Patients with functionally single ventricular circulations are challenges to evaluate by any method. The anatomy, physiology, and surgical procedures associated with these malformations are complex and varied. Conventional surgical palliation for the functionally single ventricle is usually focused toward completion of a Fontan “operation.” However, the Fontan circulation can be completed using several different surgical techniques. As a result, some of the confusion encountered when evaluating the post-Fontan patient stems from the number of different connections used to create the systemic venous to pulmonary arterial pathway. Therefore, a systematic approach is required if one is to truly understand the circulation of patients after Fontan operations. This chapter will discuss the anatomy of the Fontan circulation, its unique physiology, and an approach to the echocardiographic evaluation of these patients that not only allows a complete delineation of the Fontan circulation but also increases the likelihood of detecting late complications. In conclusion, a number of case studies illustrating the physiology and complications of this type of circulation will be reviewed.
ANATOMY AND PHYSIOLOGY OF THE FONTAN OPERATION
The first step toward understanding the patient with a Fontan circulation is to realize that this procedure is a surgical palliation, rather than a “repair.” A Fontan operation reduces ventricular workload and eliminates or reduces oxygen desaturation in the patient with a functionally univentricular heart. These goals can be achieved by many different surgical techniques, and the patients who can benefit from the Fontan operation have many different cardiac malformations. Thus, the Fontan is more of a surgical “concept” than it is a specific operative technique. This concept can be summarized as follows: A Fontan “operation” consists of any combination of surgical procedures that diverts the systemic venous return away from the ventricle, eliminates mixing of systemic and pulmonary venous blood, and creates a circulation “in series” for patients with functional single ventricle physiology. After creation of a Fontan, pulmonary venous flow remains committed to the patient’s only functional ventricle. Many types of pathways have been used to create the systemic venous diversion that defines the Fontan circulation. Regardless of how the connections were constructed, superior and inferior vena cava flows bypass the ventricle, and pass through the lungs without the assistance of a ventricular systolic pump.
When the Fontan procedure was first applied, the right atrium was used as the pathway for systemic venous flow, an anastomosis was created between the right atrial appendage, and the pulmonary artery provided the outlet from the atrium to the pulmonary arteries. This type of connection is referred to as an “atriopulmonary” Fontan and is illustrated in Figures 41.1 and 41.2. This approach worked well for patients with tricuspid atresia and others with left atrioventricular valves. However, this approach was difficult to apply to hearts with left atrioventricular valve atresia, hypoplasia, or severe left atrioventricular valve dysfunction. Although atriopulmonary connections have been superseded by newer techniques, there are many patients who still have these connections. Therefore, we need to not only understand how to recognize them, but also be aware of the unique complications associated with this connection. These complications include atrial enlargement and arrhythmias, compression of the pulmonary veins by the enlarged right atrial chamber, and atrial thrombus formation within the dilated atrium. Atrial thrombi, sluggish flow in a dilated atrial chamber (with spontaneous echo contrast), blind pouches in communication with the “left” heart (such as a ligated native pulmonary root), and residual right-to-left shunts have all been associated with an increased risk for embolic events.
Over time, less of the right atrium was included in the systemic venous pathway, allowing the right atrioventricular valve to contribute to the systemic circulation. Lateral atrial tunnels or intraatrial conduits allowed the pulmonary venous return to pass through any atrioventricular valve to reach the ventricle, expanding the spectrum of the Fontan to those with common atrioventricular valves and left atrioventricular valve abnormalities. Recently, the surgical systemic venous diversions involved in Fontan procedures have bypassed the atrium completely. This has been accomplished by combining direct superior vena cava–to–pulmonary arterial anastomoses (bidirectional Glenn connections) with intracardiac or extracardiac conduits. These nonvalved conduits connect the suprahepatic inferior vena cava to the pulmonary arteries (Fig. 41.3). A Fontan circulation that includes an extracardiac conduit has been referred to as an “extracardiac Fontan” (Figs. 41.4 and 41.5.
Figure 41.1. Diagram of an atriopulmonary Fontan connection. The underlying anatomy is that of a functionally single ventricle chamber, with right atrioventricular valve atresia and pulmonary stenosis. In this case, the superior vena cava (SVC) has been directly and bidirectionally connected to the right pulmonary artery (a “bidirectional Glenn” anastomosis). The native right atrium (RA) has been converted into a conduit for inferior vena caval flow by closure of the atrial septal defect and a right atrial appendage-to–pulmonary artery (PA) connection (black arrow). Early atriopulmonary Fontan connections often did not involve a separate SVC connection as shown here. Both of the venae cavae were left connected to the RA, and the atriopulmonary anastomosis carried all of the systemic venous return. The use of the native RA for at least part of the venous pathway is the hallmark of an atriopulmonary Fontan. The elevated venous and right atrial pressures associated with the Fontan circulation lead to prominent right atrial enlargement after this type of connection. SV, functional single ventricle.
Since the early 1990s, many Fontan operations have included a surgical “fenestration.” A fenestration is essentially a small, intentional residual atrial septal defect (Figs. 41.3 and 41.6). Atrial baffle fenestrations are usually sized between 3 and 5 mm and allow a small, continuous right-to-left atrial shunt, at the expense of mild systemic oxygen desaturation. The physiology of the fenestration’s shunt is always right to left. This is because of the absence of a ventricle in the right side of the circulation. As a result, the right atrial/pulmonary arterial pressure must be greater than the functional left atrial/pulmonary venous pressure, or there would be no driving force for pulmonary blood flow. The purpose of allowing this residual shunt via an atrial fenestration is to provide a relatively continuous source of preload to the systemic ventricle. Prior to creation of a Fontan circulation, functionally single ventricles universally have an increased preload, since they are “filled” by both the systemic and pulmonary venous returns. At the time of Fontan creation, this extra filling volume is suddenly reduced, by diverting all of the systemic veins to the pulmonary arteries. The extra volume provided to the ventricle by a fenestration eases the transition to the Fontan circulation and maintains a slightly higher cardiac output reserve even after the initial postoperative adjustment period. Fenestration has been shown to reduce the duration of pleural drainage in the immediate postoperative period. The disadvantage associated with a fenestration is that patients will remain cyanotic and therefore have a slightly increased risk for embolic complications compared with the nonfenestrated patient.
Figure 41.2. Parasternal, horizontal-plane echocardiographic images demonstrate the tomographic anatomy of atriopulmonary Fontan connection. A: Taken during an examination of the patient with normally positioned atria and levocardia. The atrial appendage has been surgically opened and connected to the pulmonary arterial confluence, creating the Fontan connection (F). These connections are found posterior to ascending aorta, just superior to the base of the heart. B: Taken during the examination of the patient with atrial situs inversus and dextrocardia. As a result, the anatomy is a mirror image of the left panel. These images were obtained from the right parasternal border. Scans in the horizontal plane began by demonstrating the dilated native right atrial chamber. The plane of sound was then progressively moved superiorly beyond the semilunar valve to reveal the Fontan connection (F) and the confluence of the right and left pulmonary arteries (RPA and LPA). Regardless of how the surgeon created the connection, a similar progression of scans (beginning with the atrium and moving toward the pulmonary arteries) should allow echocardiographic demonstration of the Fontan anastomosis. A, atrium; AAo or Ao, ascending aorta; L, left; S, superior.
Figure 41.3. Diagrams of two types of Fontan circulations that completely bypass the entire heart to create the diversion of superior and inferior vena caval flow to the pulmonary arteries. A: This panel illustrates the use of a nonvalved conduit to complete the Fontan pathway. In this case, the base of an intraatrial conduit was connected to the inferior vena caval–atrial junction. The superior end of the conduit was then connected to the pulmonary arterial confluence using an incision/anastomosis at the dome of the native right atrium. Although the atrial wall is used to complete the connection, none of the native atrial chamber is actually involved in the venous pathway. The superior vena cava (SVC) has a separate, bidirectional anastomosis to the right pulmonary artery. B: Anatomy of an extracardiac, fenestrated Fontan operation performed in a patient with prior Norwood reconstruction for hypoplastic left heart syndrome. In this case, the inferior vena cava was detached from its native connection to the right atrium and a direct anastomosis was made between the supra-diaphragmatic inferior vena cava and a nonvalved conduit (asterisk). A superior connection was then made between the conduit and the pulmonary arterial confluence. A 4-mm fenestration (dashed, black arrow) was created between the conduit and the lateral wall of the native atrium. This allowed a small residual right-to-left shunt (white arrow) to provide a continuous source of extra ventricular filling (preload), easing the transition to the Fontan circulation. Two operations had been performed prior to the Fontan completion, as described earlier. The aorta had been reconstructed during the neonatal Norwood operation, enlarging the aortic arch and fusing the hypoplastic native ascending aorta with the native pulmonary root to create a “neoaorta.” The SVC had been bidirectionally connected to the right pulmonary artery at an intermediate, second-stage operation. PA, pulmonary artery; RA, native right atrium; RV, right ventricle; SV, functionally single ventricle.
