the LAO view. The outflow tract of the right ventricle is the infundibulum or conus and is cylindrical (Fig. 4-24). The pulmonary valve is separated from the tricuspid valve by the infundibulum—a landmark difference from the left ventricle in which the aortic and mitral annuli join posteriorly. This conal segment has normal contractions and may narrow considerably in systole in right ventricular hypertrophy.

FIGURE 4-23 The right ventricle. A, Right anterior oblique cranial projection. B, Lateral projection. Systolic images of valvular pulmonic stenosis show a domed pulmonic valve and poststenotic dilatation of the main and left pulmonary arteries. P, papillary muscles; T, tricuspid valve.

FIGURE 4-24 The right ventricular infundibulum. The parietal band (arrow) is the muscle medial (A) and posterior (B) to the right ventricular outflow tract. Faint opacification of the left ventricle in the lateral projection occurred through a ventricular septal defect (arrow). A, aorta.

Left Atrium

The left atrium is behind and to the left of the right atrium. The pulmonary veins either connect with this chamber as four individual vessels, or occasionally, the superior and inferior pulmonary veins on each side have joined before connecting to the heart. The body of the left atrium has no trabeculations but the interior of the left atrial appendage may be finely striated. In contrast to the right atrium, the left atrial appendage has a narrow neck and is normally smaller (Fig. 4-25).

FIGURE 4-25 The left atrium. The catheter enters the left atrium across a patent foramen ovale. The appendage (A) is narrow and unusually long.

Left Ventricle

The left ventricle has an oval shape and finer trabeculations than the right ventricle (Fig. 4-26). The mitral valve (considered part of the left ventricle) consists of anterior and posterior leaflets. The anterior or septal leaflet covers about one third of the mitral circumference and has a smoothly rounded border. The posterior leaflet typically has several scallops and covers two thirds of the mitral circumference. The chordae can occasionally be seen as thin lucencies on a left ventriculogram. The papillary muscles may have single or multiple heads and connect about half of the distance between the base and the apex of the left ventricle. Because the papillary muscles look like filling defects on the left ventriculogram, they are occasionally confused with thrombus. However, the papillary muscles contract during systole as the adjacent wall also contracts, a criterion that distinguishes thrombus or tumor from papillary muscles. A major landmark of the left ventricle is the continuity of the mitral and aortic valves. Unlike the right ventricle, there is no muscle between the aortic and mitral valves.

FIGURE 4-26 The left ventricle in diastole and systole. A, The 30-degree right anterior oblique projection in diastole demonstrates the closed aortic valve (arrows) and the inflow of unopacified blood through the open mitral valve. B, In systole the mitral valve is closed (arrow). The anterolateral (AL) and posteromedial (PM) papillary muscles are larger and nearly touch.

The walls of the left ventricle are arbitrarily divided into five segments in the RAO projection and into four segments in the LAO view (Fig. 4-27). The septum and the anterior leaflet of the mitral valve are better seen if cranial angulation is added in the LAO projection (Fig. 4-28).

FIGURE 4-27 Left ventricular wall segments. A, Right anterior oblique view. B, Left anterior oblique view.

FIGURE 4-28 The left ventricle in the cranial left anterior oblique projection. The interventricular septum (S) is abnormally bowed toward the left ventricle because an atrial septal defect has caused right ventricular enlargement. This diastolic frame shows the open anterior (A) and posterior (P) leaflets of the mitral valve.

Tricuspid Valve

The tricuspid valve (occasionally the normal tricuspid valve has only two cusps) is considered part of the right ventricle and consists of three leaflets: septal, posterior, and anterior. The septal leaflet is the smallest and is attached by small chordae to the septum. Because this leaflet is being derived from endocardial cushion tissue, it is malformed when endocardial cushion tissue is deficient. The anterior and posterior leaflets are attached to the anterior and posterior walls respectively. On frontal projections these leaflets form the medial boundary of the right ventricle during systole but they are usually invisible in diastole except for occasional segments, which appear as thin, lucent streaks. The chordal attachments of the tricuspid leaflets insert into several papillary muscles on both the septal surface and the free wall.

