Framing

Ventricular function is a predictor of outcome after cardiac surgery and a predictor of long-term outcome in patients with cardiovascular disease. Patients with compensated CHF may have severely decreased ejection fraction (EF) with minimal symptoms. Regional ventricular dysfunction most commonly is caused by myocardial ischemia or infarction. Hence there is an imperative to detect ventricular dysfunction and institute treatment in an attempt to prevent acute or long-term consequences.

Is ventricular function normal or abnormal? Is the abnormal function global or regional? What is the coronary distribution that relates to an SWMA? Is the ventricle big or small? Is the myocardium thinned or hypertrophied? Is the abnormal function new or old? Does the medical or surgical intervention improve or decrease ventricular function?

Data Collection

Left ventricular systolic function is assessed echocardiographically based on regional and global wall motion. Methods of assessment include changes in regional wall thickness, radial shortening with endocardial excursion, fractional area change (FAC), and systolic apical displacement of the mitral annulus. Off-line measurements of EF can be calculated using Simpson’s rule. However, the EF most commonly is estimated online from the four-chamber, two-chamber, and short-axis images of the LV. Other measures include end-diastolic area (EDA), end-systolic area (ESA), and, most recently, three-dimensional analysis.

Regional assessment provides an index of myocardial well-being that can be linked to coronary anatomy and blood flow. Although the measurement of coronary blood flow is not achieved by TEE, the perfusion beds and corresponding myocardium for the left anterior descending, left circumflex, and right coronary arteries are relatively distinct and can be scrutinized by TEE using multiplane imaging. The transgastric and long-axis imaging views of the LV are the most widely used for evaluating wall motion abnormalities. Digital archival systems have gained popularity for their ability to capture a single cardiac cycle that can then be examined more closely as a continuous cine loop. Cine loops also can permit side-by-side display of images obtained under varying conditions (e.g., prebypass and postbypass). Regional myocardial ischemia produces focal changes in the corresponding ventricular walls before changes occur on the ECG.¹⁶ Changes progress from normal wall motion to hypokinesis or akinesis. Dyskinesis, thinning, and calcification of the myocardium suggest a nonacute process, likely a prior infarction.

Assessment of right ventricular systolic function is more problematic because it is less quantifiable. The crescent-shaped right ventricle (RV) is not amenable to quantitative measures of differences in chamber size during the cardiac cycle. Characterization of the RV is accomplished by comparing the size of the right ventricular chamber with that of the LV and assessing the relative contractile function of the right ventricular free wall and that of the interventricular septum.

Ventricular failure can be caused by diastolic dysfunction: the compromised ability of the ventricle to accommodate diastolic filling. Diastolic dysfunction is assessed echocardiographically by examining volumetric filling of the ventricle at the mitral or tricuspid valves and annular excursion. Normal diastolic filling is biphasic with an early (passive) component that exceeds the late (active) inflow velocities. Abnormalities of ventricular filling (e.g., impaired ventricular relaxation, restriction, constriction) produce characteristic changes in the spectral recordings of Doppler inflow velocities. Abnormalities of diastolic function can lend insight into the mechanisms of circulatory instability (see Chapter 12).

Discussion

Preexisting ventricular dysfunction suggests increased risk for surgery and poorer long-term outcome. The presence of such ventricular dysfunction may deteriorate intraoperatively, requiring the need for marked pharmacologic or mechanical support. A patient with a preoperative EF of 10% scheduled for coronary artery bypass grafting (CABG) and MV repair is at increased risk for intraoperative ischemia, acute heart failure, and difficulty maintaining hemodynamic stability during the immediate postbypass period. Anticipating such problems, consider placement of an intra-aortic balloon pump or femoral arterial catheter during the prebypass period (Figure 13-2). The same patient is likely to benefit from the administration of inotropic agents (see Chapter 32).

Figure 13-2 The prebypass transesophageal echocardiography (TEE) examination may have predictive value for postbypass circulatory management.

A 63-year-old woman with a medical history of hypertension, congestive heart failure, pulmonary edema, dilated cardiomyopathy, diabetes, and obesity was scheduled for coronary artery bypass grafting (CABG) and mitral valve (MV) repair. The preoperative evaluation documented moderate-to-severe mitral regurgitation (MR) with reversal of systolic pulmonary vein blood flow velocity. The prebypass TEE midesophageal four-chamber view showed a markedly dilated left ventricle (LV) and mildly dilated right ventricle (RV) with mildly decreased global dysfunction (A). The transgastric view was characterized by severe global dysfunction and an LV end-diastolic diameter of 6.6 cm (A). The fractional area change (FAC) was 17% [FAC = (left ventricular end-diastolic area [LVEDA] – left ventricular end-systolic area [LVESA])/LVEDA × 100]. Revascularization alone was unlikely to significantly improve MV function. The midesophageal bicommissural view of the MV (B) demonstrated marked dilation of the MV annulus (major axis = 4.8 cm) and tethering of the leaflets below the valve plane that was caused by LV chamber dilation. A femoral arterial line was inserted for monitoring of central aortic pressure and/or possibly placing an intra-aortic balloon pump. The patient underwent a CABG × 3 and MV annuloplasty for moderate MR. The separation from bypass was difficult, requiring milrinone, epinephrine, vasopressin, and placement of an intra-aortic balloon pump. TEE, which was used to initially confirm the location of the femoral guidewire (C), was later used to position the balloon pump just downstream to the left subclavian artery. Worsening of RV function that was characterized by increased central venous pressure, new-onset tricuspid regurgitation, and a hypokinetic RV can be appreciated by ventricular septal flattening and dilation of the RV (D). The LV ejection fraction did not decrease as might be expected; after correcting MR, the FAC improved slightly from 17% to 22% after bypass. Cardiac function continued to improve, and the counterpulsation device was removed without complication on the first day after surgery. The infusions of milrinone and epinephrine were continued for several days.