From an echocardiographer’s perspective, a fenestration also provides insight into the patient’s pulmonary hemodynamics. The mean pressure gradient across the fenestration can be easily measured by continuous-wave Doppler echocardiography (Fig. 41.6). The fenestration “shunt” originates in the functional right atrium (the Fontan pathway), and (in the absence of stenoses) the pressure in the Fontan pathway is equal to the pulmonary arterial pressure. The fenestration flow is directed into the functionally left atrium. In the absence of pulmonary vein stenosis, the pressure in the functional left atrium will equal the pulmonary venous pressure. Therefore, the mean pressure gradient across the fenestration will reflect the “transpulmonary gradient.” This gradient is primarily determined by the patient’s pulmonary vascular resistance, a key determinant of outcome in patients with Fontan circulations. Fontan circulations that are functioning well are associated with mean fenestration (transpulmonary) gradients of 5 to 8 mm Hg. Lower values may represent better than average Fontan physiology or may represent dehydration with artificially low right atrial pressures. The higher the gradient, the higher is the total transpulmonary resistance to flow. Gradients greater than 8 mm Hg or that have increased from the patient’s historical baseline require explanation, prompting an even more thorough evaluation than usual.
Figure 41.4. Horizontal-plane echocardiographic image from a high left parasternal, subclavicular position. The image demonstrates bilateral, bidirectional superior vena cava (SVC)–to–pulmonary artery connections. These anastomoses are often referred to as bidirectional Glenn connections or shunts. This type of connection can be used as a part of a Fontan circulation, diverting superior vena caval blood flow away from the heart and into the pulmonary arteries. These connections are often created prior to final Fontan completion, in an attempt to minimize ventricular volume overload in very young patients who would otherwise not tolerate creation of a complete Fontan. In more mature patients, if these connections are not already present, they can be made at the time of Fontan completion. Ao, ascending aorta; L, left; LPA, left pulmonary artery; RPA, right pulmonary artery; S, superior.
The physiology of the Fontan operation is unique, primarily because the redirected venous flow streams do not benefit from a ventricular pump. Forward flow through the lungs depends upon a combination of residual kinetic energy from the “previous” systemic ventricular contraction, negative intrathoracic pressure (generated by the respiratory muscles), low pulmonary arterial pressure and resistance (involving both large and small vessels), as well as active atrial and ventricular relaxation. Several of these influences are reflected in the pulmonary arterial flow patterns recorded by Doppler echocardiography in Fontan patients (Figs. 41.7 and 41.8). The respiratory influence on flow is reflected by the marked augmentation in the Doppler signal occurring during inspiration. The negative intrathoracic pressure generated during spontaneous inspiration also draws blood forward through the lungs. Conversely, positive intrathoracic pressures, seen in expiration or with mechanical ventilation, will reduce forward flow. Active ventricular diastolic relaxation also serves to augment forward flow through the lungs. As the atrioventricular valve(s) open, forward flow increases through the pulmonary arteries (Figs. 41.7 and 41.8). Reduced ventricular compliance and elevated ventricular diastolic pressure will blunt this flow, reducing overall cardiac output, especially the ability to increase output with activity (cardiac reserve). Left atrial mechanical activity also affects pulmonary arterial flow in Fontan circulations. Atrial relaxation will augment forward flow by drawing blood out of the pulmonary veins, increasing overall forward flow. In contrast, atrial contraction will generally somewhat decrease the forward flow signal. If the ventricle has good diastolic compliance, the increase in pulmonary venous pressure caused by atrial contraction is offset by the increase in ventricular filling and output associated with synchronous atrial rhythms, such as normal sinus rhythm and dual chamber pacing. However, reduced ventricular compliance with elevated diastolic pressure or the atrial contraction is not synchronized with ventricular relaxation (as in junctional rhythms or heart block) will inhibit forward flow and impair cardiac output. Pulmonary venous pressures and flow reversals associated with atrial contraction in nonsynchronous rhythms are dramatically increased, reducing forward flow (Fig. 41.9).
Figure 41.5. Echocardiographic images demonstrate the appearance of an extracardiac Fontan operation in a patient with hypoplastic left heart syndrome. A: Taken at the cardiac apex; demonstrates how the native left atrial (LA) and right atrial (not labeled) chambers now serve as a combined pulmonary venous chamber, while the extracardiac, Fontan conduit (FC) functions as the patient’s new right atrium. B: Inferior connection of the Fontan conduit to the inferior vena cava and hepatic venous confluence (HV). C: Superior connection of the Fontan conduit to the pulmonary arterial confluence in a coronal plane. The scans were obtained from the left anterior axillary line with the plane of sound directed toward the patient’s right. The Fontan connection and pulmonary arteries are seen just beyond the reconstructed ascending aorta (Ao). There was no evidence of narrowing within the venous pathway or pulmonary arteries. Color flow Doppler interrogation (D) revealed laminar flow consistent with widely patent connection from the inferior vena cava, through the Fontan conduit to the pulmonary arteries. L, left; LPA, left pulmonary artery; P, posterior; RPA, right pulmonary artery; S, superior.
Figure 41.6. Echocardiographic images demonstrate an atrial fenestration after completion of a Fontan operation in a patient with asplenia syndrome. A: An intraatrial conduit (C) diverted inferior vena caval flow to the pulmonary artery. Color flow Doppler showed a continuous, aliased jet of flow (white arrow) from the conduit into the pulmonary venous atrial chamber (LA). B: Continuous-wave Doppler interrogation revealed this flow pattern. Velocities varied during each phase of the cardiac cycle and with respiration. The mean gradient between the conduit and the pulmonary venous atrium was 6 mm Hg, representing a relatively normal transpulmonary gradient after Fontan completion. L, left; RV, right ventricle; S, superior.
Figure 41.7. Pulsed-wave Doppler recording was obtained in the left pulmonary artery of the patient with tricuspid valve atresia and an atriopulmonary Fontan connection. The tracing demonstrates three important phases to “Fontan” flow in this type of connection. Forward flow (below the baseline) is augmented by active ventricular diastolic relaxation when the systemic atrioventricular valve opens (MVO). Since the native right atrial chamber remains in the systemic venous pathway, atrial contraction will also increase forward flow velocity (AC). However, when the atrium relaxes (AR), flow actually reverses out of the pulmonary artery and returns to the atrium (signal now shown above the baseline). This to-and-fro flow contributes to the atrial enlargement seen with this type of Fontan connection. Longer recordings would also demonstrate a respiratory influence on these flows, such as negative intrathoracic pressure (caused by spontaneous inspiration) that will increase forward flow volume and velocity in the Fontan circulation. In contrast, positive expiratory pressures will blunt forward flow. AC, atrial contraction; AR, atrial relaxation; MVO, mitral valve opening; LPA, left pulmonary artery; PA, pulmonary artery.
In the setting of an atriopulmonary Fontan connection, the native right atrial contraction and relaxation also alter the pulmonary arterial flow pattern. However, unlike with the left atrium, right atrial mechanical activity is inefficient in a Fontan circulation and does not actually alter cardiac output. Any augmentation to forward flow caused by right atrial contraction is counteracted by the accompanying reversal (from the pulmonary artery back into the right atrium) that occurs during atrial relaxation (Fig. 1.7). The only real impact of right atrial contraction in these patients is to create a “to-and-fro” flow through the Fontan connection, which contributes to the progressive atrial dilation common to this type of connection.
The determinants of pulmonary blood flow just described provide clues to factors that identify successful Fontan patients and some of the complications that are poorly tolerated by those with Fontan circulations. Ventricular function (both systolic and diastolic) and pulmonary resistance are probably the most critical variables related to the success and/or failure of any Fontan circulation. Table 41.1outlines these and other factors that combine to create favorable Fontan circulations. If there are multiple negative factors, the patient is likely to struggle after the Fontan and is much more likely to develop significant complications.
Figure 41.8. Pulsed-wave Doppler recording obtained from the right pulmonary artery of the patient with hypoplastic left heart syndrome and an extracardiac Fontan connection. This tracing also demonstrates phasic flows, but there are important differences to note relative to the flow patterns seen in Figure 41.7. Ventricular diastole and atrioventricular valve opening result in augmented forward flow in both types of Fontan connections. In this recording, this is reflected by the increased forward flow seen in early diastole (TVO). There is no atrial tissue in the extracardiac Fontan pathway, therefore the to-and-fro signal seen in Figure 41.7 is not evident. However, left atrial activity can still influence the pulmonary arterial flow pattern. When the “left” (pulmonary venous) atrium contracts (AC), pulmonary venous pressures rise slightly. This reduces forward flow velocity in the pulmonary artery somewhat. In an extracardiac Fontan, left atrial relaxation (AR) promotes forward flow by drawing blood out of the pulmonary veins and into the atrium. Although pulmonary arterial flow is still phasic in an extracardiac Fontan, the normal flow velocity should rarely decrease to near zero. This is unlike the flow seen in an atriopulmonary Fontan connection, where even flow reversals are common (Fig. 41.7). Intrathoracic pressure changes will produce the same alterations in these flow patterns that were described for the atriopulmonary Fontan connection in Figure 41.7. RPA, right pulmonary artery; TVO, tricuspid valve opening.