Mitral Valve

The mitral apparatus consists of five related structures (Fig. 4-29):

FIGURE 4-29 Mitral valve. The mitral valve in diastole (A) and systole (B) in the left anterior oblique projection. The anterior leaflet (open arrows) is adjacent to the septum and aortic valve. The posterior leaflet (solid arrows) lies beside the superolateral left ventricle.

Disorders in any of these structures may result in dysfunction of the valve. Furthermore, abnormalities in the left ventricular wall adjacent to the papillary muscles, such as infarct or aneurysm, may also cause mitral regurgitation.

The mitral annulus serves as a fulcrum for the leaflets. The muscle adjacent to the fibrous portion of the mitral annulus causes this region to contract during systole. In disease states that lead to left ventricular enlargement, the mitral annulus can dilate proportionately. Radiographically, this structure when calcified serves as a marker for the adjacent left circumflex artery and the coronary sinus. The plane of the attachment of the mitral leaflets to the annulus is the posterior extent of the leaflets during systole; leaflet motion beyond this plane is abnormal and is encountered in the prolapsing mitral leaflet syndrome.

The posterior left atrial wall continues as the posterior leaflet of the mitral valve. In severe dilatation of the left atrium the posterior wall of the atrium may pull the posterior mitral leaflet superiorly, contributing in a small way to mitral regurgitation.

The anterior mitral leaflet subtends about one third of the circumference of the mitral ring and is adjacent mainly to the posterior and left aortic cusps. The anterior leaflet has a longer base-to-margin length than the posterior leaflet but the surface area of both leaflets is equal. The anterior leaflet is identified on the left ventriculogram as the posterior straight line of contrast material extending inferiorly from the aortic valve. The posterior mitral leaflet has three and occasionally up to five scalloped segments. When three segments are present, they are the middle, posteromedial, and anterolateral segments and correspond to the adjacent left ventricular papillary muscles. In the RAO view, the mitral leaflets are well defined in systole when the valve is closed but are normally not seen in diastole. The reverse happens in the LAO projection in which the mitral leaflets are seen mainly in diastole.

The chordae tendineae extend from the heads of the anterolateral and posteromedial papillary muscles, branch into several divisions, and then insert both on to the leaflets and into the subjacent left ventricular endocardium. The chordae from one papillary muscle extend to both mitral leaflets so that a rupture of the papillary muscle leads to severe instability in both valve leaflets during systole. Chordae originating from a papillary muscle typically split into three smaller chords before inserting into the leaflet (however, the division is inconstant). The clinical significance of this split is that rupture of a single chorda usually has no effect on mitral valve competency, and rupture of several chordae may produce only mild mitral regurgitation. Chordae are usually not seen on a normal left ventriculogram but may become visible if thickened because of a rheumatic process or if stretched and elongated, as in the prolapsing mitral leaflet syndrome (Fig. 4-30).

FIGURE 4-30 Mitral valve. Multiple small scallops (arrows) of the posterior mitral leaflet, which are slightly prolapsed into the left atrium, are shown.

Normally the papillary muscles contract before the adjacent left ventricular wall does, which brings the leaflets back into position for the ensuing rise in the left ventricular systolic pressure. When this sequence is altered, for example, during a premature ventricular contraction, the mitral valve becomes slightly incompetent. In systole, as the left ventricular apex moves toward the base of the heart, the papillary muscles tense and counterbalance the shortened distance between the chordal insertions and the mitral leaflets, thus preventing regurgitation. In the papillary muscle dysfunction syndrome, the area around a papillary muscle is infarcted or ischemic, which prevents contraction of this muscle. Mitral regurgitation then ensues and the leaflets can occasionally be seen prolapsing into the left atrium.