A marked decrement or unexpected decrease in global cardiac function after release of the aortic cross-clamp can be caused by poor myocardial preservation during cross-clamping or distention of the heart during bypass. The risk for such incidents can be reduced by the monitoring of the electrical activity of the heart and pulmonary artery pressures, as well as for distention of the RV and LV. Effective venting of the heart is often difficult to discern by visual inspection alone, especially with the use of minimally invasive surgery through small incisions. TEE imaging can diagnose ventricular distention produced by AV insufficiency.

Not all preexisting SWMAs benefit from coronary revascularization. Regions of akinesia and dyskinesia usually are the result of a myocardial infarction and may reflect nonviable myocardium, although “hibernating” myocardium is possible. Hypokinetic segments generally are viable and may represent active ischemia.¹⁷ Preoperative positron emission tomographic scanning can detect hibernating myocardium and may be cost-effective to guide CABG.^18–20 The detection of hibernating myocardium in an area of chronic ischemia and regional hypokinesis will direct the surgeon to revascularize the corresponding stenosed coronary artery. In contrast, an occluded coronary artery with downstream infarction may not benefit from revascularization because contractile function may be irreversibly lost. However, in this latter scenario, revascularization postinfarction may provide some benefit in decreasing the risk for ventricular aneurysm formation.²¹

Diastolic dysfunction is associated with significant increases in mortality during long-term follow-up.²² Characterization of abnormalities of diastolic function lends insight into the mechanisms of circulatory instability and hypotension. Severe left ventricular hypertrophy with a noncompliant LV and hyperdynamic systolic function may produce severe heart failure if adequate loading is not achieved. Hemodynamic indices obtained from a pulmonary artery catheter (PAC) may be misleading. The findings of a small left ventricular chamber size, blunted transmitral filling velocities, and an increased FAC demonstrate the cause of the hypotension. The decision to administer volume may be appropriate despite the increased pulmonary artery pressures.

If the intraoperative examination reveals new ventricular dysfunction, the intraoperative team must determine the cause and severity and then plan a treatment. Other causes of SWMAs such as conduction abnormalities (left bundle branch block or ventricular pacing) can be difficult to distinguish. Is the decrement in function potentially reversible with conservative therapy, or should additional intervention be considered? Treatment of myocardial ischemia may include optimizing hemodynamics; administering anticoagulants, nitrates, calcium channel blockers, or β-blockers; inserting an intra-aortic balloon pump; or instituting CPB and coronary revascularization. The presence of new-onset SWMAs after separation from CPB is worrisome for myocardial ischemia. Even the patient without coronary artery disease (CAD) remains at risk because of hypotension, a shower of air or debris into the coronary circulation, or coronary spasm. The patient with CAD undergoing CABG may have all the above risks, technical difficulties at the anastomotic site, injury to the native coronary artery (e.g., stitch caught the back wall or occlusion of the circumflex artery during MV surgery), or occlusion of the coronary graft by thrombosis or aortic dissection. The coronary arteries, grafts, and anastomoses should be carefully inspected for patency and flow. Graft patency in the operating room is difficult to determine. Techniques include manual stripping and refill, measuring coronary flow by handheld Doppler, or administration of echo contrast agents (see Chapter 12). Hybrid operating rooms have been increasing in number with the intent of providing advanced imaging of the coronary circulation at the time of surgery.²³ A new SWMA in the distribution of a new coronary graft can prompt the decision-making strategies listed in Table 13-1 (see Chapters 18, 32, and 34).

TABLE 13-1 Management Strategies for New-Onset Myocardial Ischemia after Bypass

Framing

There are many instances during the perioperative period when the patient may exhibit progressive, unremitting hemodynamic deterioration or acute cardiovascular collapse. Echocardiography offers a versatile modality to quickly and accurately diagnose the cause of hypotension and develop management strategies.

The echocardiographer may be summoned to evaluate an unstable patient in the operating room, intensive care unit, or emergency department with little or no preceding knowledge of the patient. Typically, there will be no consent for the TEE procedure; occasionally, a family member may be available to provide consent on the patient’s behalf. TEE may need to be postponed in the trauma patient with suspected cervical spine injury or esophageal injury. The trauma patient with an unstable cervical spine is at increased risk for spinal cord injury with passive movement of the head and neck. Until the cervical spine has been documented to be stable, TEE should be avoided and TTE is the alternative. The risk for further esophageal injury in patients with a penetrating trauma poses an additional challenge. Esophagoscopy before TEE is performed in patients with suspected esophageal injury. However, delay in diagnosis is not without cost. Time must be used efficiently because permanent vital organ injury relates to the magnitude and duration of hypotension and malperfusion. A number of issues should be considered to guide the discussion and development of rational management strategies.

What is the cause of the hypotension? Does the cardiac or vascular pathology detected by TEE explain the decrease in blood pressure? Is the heart big or small? Is it full or empty? What is the global function of both ventricles? Are there SWMAs? Is there fluid in the pericardium? Is the observed decrease in cardiac function the primary cause, or is it a consequence of the decreased blood pressure? Is this event related to the patient’s medical history or current operative procedure? What specific parameters of the ventricle may help explain the current episode of hypotension? What interventions or therapy can be performed to improve hemodynamics? Once therapy is initiated, what index or parameter should be monitored to guide management?