IMAGING STRATEGIES FOR EVALUATION OF PATIENTS AFTER FONTAN OPERATIONS
Most Fontan patients begin their postoperative follow-up at an older age than other patients with complex congenital heart disease. This is because of the staged nature of this surgical palliation. The fact that Fontan completion was historically performed at even older ages also contributes to this shift in demographics. As a result, the difficulties in imaging these patients are twofold. The examiner faces not only the complexity of the surgical procedure and underlying congenital heart disease, but also the reduction in image quality that is associated with increasing age and multiple prior surgical procedures. Nevertheless, the mainstay of cardiac diagnostic imaging continues to be transthoracic echocardiography. This technique provides convenience, reproducibility, and wide availability. Transesophageal echocardiography, magnetic resonance imaging, computerized tomography, and angiography all play key, but secondary, roles in obtaining structural and functional information in this patient group. These supplemental imaging strategies should be used when the clinical information required is not adequately outlined by the surface echocardiogram. The most important time to add these alternative imaging techniques to a patient’s evaluation is when he or she is clinically deteriorating, even if the changes are small.
Figure 41.9. Pulsed-wave Doppler recordings demonstrate how cardiac rhythm disturbances can also influence flows within the Fontan circulation. These signals were obtained from the right pulmonary artery (A) and the hepatic vein (B) of a patient after an extracardiac Fontan operation. These flow patterns are not normal, primarily because of the abnormal cardiac rhythm that is present. Electrocardiography reveals complete heart block with a junctional escape rhythm. The atrium is contracting more often than the ventricle in this case. Since atrial contractions are dyssynchronous in this rhythm, they will produce significant elevations of pulmonary venous pressure (cannon A waves). These pressure increases are reflected not only in the pulmonary arterial flow signal, but are also transmitted all the way back to the hepatic veins (even though there is no “atrium” within the systemic venous pathway). A, yellow arrows: Reductions in forward flow velocity caused by atrial contraction. One of the atrial contractions during this recording occurred so early (white arrow) that it actually caused flow reversal in the pulmonary artery. This phenomenon was much more evident when the hepatic venous flows were recorded (B). Cannon A waves with large flow reversals (white arrows) could be seen during nearly every cardiac cycle (yellow arrow). The one cardiac cycle in this recording in which atrial contraction occurred at approximately the “correct” time was relative to ventricular contraction. In this cardiac cycle, there was a slight decrease in forward velocity after atrial contraction, but no reversals were observed.
Knowledge of the patient’s clinical status is also helpful to choosing an imaging strategy. For example, the patient with increasing fatigue must be evaluated for worsening ventricular systolic or diastolic performance, as well as arrhythmias. The patient with a recent stroke or transient ischemic accident must have a detailed search for the source of the emboli. While imaging these patients, one must strive to obtain detailed, high-quality images. Unfortunately, we know that after the Fontan procedure, patients often have challenging acoustic windows. When surface echocardiography does not provide adequate detail, transesophageal echocardiography is often useful in visualizing the anatomic areas in question. Transesophageal echocardiography is particularly well suited to evaluating posterior structures such as the atria (to exclude thrombus formation), the Fontan connections (to exclude obstruction), and the atrioventricular valves. Magnetic resonance imaging can be helpful in identifying venous or arterial abnormalities and in quantitating ventricular function, assuming the patient does not have an electronic pacemaker. Cardiac catheterization and angiography still play an important role in the evaluation of a patient’s hemodynamic status after the Fontan operation. Catheter-derived hemodynamics remain the gold standard for determination of precise pressure measurements for comparison to the patient’s historical baseline and pulmonary vascular resistance.
We will focus the remainder of this chapter on surface echocardiographic evaluations, supplementing the discussion with examples of more invasive imaging strategies, when appropriate.
IMAGE ACQUISITION AND THE ANATOMY OF THE FONTAN RECONSTRUCTION
The most important tool in any assessment of a Fontan patient is the surgical dictation. To perform an adequate examination, one must know exactly how the Fontan circulation was created. Early atriopulmonary connections often used the right atrial appendage as the final portion of the pathway for systemic venous flow (Figs. 41.1 and 41.2). The atrial septum was closed in its natural position, and any atrioventricular connection from the right atrium to the ventricle was also closed. Similarly, the pulmonary artery, if present, was ligated and/or divided. Although the atriopulmonary Fontan achieves separation of the systemic and pulmonary venous flow streams, it leaves a large and distensible chamber (the right atrium) in the middle of the systemic venous pathway. Progressive right atrial enlargement, intraatrial thrombi, and persistent atrial arrhythmias have plagued the post-Fontan patient with this type of connection. Intracardiac thrombi can be detected by surface echocardiography (Fig. 41.10). However, transesophageal examinations are more effective in detecting these complications (Fig. 41.11), especially since the transthoracic image quality available in an older Fontan patient is often impaired.
As a result of these problems, more recent surgical connections have been modified and become more streamlined. The most common method used to create a Fontan circulation today combines bidirectional superior vena cava to pulmonary arterial anastomosis(es) with an extracardiac conduit, creating continuity between the inferior vena cava and the pulmonary artery. This type of connection, the extracardiac Fontan, is illustrated in Figures 41.3, 41.4, and 41.5. Not only does an extracardiac Fontan eliminate the dilated right atrial chamber, but it also allows use of both native atria and both atrioventricular valves in the systemic circulation. This approach simplifies creation of a Fontan circulation for patients with anomalous pulmonary venous connections and abnormalities of the atrioventricular valve(s).
These two types of Fontan pathways do not represent the only Fontan connections that are possible. This heterogeneity in surgical approach contributes to making the surgical dictation so valuable. The surgical report also allows the examiner to be confident at the conclusion of the study that all of the components of the Fontan have been assessed.
As with all studies, we tend to begin the examination of the Fontan patient from the subcostal window. Evaluations of the liver, hepatic veins, and inferior vena cava provide an insight into systemic venous pressure (dilated veins suggest elevated pressure). The presence of spontaneous echo contrast is associated with sluggish flow and/or reduced cardiac output. This same acoustic window allows the examiner to then angle the plane of sound superiorly across the diaphragm, producing a subcostal coronal image of the heart. In this position, one can visualize the systemic venous pathway, atria, and usually the dominant ventricle. Although visualization of the pulmonary arterial connection is often difficult from this transducer position, the most inferior segments of the pathway—inferior vena cava, the extracardiac conduit, and/or the inferior portion of the atrium—are usually well seen. Sweeps in both the coronal and sagittal planes should be performed. The examination should be geared to detect chamber enlargement, quantitate ventricular wall motion, scan for potential thrombi and intraatrial shunting, and locate the position of the inferior vena caval pathway to the pulmonary artery. Doppler echocardiography in this transducer position is usually limited to color flow mapping as well as interrogation of hepatic venous and atrial fenestration flow patterns.
Figure 41.10. These images were taken from an examination of a patient with tricuspid valve atresia, normally related great arteries, and restrictive, but patent, ventricular septal defect and a previous lateral tunnel Fontan connection. A: Expected anatomy as observed from the cardiac apex. The Fontan pathway (F) can be seen along the right level border of the native atrium. B: High, left parasternal horizontal-plane image. It reveals a relatively large, organized thrombus within the blind ending pouch of the patient’s native main pulmonary artery (MPA). The MPA had been “closed” with a patch at the time of Fontan completion. This created an area of limited flow, since the ventricular septal defect allowed some blood flow to reach the pulmonary arterial stump. The Fontan connection to the pulmonary arterial confluence was widely patent. These “blind pouches” (in this case the MPA) create a risk of systemic embolization when thrombi form within them, since they retain connection to the aortic circulation. In this case, the patient was anticoagulated, and the thrombus resolved without a complication. A, anterior; F, Fontan pathway/connection; L, left; LA, left atrium; LPA, left pulmonary artery; LV, left ventricle; MPA, main pulmonary artery; RPA, right pulmonary artery; S, superior.
Figure 41.11. These images were obtained during an intraoperative transesophageal echocardiogram. The patient had stenosis of his atriopulmonary Fontan connection and had developed massive right atrial enlargement. Surface echocardiography showed spontaneous echo contrast within the atrial cavity, but did not reveal the Fontan connection or evidence of organized thrombus. A: This image was obtained from the distal esophagus in a “four-chamber” orientation. It shows a large mural thrombus along the lateral right atrial wall (yellow arrow).There was also prominent spontaneous contrast noted within the atrium (dashed red arrow). B: This sagittal image of the same area shows the extent of the thrombus (yellow arrow) and the position of the superior vena caval-right atrial junction (SVC). L, left; LA, left atrium; LV, left ventricle; P, posterior; RA, right atrium; S, superior.
The transducer is then transferred to the parasternal area. The majority of the Fontan reconstruction will be visualized from acoustic windows on the anterior chest wall. The atriopulmonary and extracardiac conduit connections can usually be imaged from a parasternal window. The most convenient images of this connection are often found in the horizontal (short-axis) projections (Figs. 41.2, 41.4, 41.5, and 41.10). However, the best visualization of the Fontan reconstruction may not always be obtained from the usual left parasternal location. The examiner should pass the transducer across the entire chest, focusing on the left and right parasternal borders, as well as the left anterior axillary line (Fig. 41.12). Repositioning the patient often improves acoustic quality. Images from the left chest are often optimal when the patient is in a left lateral decubitus position. Having the patient lie on the right side facilitates right parasternal imaging. Occasionally, simply having the patient in supine position provides the best imaging window.