Pulmonary Valve

The sinuses and cusps of the pulmonary valve are designated as anterior, left, and right. The cusps of the pulmonary valve are visible on an angiogram during diastole; however, during the ejection phase the leaflets are thin and are positioned against the main pulmonary artery in such a way that they are usually not visible (Fig. 4-31). The attachments of the pulmonary cusps serve as an important landmark for identifying the upper portion of the right ventricular conus. On a lateral projection (Fig. 4-32) the angle of the blood flow through the pulmonary valve follows a vertical line roughly parallel to the sternum. In pathologic conditions, such as tetralogy of Fallot, the angle of blood flow becomes more horizontal. The practical implication is that in a normally formed heart the pulmonary valve is roughly in tangent in the posteroanterior view but in the patient with tetralogy of Fallot a steep cranial angulation is necessary to profile the valve. During right heart catheterization, a pulmonary capillary wedge pressure can be also obtained, distally in the pulmonary artery, with an occlusive balloon catheter (Swan-Ganz). A true wedge pressure can be measured only in the absence of flow. In this condition, the pressure equilibrates across the capillary bed. The pressure recorded with this technique will provide a measurement of the pulmonary venous pressure, which may be affected by mitral valve diseases or left ventricular dysfunction. In normal conditions, the pulmonary venous and left atrial pressures are equal.

FIGURE 4-31 Pulmonary valve. Cranial angulation puts the pulmonary annulus in tangent and elongates the outflow region from the infundibulum to the left pulmonary artery. The pulmonary valve is stenotic.

FIGURE 4-32 The pulmonary and tricuspid valves in the lateral projection. A, In systole the pulmonary valve is domed with a jet through it, indicating stenosis. B, The infundibulum (In) is larger in diastole so that a dynamic subvalvular infundibular stenosis is present. Note the anterior leaflet of the tricuspid valve (arrow).

Aortic Valve

The aortic sinuses of Valsalva and their cusps are named the right, left, and posterior or noncoronary (Fig. 4-33). The sinuses of Valsalva extend to the sinotubular ridge, which marks their junction with the aorta. The coronary arteries usually originate in their respective sinuses about two thirds of the distance superiorly between the aortic valve and the sinotubular ridge. The posterior sinus is next to the atrial septum so that a rupture of this sinus could extend into either atrium. The right sinus of Valsalva is adjacent to the right atrium and the right ventricle (Fig. 4-34), and the left sinus of Valsalva is next to the left atrium and the left ventricle. On the lateral projection of an ascending thoracic aortogram, the plane of the aortic valve is tilted so that the caudal extension of a line perpendicular to this plane goes posteriorly. When the aorta is “untucked,” which occurs in truncus arteriosus and in transposition of the great vessels, a line perpendicular to the plane of the aorta is more anterior (Fig. 4-35).

FIGURE 4-33 Aortic valve. Aortic valve in diastole (A) and systole (B) in the lateral projection. The arrow indicates the noncoronary sinus. The left coronary artery originates superiorly from the left sinus, and the right coronary artery originates anteriorly from the right sinus.

FIGURE 4-34 Rupture of the right sinus of Valsalva into the right ventricle. An aortic root abscess (a), which caused aortic regurgitation into the left ventricle (LV), later eroded into the right ventricle (RV).

FIGURE 4-35 The tilt of the aortic root. A, Normally a line (large arrow) through the plane of the sinotubular ridge (small arrows) points posteriorly. B, In truncus arteriosus, in which both the aorta and the pulmonary arteries originate from the truncal valve, a line (arrow) through the “untucked” valve points anteriorly.

Like the pulmonary valve leaflets, the aortic leaflets are difficult to see during systole, but during diastole they mark the boundary of the left ventricle and the sinuses of Valsalva. The leaflets may occasionally flutter in a person with a high cardiac output or in an elderly patient, possibly related to turbulence immediately beyond the cuspal attachments. During pull-back from the left ventricle to the ascending aorta, subaortic stenosis (fibromuscular dysplasia of the left ventricular outflow tract) can be diagnosed.