Data Collection

Clinicians must not underestimate the importance of medical history, chief complaint, and operative course when attempting to discern the causes of hemodynamic instability in the acute setting. Important hemodynamic indices include heart rhythm, rate, blood pressure, concentration of exhaled carbon dioxide, and central venous or pulmonary artery pressure, if available. Echocardiography can be used to develop a rational approach based on the critical factors of cardiac performance. The determinants of cardiac performance include stroke volume (SV) and heart rate (HR), as elucidated by the following equation: CO = SV × HR, where CO represents cardiac output. The three components of SV that can be affected are preload, afterload, and myocardial contractility. Although quantitative analysis is possible, qualitative online analysis generally will yield sufficient information to form the basis of initial therapeutic intervention. Preload of the ventricle can be determined by assessing EDA, afterload can be estimated by assessing the ESA, and contractility can be estimated by the velocity of circumferential shortening, FAC, or EF. Mechanical causes of hypotension must be considered (e.g., pericardial effusion).

Discussion

Initial inspection determines heart size and overall contractile function of both ventricles. Estimates of EDA and EF of the LV provide an index of ventricular load and global function. Attention to both right and left ventricular size and function helps distinguish between different inciting events.

The common causes of intraoperative or perioperative hypotension include intravascular hypovolemia, myocardial ischemia, myocardial infarction, and systemic vasodilatation, either pathologic from infection or inflammation, or iatrogenic from drug administration (e.g., vancomycin). Mechanical causes of hypotension typically are related to compressive forces impairing the heart’s ability to fill or eject (e.g., pericardial fluid, tension pneumothorax). The MV is inspected for incompetence. Acute mitral regurgitation is rare in the absence of myocardial ischemia or infarction. A diagnosis of dynamic LVOT obstruction, an uncommon cause during the perioperative period, is difficult to establish in the absence of more invasive monitoring such as TEE (Figure 13-3). Systemic hypotension with a dilated RV and a small, underfilled LV implies either primary right ventricular failure (e.g., myocardial ischemia or infarction in the distribution of the right coronary artery) or secondary right ventricular failure from acute increases in pulmonary vascular resistance (e.g., pulmonary embolus [Figure 13-4], pneumothorax, or protamine reaction).

Figure 13-3 Acute intraoperative hemodynamic deterioration.

A 65-year-old man with a medical history of hypertension, sleep apnea, and smoking was scheduled for resection of colon cancer. During the bowel resection, the patient had a profound episode of hypotension and new-onset hypoxia. Although the patient’s hemodynamic condition initially improved after administration of phenylephrine and ephedrine, the patient became more hypotensive and hypoxic with evidence of pulmonary edema. Transesophageal echocardiography (TEE) was requested and was emergently placed to diagnose the cause of cardiovascular collapse and guide management. The midesophageal four-chamber view showed a moderately hypertrophied left ventricle (A). Left ventricular (LV) systole was associated with displacement of the mitral leaflet and chordae down into the outflow tract. The resulting defect in leaflet coaptation of the mitral valve (MV) and abnormal chordal position that was noted in A were associated with overwhelming mitral regurgitation (MR) and LV outflow tract obstruction. B, Administration of inotropes (epinephrine and ephedrine) was discontinued; the management strategy was changed to volume resuscitation and pressor administration. The hemodynamics normalized and the examination was repeated 10 minutes later (C, D). Displacement of the mitral leaflet and chordae had resolved together with the findings of MR and outflow tract obstruction. The patient was believed to have experienced a profound acute decline in the systemic vascular resistance in response to the release of vasoactive substances during bowel manipulation. The initial exacerbation of the hemodynamics was created by the relative hypovolemia and use of inotrope, leading to dynamic LV outflow tract obstruction. The concomitant episodes of hypoxia and pulmonary edema, which were attributed to the severe MR and increased left atrial pressures, resolved with the decrease in MR. The application of TEE was critical in making the correct diagnosis, altering management strategies, and initiating the appropriate therapy.

Figure 13-4 Progressive hypoxia and hypotension after cardiac surgery.

A 58-year-old morbidly obese patient with a medical history of hypertension, diabetes, and smoking had recently undergone coronary artery bypass grafting × 3. Five days after surgery, the patient experienced development of new-onset atrial fibrillation together with progressive hypoxia and hypotension requiring readmission to the intensive care unit. The patient’s condition continued to deteriorate and required tracheal intubation, ventilator support, and infusions of vasoactive agents. Transesophageal echocardiography (TEE) was performed to evaluate cardiac function and rule out a pericardial effusion. The midesophageal four-chamber view (A) showed a dilated right ventricle (RV) coincident with a relatively underfilled left ventricle (LV) and abnormal positioning of the ventricular septum. The displacement of the septum into the LV was consistent with RV dysfunction and RV volume overload. Inspection of the right heart revealed a dilated hypokinetic RV and a serpiginous density that extended from the right atrium into the RV (B). The thrombus appeared to be entangled in the chordal structure of the tricuspid valve. Although no thrombus was noted in the pulmonary arteries, the diagnosis of pulmonary embolism was made and the patient underwent an emergent embolectomy. A right atriotomy was performed and a 64-cm thrombus was extracted from the RV together with additional clots from both the right and left pulmonary arteries (C). The postbypass TEE documented improved right ventricular function and filling of the LV. TEE was critical in making the diagnosis and the decisions to institute and guide therapy.

Decreased systemic vascular resistance during sepsis or a systemic inflammatory reaction is associated with decreased ESA and increased ventricular contractility (increased velocity of circumferential shortening), with concomitant increases in the EF and FAC. The increase in cardiac performance can be quantified by measuring the cardiac output (CO). In cases of hypotension associated with a markedly increased CO, the treatment of choice would be the administration of a vasopressor, such as phenylephrine or vasopressin. If decreased systemic vascular resistance and CO are present, administration of a positive inotrope having vasopressor actions, such as epinephrine or norepinephrine, might be more appropriate.