In addition to routine imaging of the intracardiac structures (focused on the valves, atria, and ventricles), serial images following the inferior vena caval pathway toward the Fontan anastomosis should be obtained. The most convenient way of tracing this pathway is to use a series of horizontally oriented (short-axis) scans obtained by gradually moving the transducer from the inferior costal margin to a more superior position on the chest. When the pulmonary arterial confluence is visualized, the examiner can be confident that the position of the inferior Fontan connection will be slightly below that level. Regardless of how the connection is visualized, it is often found just posterior and slightly to the right of the ascending aorta. If one follows the flow stream beyond the connection, it will lead to the pulmonary arterial bifurcation. Color flow Doppler is often quite helpful not only in tracing the vena caval flow to the Fontan, but also in defining the transition between Fontan pathway and the branch pulmonary arteries. Since venous and pulmonary arterial flows in the Fontan circulation have relatively low velocities, reducing the Nyquist limit to 60 cm/s or less is recommended.
Figure 41.12. Echocardiographic images of a nonstandard view of the Fontan connection and pulmonary arterial confluence in a patient after an extracardiac Fontan for “double-inlet left ventricle.” As described in the text, the transducer was positioned along the left anterior axillary line just lateral to the pectoralis major muscle group. The image reveals the Fontan connection (FC) and both pulmonary arteries just beyond the aorta (Ao). Since this is a coronal scan originating from the left lateral surface of the chest, the right pulmonary artery (RPA) is seen in the far field. Doppler interrogation of the vessel produces excellent signal quality, since RPA flow is parallel to the plane of sound from this transducer position (laminar, blue flow signal, right), but spatial resolution of the vessel walls is less optimal. In contrast, the left pulmonary artery (LPA) is clearly seen in the two-dimensional scan in the left panel, but Doppler flow is difficult to demonstrate since the long axis of the vessel is perpendicular to the imaging plane. L, left; P, posterior.
The scans used to define the inferior pathway to the Fontan can be continued more superiorly to detect any connections from the superior vena cava that may have been created as a part of the patient’s surgical palliation. Once the inferior pathway, the pulmonary arteries, and the superior pathway(s) have been identified in horizontal images, the scan planes can be shifted into a more sagittal display to show the long axis of these connections and pathways. Documentation of the venous pathway’s diameter at the level of the inferior vena cava, the mid-atrium, the pulmonary arteries, and the superior vena cava is helpful in determining whether any significant obstructions are present. Color flow and spectral Doppler interrogation of the suture lines and any fenestrations help define any obstructions and assess the transpulmonary resistance.
Nonstandard acoustic windows can greatly enhance the examination. The “inferior” Fontan connection (atriopulmonary Fontan or extracardiac conduit) can often be visualized using a modified left anterior axillary window. This image is obtained with the transducer positioned just lateral to the pectoralis major muscle, near the axilla. The plane of sound is angled “back” toward the midline, resulting in an imaging plane that is nearly parallel to the long axis of the right pulmonary artery (Fig. 41.12).
Attention is then shifted to the apical window. The long axes of the Fontan connections and venous pathways are often not seen well in this orientation. Some short-axis images of Fontan conduits (extracardiac or intracardiac) and lateral tunnels are usually available at the apex. Residual right-to-left shunts from the Fontan pathway or native right atrium are easily detected by color flow Doppler from this window. Any right-to-left atrial shunt lesions should also be interrogated with continuous-wave Doppler, to determine the mean gradient of that flow. That gradient will reflect the drop in pressure from the Fontan pathway to the pulmonary veins or transpulmonary gradient. This value provides crucial insight into the resistance to blood flow within the pulmonary arterial bed.
Finally, the transducer is moved to the high parasternal and suprasternal windows. The upper systemic veins (superior vena cava, jugular, and innominate veins) and aortic arch are evaluated as has been described elsewhere.
ASSESSMENT OF VENTRICULAR PERFORMANCE
The most important variable related to ongoing success of a Fontan operation is ventricular performance. The systemic ventricle must maintain normal (or nearly normal) systolic contractility and low diastolic filling pressure. The complex geometry of functional single ventricles makes the use of a single method of assessment impractical. In hearts with primarily left ventricular morphology, standard methods of calculating ejection fraction, circumferential fiber shortening, and wall stress continue to be clinically useful. When ventricular morphology is more complicated, one must resort to other methods of assessment. An important point to remember is that any method used must be reproducible, both between examiners and over time. In these situations, we have used methods that do not depend on ventricular geometry. Determinations of fractional area change in multiple planes (usually parasternal short-axis and the apical “four-chamber” views) are helpful in documenting ventricular wall motion (Fig. 41.13). Normal values for both right and left ventricular fractional area change are greater than 40%. The myocardial performance index and Doppler-derived systolic rates of ventricular pressure change (dP/dT) can also provide information about ventricular performance. All of these “nonstandard” parameters gain added value when followed over time to outline trends in the same patient.
Detailed discussions of the methodologies used to assess ventricular diastolic filling pressure can be found in the section of Chapter 3 discussing evaluation of ventricular diastolic function. Pulsed-wave Doppler techniques have been validated in patients with single ventricle physiology. Tissue Doppler and myocardial deformation imaging are likely to provide additional insights, but require validation in this group. Pulmonary venous atrial reversal duration and atrioventricular valve diastolic deceleration time have been the most useful diastolic parameters in Fontan patients. Significant prolongation (>28 ms) of the pulmonary venous atrial reversal relative to atrial forward flow duration into the ventricle has been shown to be a reliable sign of elevated filling pressure, even in the Fontan circulation. Reductions in deceleration time have been associated with prolonged chest tube drainage after Fontan creation and increased mortality risk in patients with protein-losing enteropathy. It is our impression that gradual changes in these parameters provide early warning of impending significant ventricular dysfunction. Therefore, thorough diastolic filling assessments should be performed regularly and followed serially in these patients.
Figure 41.13. Diastolic (A) and systolic (B) frames taken from the apex of a common ventricle after Fontan operation. A portion of the extracardiac conduit (asterisk) can be seen adjacent to the posterior and rightward “corner” of the common atrium (A). Ventricles with mixed or right ventricular morphology do not follow standard geometric conventions. In these cases, an index of ventricular function can be obtained by simply comparing the diastolic and systolic areas of the ventricular cavity. Once these areas are traced, as in the figure, a two-dimensional, systolic fractional area change (FAC) can be calculated by dividing the difference between the two areas by the diastolic area (see formula). This measurement is analogous to the linear systolic shortening fraction determined by M-mode. Similar to shortening fraction and ejection fraction, this parameter will be influenced by alterations in preload and afterload, but it offers a simple, reproducible measurement that is related to systolic function, even in geometrically complex ventricles. The measurement can be obtained from one or multiple planes and normal values are generally greater than 40%.
FAC = (diastolic area – systolic area)/diastolic area
The ventricle illustrated in this figure had diastolic and systolic areas of 15.7 and 9.7 cm2, respectively. The difference is 6 cm2 and the FAC is therefore equal to 6 divided by 15.7 (apical FAC = 38%). L, left; S, superior.
VENOUS AND ARTERIAL PATHWAY OBSTRUCTIONS
Obstructions at any level are less well tolerated by patients with Fontan physiology than those who have biventricular circulations. Subaortic stenosis, coarctation, and pulmonary arterial distortions represent the most common obstructive problems seen after the Fontan operation. Assessments of ventricular outlet and aortic arch obstructions are performed using the same techniques described in the chapters focused on these abnormalities as seen in “biventricular” hearts. However, the examiner needs to recognize that even mild degrees of stenosis can have significant adverse effects in the Fontan patient. Subaortic and arterial obstructions with even the low gradients (15 to 25 mm Hg) can negatively affect cardiac performance and output reserve in Fontan patients.
Pulmonary arterial stenosis requires a different method of evaluation in the Fontan patient than in those with biventricular circulations. Since there is no pulmonary ventricular pump, flow in the pulmonary vascular bed is nonpulsatile. Thus, stenoses in the Fontan pathway and pulmonary arteries will display hemodynamics that are more similar to venous obstructions. As a result, Doppler flow gradients are often misleadingly low. Therefore, the focus of the pulmonary arterial examination should be to carefully define the size of the pathways leading to the Fontan connections, the size of the connections themselves, and the size of the pulmonary arteries centrally (Fig. 41.14). Any abrupt decrease in vascular diameter should be considered a potential stenosis. Particular attention should be paid to areas in which the pathway narrows and then enlarges again downstream. These areas should be interrogated with both continuous and pulsed-wave Doppler. A reduction in flow variability either proximal to (Fig. 41.15) or beyond the narrowing is suggestive of obstruction. If there is an increase in velocity through the narrowed segment, a mean gradient (measured over multiple cardiac cycles) should be calculated (Fig. 41.13). A mean gradient of greater than 3 mm Hg should be considered significant and at a minimum warrants additional evaluation and possibly intervention.