The distribution of the right coronary arterial system of most patients (right dominant system) includes the RV and the posterior descending coronary artery, which provides blood supply to the inferior and inferoseptal walls of the LV. Acute right ventricular dysfunction is not uncommon after the release of the aortic cross-clamp. Preservation of the RV is less reliable compared with the LV because of its exposure to ambient room temperature and variability in its coronary circulation. Open-chamber procedures increase the risk for right ventricular dysfunction because of retained intracardiac air. In the supine patient, the right coronary ostium is located in the least-dependent portion of the aortic root, predisposing it for the embolization of air bubbles. Air embolization to the right coronary artery produces acute ST-segment changes, marked global right ventricular dysfunction, and SWMAs of the inferior wall of the LV (Figure 13-5). Conservative treatment includes increasing the blood pressure to promote coronary perfusion while continuing CPB.

Figure 13-5 Air embolism after open-chamber procedure.

A 57-year-old patient underwent surgical intervention to treat severe mitral regurgitation (MR) and two-vessel coronary artery disease. After performing the mitral valve (MV) repair, the right superior pulmonary vein vent catheter was advanced through the mitral valve to facilitate deairing. The patient was positioned in Trendelenburg and the aortic cross-clamp removed. After ventricular ejection, the left atrium and ventricle (LV) were relatively clear of air and the left atrial vent was removed. An aortic vent that had previously served as the cardioplegia cannula was used to vent the residual air from the ascending aorta. The patient was separated from bypass in normal sinus rhythm and maintained good hemodynamics. The initial postbypass transesophageal echocardiogram (TEE) showed normal ventricular function and filling (A). Shortly after starting administration of protamine, the blood pressure decreased and the electrocardiogram demonstrated ST-segment changes consistent with ischemia. The protamine administration was stopped, and vasopressors and inotropes were quickly administered. The TEE documented reduced cardiac function in the right ventricle and hypokinesis in the inferior and inferoseptal walls of the LV. The transgastric short- and long-axis views of the LV (B, C) showed that the myocardium in the distribution of the right coronary artery, as designated by the arrows, was characterized by increased echogenicity. Note the significantly elevated ST segments observed in B and C compared with the baseline (A). The absence of pulmonary hypertension and adequate diastolic filling of the LV supported a diagnosis other than acute anaphylactic protamine response or pulmonary embolism. The most likely culprit was air embolization causing transient myocardial ischemia. Air bubbles that migrated from the LV chamber embolized into the ostium of the right coronary artery, which lies at the most anterior aspect of the sinus of Valsalva. TEE served a crucial role in quickly evaluating and diagnosing the cause of hypotension. The hemodynamics, which were temporarily supported by boluses of vasopressors and inotropes, stabilized, and the protamine dose was completed. The patient was transferred to the intensive care unit without any further incident.

Framing

Small effusions are common after cardiac surgery, especially after removal of chest tubes. In isolation, pericardial effusion may not require surgical intervention. Cardiac tamponade occurs when the pressure exerted by the presence of a pericardial effusion (or any structure adjacent to the heart) compresses the heart, limits diastolic filling of any of the chambers of the heart, and impairs CO. Cardiac tamponade is an emergency, necessitating immediate diagnosis and intervention.

Is a pericardial effusion present? If so, does it contribute to cardiac dysfunction and the patient’s present hemodynamic distress? What is the cause of the effusion (acute or chronic)? Is the pericardial effusion the sequelae of another cardiovascular event or process (i.e., aortic dissection or cardiac catheterization)? Are there invaginations of the free walls of the right atrium (RA), RV, or LA? Is the effusion free flowing or loculated? Where is it located? Is echocardiographic assessment consistent with tamponade? What is the coagulation status?

Data Collection

Echocardiography is the standard modality for diagnosis of pericardial fluid. However, the diagnosis of cardiac tamponade is a clinical diagnosis based on hemodynamics and the patient’s condition. Low CO, hypotension, equalization of pressures, and high venous pressures are all signs of cardiac tamponade. Echo findings consistent with tamponade include presence of pericardial fluid, compression of the atria, compression of the RV, and loss of normal respiratory variability of ventricular inflow velocities.

The TEE examination quickly determines whether pericardial effusion is present, the location of the effusion (loculated, free-flowing), and impact on chamber filling. Location of the effusion is paramount should pericardiocentesis be deemed necessary for acute decompression. Right atrial and right ventricular collapse are the most sensitive signs of increased pericardial pressure. The effusion need not be a large circumferential effusion to significantly impact cardiac function. Postcardiotomy pericardial clot may be smaller and more compartmentalized than a chronic circumferential effusion that may be free flowing. Interpreting the clinical significance on cardiac hemodynamics may be complicated by factors such as lability of hemodynamics, decreased intravascular volume, depressed cardiac function, mechanical ventilation and pulmonary dysfunction, soft-tissue changes, and chest tubes that obstruct some of the echocardiographic windows. Doppler is used as a complementary method of demonstrating the hemodynamic derangements of tamponade and to determine the clinical significance of an effusion. The echocardiographer should interrogate the phasic respiratory variation of blood flow through the tricuspid valve and MV. Although not specific for tamponade, the changes in respiratory variation of inflow velocity are the hallmark of increased pericardial pressure. Significant respiratory variation of blood inflow velocities also may be seen in constrictive pericarditis or conditions associated with changes in intrathoracic pressure, such as increased work of breathing, asthma, or positive-pressure ventilation, and may be exacerbated in patients on positive end-expiratory pressure. Other important data pertaining to the cause and possible intervention include coagulation status.