Figure 41.14. Branch pulmonary arterial stenosis after an atriopulmonary Fontan operation. This echocardiographic image and the Doppler signal demonstrate a discrete, but severe proximal left pulmonary artery (LPA) stenosis (yellow arrow). The narrowed segment is short, but the internal pulmonary artery diameter is less than 50% of the downstream diameter. Right: Continuous-wave Doppler interrogation of the flow crossing this stenosis produced this signal. Flow velocities are elevated, relative to what is usually seen in the Fontan patient. However, maximum flow velocities only reach 1.7 m/s. The mean gradient, averaged over multiple cardiac and respiratory cycles, was 7 mm Hg. Note that although the flow is somewhat phasic, the velocity profile never reverses during atrial relaxation. In fact, it never even approaches the baseline, as one would expect in a patient with an atriopulmonary connection. The combination of the absence of the ventricle in the pulmonary circulation of the Fontan patient, this luminal narrowing, and its associated pressure gradient, significantly limited the patient’s ability to increase his cardiac output and had been associated with a progressively enlarging right atrial chamber. His exercise capacity improved after placement of a left pulmonary artery stent and revision of the Fontan connection using an extracardiac conduit. A, anterior; Ao, aorta; FC, Fontan connection; L, left.
PROBLEMS ASSOCIATED WITH ATRIAL ENLARGEMENT
Direct atrial–to–pulmonary arterial Fontan connections can develop complications that are rarely seen with either intraatrial or extracardiac conduits. The native atrial tissue left in the systemic venous circulation is under a higher than normal distending pressure. In addition, since there are no valves in the Fontan pathways, atrial contraction and relaxation will produce an additional stimulus for atrial enlargement. As a result, these atria will dilate, usually in a progressive fashion. These dilated chambers can compress nearby vascular structures. The pulmonary veins are most susceptible to this compression, because their distending pressures are lower than those within the Fontan circuit (Fig. 41.16). Pulmonary venous compression/obstruction is poorly tolerated by patients with Fontan physiology. Flow through the segments of lung drained by the compressed pulmonary veins will be reduced. This will certainly reduce the patient’s ability to increase cardiac output, and in severe situations may even decrease the resting cardiac output.
The dilated atria seen in patients with atrial pulmonary connections create areas where blood flow is sluggish. These atria are therefore prone to mural thrombosis formation (Figs. 41.10 and 41.11). These thrombi are rarely obstructive, but can contribute to embolic disease. Last, it is thought that distended atrial tissue is at increased risk for generating abnormal, reentrant tachyarrhythmias, particularly atrial flutter. These abnormal atrial rhythms can be detected echocardiographically by careful analysis of atrial wall motion or the atrial influences on venous or pulmonary arterial flow patterns (Figs. 41.9 and 41.17).
ECHOCARDIOGRAPHIC CLUES TO A FAILING FONTAN CIRCULATION
A number of echocardiographic findings can provide clues to deteriorating cardiovascular function (Table 41.2). Most of these abnormalities can be detected by standard echocardiographic techniques. The gradual rate at which some of these changes occur make it necessary to track these parameters, particularly chamber sizes and indices of contractility, consistently over time. When present, these dysfunctional findings should prompt additional investigation, even if the primary problem is not evident to standard interrogations.
A few additional simple scans are required to document the presence or absence of ascites and effusions. The inferior vena cava of Fontan patients will be more prominent than in patients with biventricular circulations. This is because of the expected increase in systemic venous or “right atrial” pressure that accompanies direct connection of the systemic veins to the pulmonary arteries. However, systemic venous pressures will increase even further in the face of decreasing cardiac output or ventricular function. Consequently, the inferior vena cava will tend to progressively enlarge in patients with deteriorating cardiac function and progressively elevating right atrial pressures. As a result, serial evaluation of the inferior vena cava diameter should be included in the postoperative evaluation of all Fontan patients. This evaluation can simply include measurement of expiratory diameter, with adequate two-dimensional visualization to allow exclusion of thrombus and spontaneous contrast. The most convenient approaches for obtaining these images involve acquiring a sagittal-plane view of the inferior vena cava from either the central subcostal transducer position or a more lateral transducer position over the body of the liver.
Figure 41.15. Alterations in superior vena caval flow patterns in the presence of downstream Fontan obstructions. The best method of determining the presence or absence of stenosis within a Fontan pathway is to directly visualize the vessels involved. However, image quality after the Fontan operation is often suboptimal. Analysis of venous Doppler flow patterns can provide clues regarding the status of downstream venous connections to the pulmonary circulation. Left: Three pulsed-wave Doppler signals were obtained from the mid-portion of the superior vena cava (SVC) in three different patients. All three had significant downstream stenoses in their circulations. The top signal (1) was obtained from the patient with near-complete occlusion of the SVC. There is little phasic change in the flow pattern. Flow does increase during ventricular diastole, but never returns to baseline. Even respiratory variation is absent; this is extremely abnormal in the venous circulation. The middle (2) and bottom (3) signals were obtained from patients with less obstruction. There is some respiratory variation and augmentation of forward flow during ventricular diastole. However, similar to the top signal (1), flow velocities never return toward the baseline, indicative of a pressure in the SVC which is constantly greater than the chamber or vessel into which it flows. Normal flows are illustrated by the two signals in the right side of the figure. Since SVC pressure equalizes with downstream pressures in the absence of obstruction; the resulting signal will show phasic, low-flow velocities or even flow reversals after atrial contractions. Top right: Signal was obtained in a patient with an atriopulmonary Fontan connection and a widely patent SVC. The patient was in a junctional rhythm during the recording, and as a result the atrial reversals are seen after the QRS complex. Nonetheless, atrial activity is shown to influence the flow pattern in the upper portions of the SVC, confirming a widely patent connection between those two points in the cardiovascular system. Bottom right: Signal was obtained from the patient with a bidirectional SVC–to–right pulmonary artery anastomosis. There are no reversals seen, but there is respiratory variation and appropriate reductions and augmentations to forward flow not only during ventricular diastole, but also with left atrial contraction and relaxation (Fig. 38.8). Evidence of atrial activity in the vena caval flow pattern again confirms the patency of the connection in this case. AP, atriopulmonary; BDCPA, bidirectional cavopulmonary anastomosis.
Echocardiographic evaluation of cardiac rhythms is usually confined to the prenatal practice. However, the electrocardiogram of the post-Fontan patient can be difficult to interpret. Therefore, the echocardiographer should report the presence of normal atrial activity when it is evident. Abnormally rapid or dyssynchronous atrial contractions can often be detected by careful analysis of the venous Doppler flow tracings. Pulmonary venous and arterial flow signals will show an increased number and rate of atrial “indentations” in the forward flow signal, or in more extreme cases actual reversals, when the patient is in atrial flutter or junctional rhythm (Fig. 41.17). The abnormal flow pattern (atrial activity) can be easily related to the QRS complex on the simultaneous rhythm strip to determine the cardiac rhythm. Alternatively, if atrial wall motion is evident, M-mode tracings of the atrial appendage can provide the same type of information.
Figure 41.16. Pulmonary venous compression secondary to right atrial enlargement. Images display massive right atrial (RA) enlargement in a patient with a very complex past surgical history. The primary lesion was that of the “double-inlet left ventricle.” His surgical history included pulmonary arterial reconstructions, subaortic resection, several palliative shunts, and eventually an atriopulmonary Fontan completion. As expected, his ventricular diastolic compliance was poor and led to multiple complications. At the time of this examination, the patient’s RA diameter was greater than 7 cm. A:Horizontal-plane image shows the Fontan connection to be widely patent with normal flow seen in both of his pulmonary arteries. B: Relationship of the dilated right atrium to the left atrium (LA). The LA was significantly smaller than the right atrium. Color Doppler detected aliased flow in the left lower pulmonary vein (LP vein), and continuous-wave Doppler revealed a mean gradient of 5 mm Hg between the left pulmonary vein and the LA. The lumen of the right pulmonary vein (RP vein) was difficult to appreciate and Doppler flow could never be detected. This kind of compression severely limits the amount of cardiac output that can pass through the lungs. As might be expected, this patient had symptoms of low cardiac output and tremendous sodium retention with large amounts of ascites and recurrent pleural effusions. Unfortunately, despite an attempted surgical revision of his Fontan pathway, he did not survive. A, anterior; L, left.
Figure 41.17. Detecting atrial arrhythmias using echocardiography. Atrial electrical activity is often difficult to visualize on the standard electrocardiogram of a Fontan patient. The images in this figure were taken from an examination of a patient with hypoplastic left heart syndrome after an atriopulmonary Fontan operation who had developed atrial flutter with 3:1 atrioventricular conduction. Although the rhythm strip does not show discernible P waves, the M-mode tracing (bottom) shows multiple contractions of the left atrial appendage during each cardiac cycle (yellow arrows). Similarly, the pulsed-wave Doppler flow signal (top) shows multiple reductions in forward flow (white arrows) that cannot be explained by the cardiac cycle. When a Fontan patient shows rapid narrow complex tachycardia or a relatively constant heart rate with no visible P waves or a prolonged PR interval, one must suspect atrial flutter. Careful analysis of the venous and pulmonary arterial flow patterns or M-mode interrogation of the atrial wall can clarify the origin of the rhythm in these cases. LA, left atrium; RPA, right pulmonary artery.