Discussion

The American College of Cardiology/American Heart Association/American Society of Echocardiography (ACC/AHA/ASE) Task Force assigned a Class I recommendation to the use of echocardiography in patients with suspected bleeding in the pericardial space. Echocardiography is portable, quick, and noninvasive, yet it is a sensitive and specific modality for the detection and impact of a pericardial effusion. Pericardial effusions can be diagnosed and cardiovascular effects determined by TTE or TEE. However, during the postcardiac surgery period, the presence of positive-pressure ventilation, chest tubes, and bandages may severely limit the capability of TTE to assess fluid in the pericardium.

The effusion need not be a large circumferential effusion to significantly affect heart function. Loculated effusion may impinge only on the LA and may not be discernible by the traditional acoustic windows used by TTE. A hemodynamically significant localized hematoma compressing only the LA may not produce right atrial and right ventricular collapse, or the constellation of equalization of pressures. Small effusions are common after cardiac surgery, especially after removal of chest tubes, and in heart transplant recipients in whom there is a mismatch between heart size and pericardial cradle. The presence of a pericardial effusion in the nonpostcardiotomy patient must lead to a search for the cause of the effusion. Pericardial effusion mandates close scrutiny of the aortic root for possible aortic dissection. Pericardial effusion in a trauma patient is worrisome for cardiac rupture, ventricular contusion, or foreign body injury.

Acute cardiac tamponade in the nonpostcardiotomy patient can develop after introduction of as little as 60 to 100 mL blood. Causes might include type A aortic dissection, myocardial infarction with rupture, acute pericarditis, bleeding from malignancy, myocardial contusion, or myocardial perforation from penetrating trauma. These life-threatening conditions may present with hypotension, tachycardia, plethora, and jugular venous distention. Other classic findings include narrowed pulse pressure, pulsus paradoxus, widening of the mediastinum on chest radiography, and electrical alternans on the ECG. Treatment is immediate decrease in pericardial pressure that could be accomplished through the removal of a relatively small volume of fluid. This temporizing measure can be life-saving until more definitive therapy is instituted.

Not all pericardial effusions require immediate intervention. Development of cardiac tamponade is related to the rate of accumulation of pericardial fluid and the capacity for the pericardium to stretch and accommodate fluid. Chronic pericardial effusions, which occur in cases of malignancy, uremia, connective tissue disease, Dressler syndrome, and postinfection pericarditis, uncommonly require emergent intervention. Acute pericardial effusions that occur postcardiotomy are usually more ominous and often result in hemodynamic compromise, requiring treatment (see Chapters 22 and 34).

Hemodynamics may improve temporarily with the administration of volume, altering intrathoracic pressure (decreasing peak inflation pressure), but still may require drainage of the effusion. Chronic malignant effusions will improve after pericardiocentesis but often require a pericardial window for more definitive therapy. Effusions resulting from acute aortic syndromes or cardiac trauma require timely surgical intervention. Postcardiac surgery patients may require urgent re-exploration for evacuation of pericardial hematoma and to address the cause of continued bleeding. If hemodynamics improve after sternotomy but minimal clot is found, the physiologic tamponade may be related to generalized tissue edema and pulmonary dysfunction. In cases of poor cardiac function, the sternal incision may need to remain open and covered with a sterile dressing until edema recedes and cardiac function improves.

Framing

Ischemic heart disease is the most common cause of mitral insufficiency in the United States. Mechanisms of valve incompetence are varied and include annular dilatation, papillary muscle dysfunction from active ischemia or infarction, papillary muscle rupture, or ventricular remodeling from scar, often leading to a tethering effect of the subvalvular apparatus. Mitral regurgitation leads to pulmonary hypertension, pulmonary vascular congestion, and pulmonary edema with functional disability. Ventricular function deteriorates as the LV becomes volume overloaded with corresponding chamber dilatation. Left untreated, severe mitral regurgitation from ischemic heart disease has a poor prognosis, hence the imperative for diagnosis and treatment.^24–26 Less certain is the impact of lesser degrees of mitral insufficiency on functional status and long-term morbidity and mortality. Patients presenting for CABG often have concomitant mitral regurgitation of a mild or moderate degree. The intraoperative team is confronted with the decision whether to surgically address the MV during the coronary operation.

Does mitral regurgitation warrant mitral surgery? What is the mechanism of the regurgitation? What are the grade and chronicity of the mitral regurgitation? Is the mitral regurgitation likely to improve by coronary revascularization alone?

Data Collection

Pertinent data, including preoperative functional status and evaluation, need to be considered to appropriately interpret and place the intraoperative data in context. The preoperative echocardiogram and ventriculogram need to be reviewed. The intraoperative hemodynamic data are coupled with TEE information to complete the dataset needed to move forward with the decision-making process. The severity of mitral regurgitation on TEE is measured by the vena contracta, maximum area of the regurgitant jet, regurgitant orifice area, and pulmonary vein blood-flow velocities. Valvular disease causes changes in other cardiac structures. Chronic mitral regurgitation may be associated with a dilated LA, pulmonary hypertension, and right ventricular dysfunction. Wall motion assessment and the ECG are used for detecting reversible myocardial dysfunction that may benefit from revascularization. The hemodynamic and TEE data are coupled with provocative testing of the MV in an attempt to emulate the working conditions of the MV in an awake, unanesthetized state. It is not uncommon that preoperative mild-to-moderate mitral regurgitation with a structurally normal valve totally resolves under the unloading conditions of general anesthesia.^27–29

Discussion

Most cases of ischemic mitral regurgitation are categorized as “functional” rather than structural. In a study of 482 patients with ischemic mitral regurgitation, 76% had functional ischemic mitral regurgitation, compared with 24% having significant papillary muscle dysfunction.³⁰ The mechanism of ischemic mitral regurgitation is attributed to annular dilatation, secondary to left ventricular enlargement and regional left ventricular remodeling with papillary muscle displacement, causing apical tethering and restricted systolic leaflet motion.³¹ The importance of local left ventricular remodeling with papillary muscle displacement as a mechanism for ischemic mitral regurgitation has been reproduced in an animal model.³²

The mitral regurgitation is prioritized in accordance with the principal diagnosis (e.g., CAD), comorbidities, functional disability, and short- and long-term outcome. Ischemic mitral regurgitation is quantified and the mechanism of valve dysfunction is defined. Intraoperative mitral regurgitation is compared with preoperative findings. Discrepancies between the preoperative and intraoperative assessment of the valve may reflect the pressure and volume unloading effects of general anesthesia. In patients with functional ischemic 1 to 2+ mitral regurgitation, the MV often is not repaired or replaced. However, the need for surgical intervention in patients with 2+ mitral regurgitation under anesthesia remains a point of debate and has not been definitively answered by prospective studies. MV surgery typically is recommended to improve functional status and long-term outcome for patients with 3+ ischemic mitral regurgitation or greater.³⁰ Ignoring significant ischemic mitral regurgitation at the time of CABG can limit the functional benefit derived from surgery.

The risks to the patient of not surgically altering the MV and anticipated residual regurgitation are weighted against the risk for atriotomy, mitral surgery, extending CPB and aortic cross-clamp times, and the likelihood that the CABG surgery will be successful at decreasing the severity of mitral regurgitation. Added risk includes commitment to a mechanical prosthesis should a reparative procedure prove unsuccessful. Mitral regurgitation caused by acute ischemia may resolve after restoration of coronary blood flow (Figure 13-6). The reversibility of the regurgitation is difficult to predict: Factors supporting reversibility (and, hence, no immediate need to surgically address the valve) include a structurally normal MV, normal left atrial and left ventricular dimensions, including the mitral annulus, and SWMAs associated with transient regurgitation and pulmonary edema. Revascularization of the culprit myocardium with improvement in regional function may be all that is necessary to restore normal mitral coaptation.^33,34 Myocardial infarction with a fixed wall-motion defect or aneurysm, chronically dilated left-sided heart chambers, dilated annulus, or other structural abnormalities that are not reversible (ruptured papillary muscle or chordae, leaflet prolapse, leaflet perforation) suggest myocardial revascularization is unlikely to correct the valvular incompetence.

Figure 13-6 Evaluation of mitral regurgitation (MR) in a patient undergoing coronary artery bypass grafting.

A 63-year-old man was scheduled to undergo off-pump coronary artery revascularization. The patient had a history of progressive congestive heart failure without evidence of acute pulmonary edema. The physical examination was significant for diffuse laterally displaced point of maximum impulse (PMI) and a systolic murmur at the apex that radiated to the axilla. The patient received an intraoperative transesophageal echocardiography (TEE) examination to evaluate the severity of MR. The left ventricle (LV) was significantly dilated with an LV end-diastolic dimension of 7 cm (A) and had depressed systolic function with an estimated ejection fraction of 40%. The MR was characterized by color-flow Doppler imaging to be a central jet of mild-to-moderate severity. The grading of MR was based on the area of the regurgitant jet and the vena contracta (B) viewed in a bicommissural view. The pathogenesis of MR was believed to be functional and resulted from restricted leaflet mobility caused by the dilated LV. The coaptation of the anterior and posterior leaflets was below the valve plane (C). The absence of reversal of pulmonary vein blood flow measured in the left lower pulmonary vein (C) supported the assessment of moderate MR. Because the annulus was not significantly dilated (the minor axis measured 2.97 cm) and the MR graded as only mild to moderate, the surgeon proceeded with his initial plan of off-pump coronary artery bypass grafting. The MR decreased immediately after revascularization, and the patient’s symptoms were expected to further improve with afterload reduction.

The decision whether to proceed with mitral surgery in the setting of ischemic heart disease is institution and surgeon dependent. Centers may elect to surgically address any degree of mitral regurgitation detected during the preoperative or intraoperative work-up of a patient scheduled for CABG surgery. Less aggressive sites elect to proceed with coronary revascularization, followed by repeat scrutiny of the ventricular wall motion and MV. If revascularization has not corrected the mitral regurgitation, the surgeon proceeds with CPB and mitral surgery. With the advent of off-pump coronary artery bypass surgery, this process has gained another level of complexity because decisions to proceed with mitral repair will commit the patient to CPB. Off-pump mitral surgical procedures may be possible in the near future. A device that decreases the minor annular axis by adjusting the length of an artificial subvalvular chord that spans the ventricle between anterior and posterior epicardial pads has been used to treat functional mitral regurgitation.^35,36 Likewise, percutaneous endovascular procedures to “clip” the free edges of the anterior and posterior mitral leaflets akin to the Alfieri surgical technique are aimed at noninvasively improving leaflet coaptation.^37,38 Short- and long-term outcome data regarding such investigational techniques remain forthcoming (see Chapters 3, 18, 19, 26, and 27).

Framing

Myxomatous degeneration of the MV is a common cause of mitral regurgitation. Patients often are young and otherwise quite healthy. The diagnosis is commonly established before surgery, and patients are scheduled for elective surgery unless acute leaflet or chordal rupture leads to acute pulmonary edema and emergency surgery. The decisions to be addressed in this setting include: Can the surgeon perform the MV repair to address the mitral regurgitation, or is valve replacement necessary? What does the surgeon need to know to assess the possibility of repair and how to accomplish it? What are the possible complications of MV repair? Is the MV repair acceptable?