In some situations, pulmonary arterial and hepatic venous flow signals can also provide evidence of atrial contractions. When the native right atrium remains in communication with the venae cavae in patients (as with atriopulmonary connections), atrial flow reversals will be seen in the hepatic veins and can be used to assist in determining the rhythm in the same way that was described for the pulmonary veins (Fig. 41.7). In atrioventricular dyssynchronous rhythms, like complete heart block or junctional rhythm, hepatic “cannon” atrial reversals caused by left atrial contractions can be seen even in patients with extracardiac conduits (Fig. 41.9).
As in all patients, comprehensive examinations of valvar function need to be performed in patients after Fontan procedures. These assessments have been covered elsewhere in this textbook, and detailed discussion of the methods used will not be repeated here.
ILLUSTRATIVE CASES INVOLVING PATIENTS WITH FONTAN CIRCULATIONS
The images for this case were taken shortly after creation of an extracardiac, fenestrated Fontan circulation in a 3-year-old with tricuspid valve atresia. The patient has been dismissed from the hospital and was steadily improving. Figure 41.18 contains apical four-chamber views of the extracardiac conduit (red arrow), pulmonary venous atrium (native right and left atrium), mitral valve, and left ventricle. The aliased color flow signal in the top right panel (yellow arrow) and the continuous-wave Doppler signal in the bottom panel were generated by flow through the fenestration. Both Doppler signals confirm the right-to-left shunt flow. The spectral Doppler tracing shows that the mean “Fontan–to–left atrial” or transpulmonary gradient is 7 mm Hg. The velocity of the right-to-left shunt decreases after left atrial contraction (white arrows), but fenestration flow is continuous, as one would expect in a patient after Fontan creation. Figure 41.19 documents the patency of the anastomosis between the extracardiac conduit and the pulmonary artery. Figure 41.20 confirms the adequacy of the preexisting bidirectional right superior vena cava–to–right pulmonary artery connection (asterisk) and reveals a normal phasic flow pattern in the superior vena cava (lower pulsed-wave Doppler tracing). The Fontan conduit–to–inferior vena cava connection is also appreciated in these images (red arrows).
Figure 41.18. Case 1. Apical “four-chamber” views and continuous-wave Doppler flow signal generated by the atria I fenestration described in the text and obtained from a patient with tricuspid valve atresia after a fenestrated, extracardiac Fontan completion. CW, continuous-wave; FC, Fontan conduit; L, left; LA, left atrium; LV, left ventricle; S, superior; Ρ, pressure gradient.
These images were taken from an examination of a 7-year-old with hypoplastic left heart syndrome. He had a history of neonatal Norwood operation, followed by bidirectional right superior vena cava–to–right pulmonary artery anastomosis at 7 months of age, and Fontan completion at 3 years of age. Figure 41.21 shows the apical four-chamber anatomy. Unlike the patient in Case 1, pulmonary venous flow must pass from the native left atrium and cross the plane of the resected atrial septum to reach the right atrium and tricuspid valve. The fenestration had spontaneously closed, leaving no evidence of residual intracardiac shunt. There is a trivial amount of tricuspid valve regurgitation detected. Pulmonary venous and tricuspid valve pulsed-wave Doppler recordings are consistent with low filling pressure. Ventricular function was normal with an apical fractional area change value of 42%. The arterial reconstruction associated with the Norwood operation is seen in Figure 41.22. The patency of these connections must be confirmed, as any residual obstruction will create unnecessary afterload, potentially compromising ventricular function. The connection between the extracardiac conduit and the pulmonary arterial confluence is seen in Figure 41.23. The connection is not restrictive, although the branch pulmonary arteries are somewhat small (common after the Norwood reconstruction). The pulsed-wave Doppler signal in the bottom panel shows a normal flow pattern within the right pulmonary artery. Appropriate variation is seen with both respiration and “left” atrial and ventricular activity.
Figure 41.19. Case 1. Coronal views of the Fontan connection from the left lateral axillary transducer position. Two-dimensional scans (A) and color Doppler (B) showed no narrowing in the Fontan connection (FC) or the pulmonary arteries. Ao, aorta; L, left; LPA, left pulmonary artery; P, posterior; RPA, right pulmonary artery.
Figure 41.20. Case 1. Top: Long-axis (sagittal) images of the superior vena cava (SVC) and Fontan conduit (FC). The preexisting SVC–to–right pulmonary artery anastomosis is marked (asterisk). The FC–to–inferior vena cava transition is highlighted (red arrow). No narrowings were noted anywhere within the systemic venous diversion. Color flow Doppler profiles and the pulsed-wave Doppler signal from the superior vena cava were normal (right and bottom). Note the phasic changes in the Doppler signal that relate to both cardiac activity, as well as the respiratory cycle. A, anterior; Ao, ascending aorta; S, superior.
Figure 41.21. Case 2. Apical “four-chamber” views and pulsed-wave Doppler flow signals recorded at the tips of the tricuspid valve leaflets and within the right lower pulmonary vein of a patient with mitral valve atresia after a neonatal Norwood operation and eventual fenestrated, extracardiac Fontan completion. The fenestration had spontaneously closed 2 years before this examination. Color flow interrogation of the tricuspid valve showed only mild regurgitation (top right). Pulsed-wave Doppler signals at the tricuspid valve and in the pulmonary veins showed normaI diastolic filling. Forward flow into the ventricle with atriaI contraction was prominent and there was no evidence of reversed flow in the pulmonary vein after the P wave. These findings are consistent with low ventricular filling pressure, a key component of favorable physiology in patients with Fontan circulations. FC, Fontan conduit; L, left; LA, left atrium; P, pulmonary; RV, right ventricle; S, superior; TV, tricuspid valve.
These images were taken during an examination of a 56-year-old woman with tricuspid valve atresia and normally related great arteries. As a 5-year-old, she had creation of a right Blalock-Taussig shunt. At age 39, an atriopulmonary Fontan was created. Her past history was also remarkable for three episodes of endocarditis prior to the Fontan operation and multiple episodes of atrial tachyarrhythmias, now controlled with chronic amiodarone administration. Her atrial arrhythmias began approximately 12 years after completion of the Fontan. She presented for this examination with increasing fatigue and shortness of breath. Her echocardiogram revealed a distended inferior vena cava with a moderate amount of spontaneous contrast within it. Figure 41.24 contains the apical four-chamber view. Her right atrium was massively enlarged, the left atrium and ventricle were moderately enlarged, and ventricular ejection fraction was calculated to be 55%. The right panels of Figure 41.24 show that her atriopulmonary Fontan connection was widely patent and there was no distortion of her proximal pulmonary arteries. Pulsed-wave Doppler interrogation of the right pulmonary artery and hepatic vein showed phasic flow changes because of atrial contraction and relaxation, further confirming the patency of the Fontan connection (Fig. 41.25). Doppler interrogation of the mitral inflow tract (Fig. 41.26) revealed limited atrial forward flow duration (yellow arrows) and the presence of a pathologic mid-diastolic L-wave (white arrow, Fig. 41.26, top). Pulmonary venous flow pattern (Fig. 41.26, bottom) showed brief early diastolic augmentation (after mitral valve opening) with prominent and prolonged flow reversal after atrial contraction (yellow arrows, Fig. 41.26, bottom). Note that her heart rate was only 45 beats per minute during the examination. Exercise testing showed a blunted heart rate response, reaching a maximum of only 85 beats per minute. This degree of bradycardia is poorly tolerated in the setting of reduced ventricular compliance, since increasing heart rate is one of the few methods available to increase cardiac output in these patients. Since no surgical or catheter intervention would have altered her physiology, her medication regimen was adjusted. Her beta-blocker was discontinued and the amiodarone dose was decreased in an attempt to increase her heart rate. Her diuretic therapy was intensified. These maneuvers resulted in a small improvement in her symptoms, but unfortunately her activities remained limited by her sinus node and severe diastolic dysfunction. Pacemaker implantation is currently being considered as a method to improve her heart rate response to exertion.
Figure 41.22. Case 2. Complex arterial reconstruction involved in the Norwood operation. Examination of these patients both before and after the Fontan operation must confirm the absence of obstruction throughout the thoracic aorta. A: Sagittal-plane image shows the proximal arterial roots near the base of the heart. The native pulmonary artery (PA) and hypoplastic ascending aorta (Ao) have been surgically merged into one functional “neoaortic” root. B: Oblique, high left parasternal transducer position. It shows the native pulmonary valve, ascending “neoaorta” (NeoAo), transverse aorta arch (TrAo), and upper descending aorta (DAo). In this case, the arterial reconstructions were widely patent and the native pulmonary valve showed no evidence of regurgitation. Narrowings of the connection between the native pulmonary artery and the hypoplastic native aorta can compromise coronary blood flow, while residual stenosis in the arch reconstruction (coarctation) causes an unacceptable pressure load on the systemic right ventricle (RV). A, anterior; LA, left atrium; S, superior.