Data Collection

The surgeon cannot rely on visual inspection of the native MV in the flaccid arrested heart to discern the mechanism of mitral regurgitation. Unless the patient had a preoperative TEE as part of the preoperative evaluation, the intraoperative study often is critical in assisting the surgeon to plan the appropriate surgical correction.

The intraoperative TEE examination targets the mitral regurgitation, LA, LV, and RV. Two-dimensional and color-flow Doppler remain the standard, although increasing application of three-dimensional echo has targeted mitral surgery as a niche. Three-dimensional echo has made significant advances in miniaturization, speed of data processing, and user interfaces. In the past, it was a technology seeking an application. Mitral surgery for myxomatous disease appears to be one of its prime applications. Myxomatous mitral regurgitation typically is amenable to valve repair. Typical findings of myxomatous valves include excessive leaflet motion, redundant leaflet tissue, and a dilated mitral annulus. Leaflets commonly prolapse into a dilated LA, the degree of which is based on the chronicity of the illness. Chordal rupture is common and leads to flail leaflets and severe mitral regurgitation. The imperative is to be exact in the descriptive anatomy of the MV. The accepted nomenclature of the anatomy of the MV is stated in a consensus guideline.¹⁵ The locus of flailed scallops or leaflets and regurgitant orifice, the width of the anterior and posterior leaflets, the bisecting widths of the mitral annulus (minor and major axes), the locus of a perforation, severity of annular calcification (less common in isolated ischemic or myxomatous disease), and the size of the LA all contribute significantly to the planned repair. The findings of long and redundant anterior and posterior leaflets increase the risk for postoperative SAM of the MV. Maslow et al³⁹ examined the predictors of LVOT obstruction after MV repair. In this study of patients who were undergoing repair for myxomatous valve disease, 11 of 33 patients experienced development of SAM and outflow tract obstruction. The major predictive factors were smaller anteroposterior length ratio (annulus to coaptation; 0.99 vs. 1.95) and short distance from septum to MV coaptation point (2.53 vs. 3.00 cm). The surgeon may elect to perform a posterior sliding valvuloplasty to move the point of coaptation laterally, thereby decreasing the risk for outflow obstruction and mitral incompetence associated with SAM.

Although assessment of valvular incompetence is made and the need for repair/replacement generally is decided before surgery, the severity and pathogenesis of mitral regurgitation should again be determined intraoperatively. The severity of mitral regurgitation can be quite variable and depend on the hemodynamic loading conditions and its pathogenesis. Discrepancies between preoperative and intraoperative assessment of mitral regurgitation are less likely in patients with structural abnormalities (e.g., ruptured chordae or leaflet perforation) of their MVs, in contrast with patients with functional mitral regurgitation (e.g., ischemic mitral regurgitation).

The repaired MV is scrutinized closely, with the ventricular vent removed at the discontinuation of CPB under typical loading conditions. The systolic and diastolic functions of the valve are examined for residual regurgitation, stenosis, and the presence of outflow tract obstruction. The postbypass examination is critical for determining the acceptability of the repair or guiding the subsequent revision, should the initial repair be unacceptable. Systolic valve dysfunction after repair typically produces residual mitral regurgitation. Residual mitral regurgitation after mitral repair is not uncommon and may not necessitate revision if the grade is trace or mild. Diastolic mitral function is ascertained by Doppler flow measurements across the mitral orifice to provide assurance that the repaired valve has not been rendered stenotic. Peak transvalvular blood-flow velocities and pressure half-times are measured to calculate valve gradients and valve areas.

Discussion

There is a growing imperative to repair instead of replace myxomatous MVs. The intraoperative echocardiographer soon becomes familiar with the abilities and limitations of his or her surgical counterparts. Outcomes may be quite dependent on the ability of the individual surgeon, more so than in valve replacement surgery. Hence it may be prudent to track short-term (intraoperative) results of these operations because outcomes may be less defined by national databases and more so by individual provider. In general, the likelihood of a successful repair is based on the severity and extent of involvement of the mitral leaflets. Isolated prolapse of the middle scallop of the posterior mitral leaflet associated with eccentric mitral regurgitation that overrides an otherwise normal anterior leaflet is associated with a high success rate. However, cases of extensive leaflet degeneration with bileaflet prolapse, multiple chordal ruptures from both leaflets, leaflet destruction from preceding endocarditis, two or more regurgitant orifices, and extensive calcification are associated with a significantly lower success rate for repair.⁴⁰ Many patients leave the operating room with a prosthetic MVR with preservation of the subchordal apparatus with excellent long-term results. Retention of the subvalvular apparatus preserves longitudinal shortening of the LV and decreases the incidence of HF in the long term.^41,42 In a comparative study, the EF of patients without chordal transfer decreased 24% after surgery compared with patients having chordal transfer who maintained preoperative function.⁴²

Some degree of residual mitral regurgitation after repair is common. Most surgeons will not accept residual mitral regurgitation of 2+ or greater and will readdress the valve surgically. Assessment of valve function postbypass under unloaded conditions (decreased afterload with relative hypovolemia) may erroneously predict mitral regurgitation in the long term or under provoked conditions of exercise. Hypotension with decreased left ventricular contractility and chamber dilatation often responds to inotropic agents with improved coronary perfusion and decreased mitral regurgitation.

The mechanism of the failed repair is paramount to the decision process and treatment. Residual mild regurgitation associated with persistent leaflet prolapse may warrant further leaflet resection. Mild residual regurgitation in the setting of a “normal” MV may sway the surgeon to accept it. The presence of a central jet of mitral regurgitation may require resizing (smaller) or considering an edge-to-edge repair (i.e., Alfieri). The decision of re-repair or valve replacement often is a difficult one, weighing the balance between the risks for reoperation versus the risk for residual mitral regurgitation. The final decision of what is acceptable residual valvular regurgitation is patient specific (e.g., age, anticipated level of activity, ability to tolerate a return to CPB).