Figure 41.23. Case 2. Two-dimensional and color flow images show widely patent Fontan connection (FC) with normal flow in both the proximal right and left pulmonary arteries (RPA and LPA). The pulsed-wave Doppler tracing shows normal phasic activity in the right pulmonary artery as previously described. AC, atrial contraction; Ao, aorta; AR, atrial relaxation; FC, Fontan connection; L, left; LPA, left pulmonary artery; P, posterior; RPA, right pulmonary artery; TVO, tricuspid valve opening.
Figure 41.24. Case 3. Echocardiographic images show the apical “four-chamber” anatomy, as well as the atriopulmonary Fontan connection and pulmonary arterial confluence in this 56-year-old patient. The right atrium (RA) is tremendously dilated. The Fontan pathway was widely patent. Other pertinent findings are summarized in the text. A, anterior; Ao, aorta; FC, Fontan connection; L, left; LA, left atrium; LPA, left pulmonary artery; LV, left ventricle; P, posterior; RPA, right pulmonary artery.
Figure 41.25. Case 3. Pulsed-wave Doppler tracings. Both the hepatic vein and right pulmonary artery (RPA) signals show the phasic flows related to atrial contraction and relaxation (arrows) that we expect with patent Fontan connections. However, the degree of forward flow reduction during the expiratory phase of the respiratory cycle was more prominent than usual (red arrow), suggesting elevated systemic venous pressure.
Figure 41.26. Case 3. Pulsed-wave Doppler recordings revealed the primary hemodynamic problems afflicting this patient. A: Mitral valve (MV) inflow Doppler signal. Mid-diastolic deceleration time was relatively normal (185 ms), but there was a prominent mid-diastolic L-wave (white arrow), indicating significantly abnormal/delayed left ventricular relaxation. The transmitral atrial filling wave (yellow arrows) is abbreviated and ends at approximately the same time as the S wave on the ECG. B: Right lower pulmonary vein (P vein). It shows extremely low-velocity diastolic forward flow into a prominent, prolonged atrial reversal (yellow arrows). Venous reversal extends well beyond the S wave. In fact, the reversal duration was 60 ms longer than the atrial forward flow duration. These findings are consistent with reduced diastolic ventricular compliance and significantly elevated ventricular end-diastolic and mean left atrial pressures. One should also note that the patient’s heart rate was quite slow (45 bpm). This type of bradycardia may be poorly tolerated in the face of such severe ventricular diastolic dysfunction.
These transesophageal images were taken during examination of a 31-year-old man with tricuspid and pulmonary valve atresia. He complained of increasing fatigue and exercise intolerance. He had undergone creation of an atriopulmonary Fontan connection 24 years prior to this examination. Before the Fontan completion procedure, he had been palliated with a Waterston shunt. As a result, his pulmonary arterial confluence was significantly distorted. His Fontan connection included a direct right atrial–to–right pulmonary artery anastomosis and placement of a supplemental 22-mm nonvalved conduit from the right atrium to the left pulmonary artery to bypass the acquired stenosis of his pulmonary arterial confluence. His transthoracic echocardiogram showed good ventricular function, mild mitral regurgitation, and a very large right atrium. The Fontan connections could not be visualized by surface echocardiography and this transesophageal echocardiogram was performed. Figure 41.27 shows two images from the distal esophageal four-chamber window. Scans from this position confirmed satisfactory systolic function and only a mild degree of central mitral valve regurgitation. Figure 41.28demonstrates that the right atrial–to–right pulmonary artery connection was quite small, measuring only 12 mm in diameter. The color flow signal (Fig. 41.28, top right) showed aliasing at the distal anastomosis. Continuous-wave Doppler revealed reduced variability and an elevated mean gradient (4 mm Hg) across the connection (Fig. 41.28, bottom). The 22-mm conduit could not be identified. However, flow was detected in the area of the left pulmonary artery, just posterior and lateral to the ascending aorta (Fig. 41.29). Cardiac catheterization subsequently revealed that the left pulmonary artery was stenotic, but retained its connection to the right pulmonary artery and was of normal size distally. The right atrial–to–left pulmonary artery conduit was actually patent, but had an extremely narrow lumen (Fig. 41.30). This case dramatically illustrates the need to use multiple imaging modalities when evaluating a patient with a failing Fontan circulation. Revision of the patient’s Fontan connection was recommended. A stent was placed in the pulmonary arterial confluence intraoperatively, enlarging the connection between the right and left pulmonary artery. The Fontan connection was then revised to include an intraatrial conduit, an enlarged atrial–to–right pulmonary artery connection, and a small atrial fenestration. The images in Figure 41.31 were taken from the postoperative transesophageal echocardiogram performed during this revision. The left panel shows the stent enlarging the left pulmonary artery, and the right panel illustrates the associated improvement in the color flow Doppler pattern within the Fontan connection and proximal left pulmonary artery.
Figure 41.27. Case 4. “Four-chamber” views taken from the distal esophageal position during a transesophageal echocardiogram in a patient with tricuspid and pulmonary valve atresia. The atrial septation patch appears to adequately divide the right (RA) and left atrium (LA). All of the cardiac chambers were somewhat enlarged, and color flow Doppler revealed mild mitral regurgitation (right, white arrow). L, left; LV, left ventricle; S, superior.
Figure 41.28. Case 4. Sagittal-plane images (top) demonstrate the connection between the dilated right atrium and the pulmonary arterial confluence. The pulmonary arteries are not completely visualized in this plane, but the right atrial appendage “Fontan” connection is well visualized (RAA). The 1.0 × 1.5-cm diameter of this pathway is very small for this 31-year-old man. The color flow signal (top right) shows aliasing consistent with elevated flow velocity in this area. As expected, the continuous-wave Doppler signal (bottom) shows an abnormal flow pattern. The decreased velocity at either end of the tracing was caused by translational motion and does not reflect the true flow signal. When the Doppler beam was adequately aligned (center of the tracing), the flow velocities were elevated and did not reflect the atrial contraction and relaxation patterns expected in this atriopulmonary Fontan connection. The mean right atrial–to–pulmonary artery Doppler gradient was 4 mm Hg. This may seem like a “low” value, but in the unforgiving setting of a Fontan circulation, it represents an important obstruction. A, anterior; Ao, aorta; CW, continuous-wave; S, superior.
Figure 41.29. Case 4. Horizontal-plane image obtained just posterior to the ascending aorta (Ao). The color flow signaI outlines the entire diameter and length of the proximal left pulmonary artery (LPA). This vessel was only 5 mm in diameter, clearly too small to provide adequate flow in this adult male. The aliased flow pattern suggests that this stenosis/hypoplasia is causing an additional obstruction to transpulmonary flow (above and beyond that caused by the Fontan obstruction shown in Fig. 38.28). A, anterior; L, left.
Figure 41.30. Case 4. A: Angiogram shows an anteroposterior (AP) projection of a contrast injection made just proximal to the Fontan anastomosis (horizontal white arrow). The Fontan connection is indeed narrow as was seen on the transesophageal echocardiogram (Fig. 38.28). The branch pulmonary arteries bifurcate to the right and left of the Fontan connection. The right pulmonary artery is normal. The left pulmonary artery stenosis, seen in Figure 38.29, can be appreciated here as well (red arrow). However, it is now clear that the distal left pulmonary artery is significantly larger and that relief of these narrowings may improve transpulmonary flow. B: Angiogram shows a lateral projection of an injection in the supplemental, 22-mm graft connecting the right atrium to pulmonary artery. This connection could not be detected by any echocardiographic examination and was thought to have been occluded. These injections showed that it was slightly patent (yellow arrow) but had narrowed significantly, measuring only 5 to 6 mm in diameter. L, left; P, posterior; PA, pulmonary artery; S, superior.
Figure 41.31. Case 4. Postoperative transesophageal echocardiogram performed during revision of this patient’s Fontan connection. After initiation of cardiopulmonary bypass, an intravascular stent was placed across the left pulmonary artery stenosis and dilated to a 15-mm diameter. An intraatrial conduit was placed to eliminate the dilated atrial chamber from the Fontan circulation and direct both vena caval flows to the enlarged Fontan connection. A: Much larger communication between the upper portion of the right atrial conduit (RA) and the pulmonary arterial confluence. The left pulmonary artery (LPA) stenosis, posterior to the aorta (Ao), has been completely eliminated. B: Color flow Doppler no longer shows any evidence of aliasing in these areas. A, anterior; L, left.
Understanding and evaluating patients with the Fontan circulation requires a thorough understanding of not only the surgical anatomy, but also the unique cardiovascular physiology of these patients. Unfortunately, these patients often have challenging acoustic windows. When surface echocardiography does not provide adequate detail, alternative imaging strategies should be used. The systematic echocardiographic approach described in this chapter allows not only a complete delineation of the cardiovascular anatomy but also the physiology associated with Fontan circulation. Serial, longitudinal follow-up studies using these techniques should increase the likelihood of early detection of late complications of the Fontan operation, leading, it is hoped, to improved therapeutic outcomes.
Bartz PJ, Driscoll DJ, Dearani JA, et al. Early and late results of the modified Fontan operation for heterotaxy syndrome: 30 years of experience in 142 patients. J Am Coll Cardiol. 2006;48:2301–2305.
Cetta F, Feldt RH, O’Leary PW, et al. Improved early morbidity and mortality after Fontan operation: the Mayo Clinic experience, 1987 to 1992. J Am Coll Cardiol. 1996;28:480–486.
Choussat A, Fontan F, Besse P, et al. Selection criteria for Fontan procedure. In: Anderson FH, Shinebourne EA, eds. Pediatric Cardiology. White Plains, NY: Churchill Livingstone; 1978:559–566.
Cook AC, Anderson RH. The functionally univentricular circulation: anatomic substrates as related to function. Cardiol Young. 2005;15:7–16.
Earing MG, Cetta F, Driscoll DJ, et al. Long-term results of the Fontan operation for double-inlet left ventricle. Am J Cardiol. 2005;96:291–298.
Fontan F, Baudet E. Surgical repair of tricuspid atresia. Thorax. 1971;26:240–248.
Fyfe DA, Kline CH, Sade RM, et al. Transesophageal echocardiography detects thrombus formation not identified by transthoracic echocardiography after the Fontan operation. J Am Coll Cardiol. 1991;18:1733–1737.
Hagler DJ, Seward JB, Tajik AJ, et al. Functional assessment of the Fontan operation: combined M-mode, two-dimensional and Doppler echocardiographic studies. J Am Coll Cardiol. 1984;4:756–764.
Harrison DA, Liu P, Walters JE, et al. Cardiopulmonary function in adult patients late after Fontan repair. J Am Coll Cardiol. 1995;26:1016–1021.
Huhta JC, Hagler DJ, Seward JB, et al. Two-dimensional echocardiographic assessment of dextrocardia: a segmental approach. J Am Cardiol. 1982;50:1351–1360.
Madan N, Robinson BW, Jacobs ML. Thrombosis in the proximal pulmonary artery stump in a Fontan patient. Heart. 2002;88:396.
McGoon DC, Danielson GK, Ritter DG, et al. Correction of the univentricular heart having two atrioventricular valves. J Thorac Cardiovasc Surg. 1977:74:218–226.
Milanesi O, Stellin G, Colan SD, et al. Systolic and diastolic performance late after the Fontan procedure for a single ventricle and comparison of those undergoing operation at <12 months of age and at >12 months of age. Am J Cardiol. 2002;89:276–280.
Olivier M, O’Leary PW, Pankratz VS, et al. Serial Doppler assessment of diastolic function before and after the Fontan operation. J Am Soc Echocardiogr. 2003;16:1136–1143.
Rosenthal DN, Friedman AH, Kleinman CS, et al. Thromboembolic complications after Fontan operations. Circulation. 1995;92(suppl.):II-287–II-293.
Silvilairat S, Cabalka AK, Cetta F, et al. Protein-losing enteropathy after the Fontan operation: associations and predictors of clinical outcome. Congenit Heart Dis. 2008;3:262–268.
Sluysmans T, Sanders SP, van der Velde M, et al. Natural history and patterns of recovery of contractile function in single left ventricle after Fontan operation. Circulation. 1992;86:1753–1761.
Van Praagh R, Vlad P. Dextrocardia, mesocardia and levocardia: The segmental approach to diagnosis in congenital heart disease. In: Keith JD, Rowe RD, Vlad P, eds. Heart Disease in Infancy and Childhood. 3rd ed. New York: Macmillan; 1978:638.
1.What are the goals of a Fontan operation?
A.To eliminate arrhythmias and ventricular outflow obstruction
B.To reduce myocardial oxygen demand and eliminate AV valve regurgitation
C.To minimize cyanosis and reduced ventricular workload
D.To improve atrial contractility and coronary blood flow
2.What is a Fontan operation?
A.A connection between the right atrial appendage and the pulmonary artery
B.A procedure that normalizes the circulation in patients with single ventricular physiology
C.A procedure that can only be performed in teenagers
D.Any procedure which diverts the inferior and superior systemic venous return directly into the pulmonary artery without the driving force of a ventricular pump
3.Atriopulmonary Fontan connections have a high incidence of late
C.onset heart block.
D.onset pulmonary arterial hypertension.
4.What does a fenestrated Fontan operation include?
A.A simultaneous repair of a fenestration in the atrioventricular valve
B.A simultaneous creation of a small “residual” atrial septal defect
C.A simultaneous creation of a small “residual” ventricular septal defect
D.None of the above
5.The mean Doppler gradient derived from the flow through a Fontan fenestration can be used to estimate:
A.The ventricular systolic pressure
B.The cardiac output
C.The systemic arterial resistance
D.The transpulmonary pressure gradient
6.Pulmonary arterial Doppler flow patterns in a patient with an extracardiac Fontan circulation will not have which of the following features?
A.Atrial reversal (post atrial contraction)
B.Early diastolic augmentation of forward flow
C.Late diastolic augmentation of forward flow (post atrial contraction)
D.Systolic augmentation of forward flow
7.In a Fontan patient, progressive dilation of the inferior vena cava is:
B.Associated with deteriorating function of the Fontan circulation
C.Seen only in those with extracardiac connections
D.Seen only in those with atriopulmonary connections
8.Evaluation of diastolic ventricular performance in Fontan patients:
A.Is not possible
B.Is not different from the evaluation of a biventricular circulation
C.Should focus on myocardial strain analysis
D.Is best understood by comparison of serial examinations
9.Which of the following parameters is associated with an increased mortality risk in Fontan patients with protein losing enteropathy?
A.Aortic valve area < 1.4 cm2
B.Reduced AV inflow deceleration time
C.Pulmonary arterial stenosis
D.Prolonged isovolumic relaxation time
10.Which of the following would not be a parameter associated with a successful Fontan circulation?
A.Normal ventricular diastolic function
B.Low pulmonary arterial resistance
1.Answer: C. By separating the systemic and pulmonary venous circulations, a Fontan operation will generally eliminate cyanosis, and decrease the excess volume work required of the single ventricle when the circulations are still mixing.
2.Answer: D. Fontan operations can be “created” in a large number of ways, not just at a RA to PA connection. However, the Fontan circulation is not normal and can be performed in young children (those over two years), as well as teens and select young adults.
3.Answer: A. Due to the combination of extensive atrial manipulation during surgery, elevated atrial pressures and chronic atrial enlargement, these patients have an ever increasing incidence of atrial arrhythmias. While ventricular arrhythmias and heart block can occur, they are not common in any Fontan patient group. Pulmonary hypertension is essentially incompatible with adequate function of a Fontan circulation and is exceedingly rare, and when present the degree of absolute pressure elevation is mild (although the clinical consequences may be devastating).
4.Answer: B. A fenestration is a surgically created defect in the Fontan septation that allows a small residual right-to-left shunt between the functional systemic and pulmonary venous atria. A shunt at this level provides a consistent source of preload to the systemic ventricle. Repairs of AV valve can be performed during the Fontan procedure but are additional procedures and are not a part of the fenestrated Fontan itself. Since there is only one functional ventricle in these patients, the ventricular septum rarely is involved in the procedure.
5.Answer: D. Because fenestration flow originates in the “Functional RA,” and that directly connects to the pulmonary artery, the pressure drop from RA to LA is the same as the difference in pressure between the PA and the pulmonary veins (in the absence of PA and Pulmonary vein stenosis). This pressure difference is directly related to pulmonary vascular resistance, which is a very important determinant of clinical success in these patients. Options A, B, and C are not as directly linked to fenestration flows and cannot be estimated by this Doppler signal.
6.Answer: C. The extracardiac Fontan excludes all contractile tissue from the systemic venous circulation and therefore it is not possible for atrial contraction to augment forward flow. Forward flow is augmented both in systole (AV valve anular descent) and in early diastole (AV valve opening) by events related to the systemic ventricle that improve forward flow in the pulmonary veins, and thus impact arterial flow. And in some situations, usually associated with marked left atrial pressure elevation or retrograde atrial activation (junctional rhythm), atrial reversal can actually be observed to be transmitted through the pulmonary veins into the arteries.
7.Answer: B. While all patients with a Fontan circulation will have some degree of IVC enlargement, there should not be a progressive increase over time. When this occurs, it is strongly associated with increasing systemic venous pressure. IVC enlargement is not a specific sign of a particular complication(s), but rather is an early warning marker that should lead to thorough re-evaluation of all aspects of the patient’s cardiovascular system. This type of change in the IVC occur in patients with any type of Fontan connection.
8.Answer: D. The unique physiology and anatomy of the functionally univentricular heart makes any assessment more challenging than those of biventricular circulations. However, careful serial studies have provided significant insights into diastolic performance in these patients. The variable anatomy/procedures seen in this group make serial assessments with comparison to a baseline the most easily interpreted. Diastolic strain analysis has not yet been validated in these patients.
9.Answer: B. While any abnormality can impair the performance of a Fontan circulation, the only late echocardiographic variables that have been associated with either development of PLE or mortality risk in those with PLE are reduced deceleration time and ejection fraction/systolic function.
10.Answer: D. Options A, B, C are all associated with an optimal Fontan circulation. Systemic hypertension increases the afterload on the functional single ventricle and can lead to increased wall thickness, increased mass, and decreased compliance.