The finding of LVOT obstruction caused by SAM after mitral repair often is caused by the displacement of leaflet coaptation toward the septum, resulting in the anterior leaflet to paradoxically move into the LV, rather than toward the LA late in systole (Figure 13-7).⁴³ Displacing the anterior leaflet into the LVOT during ventricular ejection (Venturi effect) produces outflow tract obstruction, early closure of the AV, and mitral regurgitation. The incidence of postoperative SAM and the need to revise the surgical repair have decreased as understanding of the predisposing factors and management strategies has improved. SAM may be intermittent and dependent on loading conditions. If possible, patients should be examined after adequate volume resuscitation and with minimal inotropic support. Most cases of SAM resolve with conservative measures including β-blockade, vasoconstriction, and fluid administration. In a single-center retrospective study of 2076 patients who underwent MV repair, the incidence rate of intraoperative SAM was 8.4% (174 cases).⁴⁴ Revision of repair or valve replacement related to SAM during initial operation was undertaken in only two patients. However, in the case of persistent severe SAM with a high LV-to-aorta gradient, returning to CPB and revising the mitral repair (e.g., posterior sliding mitral valvuloplasty to “slide” the locus of coaptation laterally), performing an edge-to-edge repair or possibly replacing the MV should be considered. Patients who transiently demonstrate SAM or turbulence in the outflow tract in the operating room may be at increased risk during the immediate postoperative period (see Figure 13-3). An important role of the clinical echocardiographer is to recognize potentially important findings that may have an impact on subsequent patient care and long-term follow up (see Chapters 12, 19, and 22).

Figure 13-7 Complex mitral valve (MV) replacement/repair revision.

A 58-year-old woman with severe mitral regurgitation (MR) and a history of congestive heart failure, pulmonary edema, and hypertension was scheduled for surgical repair of mitral insufficiency. The prebypass transesophageal echocardiogram (TEE) characterized the MV as having severe MR with mildly thickened, myxomatous leaflets and several ruptured chordae. The markedly dilated annulus was consistent with the chronic disease process. Prolapse of the posterior leaflet resulted in the MR jet overriding the anterior leaflet (A). Although the distance from the septum to the coaptation point of the MV was greater than the value cited by Maslow as predictive of postbypass SAM, the ratio of the length of the anterior and posterior leaflets (0.89) suggested a risk for LV outflow obstruction (B).³⁹ The midesophageal long-axis view demonstrates several ruptured chordae (B) that resulted in MR having a vena contracta of 0.65 cm. Blunting of the systolic component of pulmonary vein blood flow velocity corroborated the diagnosis of significant MR (C). The surgeon performed a quadrangular resection of the posterior leaflet and secured the annulus with a No. 30 Physio ring. The postbypass imaging of the midesophageal long-axis view demonstrated a coaptation defect, designated by the arrow, and nonlaminar flow in the LV outflow tract (D). Shift of the coaptation point medially created laxity of the redundant chordae that were drawn into the outflow tract by systolic LV ejection (D). The outflow tract obstruction was characterized by a pressure gradient of 54 mm Hg as determined by continuous-wave Doppler through the outflow tract using the transgastric long-axis view (E). Neither the MR nor the outflow tract obstruction was effectively addressed by volume loading and decreasing the inotropic support. Circulatory bypass was reinstituted, and the surgeon revised the previous repair by further resecting the posterior leaflet and enlarging the annular ring to a No. 34. The patient was successfully separated from bypass using minimal support and had resolution of LV outflow tract obstruction, return of normal hemodynamics, and reduction of MR to trace severity (F).

Framing

The finding of a persistent left-sided superior vena cava (SVC) is not a common incidental finding, but its diagnosis has important implications for the conduct of circulatory management during CPB.

What echocardiographic findings suggest persistent left-sided SVC? What confirmatory test can be performed? Does the finding of an anomalous left-sided SVC have implications for the conduct of CPB?

Data Collection

The echocardiographer should suspect a persistent left-sided SVC if the coronary sinus is significantly dilated or if significant difficulty was encountered while attempting to place a pulmonary artery catheter (PAC). Because the differential diagnosis of a dilated coronary sinus includes pathology associated with increased right-sided pressures, confirmation should be obtained by injecting agitated saline contrast into an intravenous catheter in the left arm. In the case of a persistent left-sided SVC, opacification of the coronary sinus occurs before that of the RA or RV (Figure 13-8). Once the diagnosis is confirmed, the echocardiographer should look for other associated congenital anomalies, including atrial septal defects (ASDs) or an unroofed coronary sinus with communication between the coronary sinus and the floor of the LA.

Figure 13-8 Detection of occult persistent left superior vena cava alters circulatory management.

A 57-year-old patient was scheduled to undergo an aortic valve replacement for severe aortic stenosis. The intraoperative transesophageal echocardiography examination documented a bicuspid aortic valve (AV) with a dilated aortic root and ascending aorta, preserved left ventricular (LV) systolic function, moderate LV hypertrophy, and dilated coronary sinus (CS) (A). The diagnosis of persistent left superior vena cava was considered. An agitated mixture of saline, blood, and air was injected into the intravenous access in the left arm, and the diagnosis was confirmed when the coronary sinus was opacified before the right atrium (RA) and ventricle (RV). The diagnosis of a persistent left superior vena cava altered conduct of bypass by negating the standard use of retrograde cardioplegia and by relying solely on antegrade cardioplegia.

Discussion

Left-sided SVC is the consequence of arrested embryologic development that results in the left brachiocephalic vein emptying into the coronary sinus with subsequent flow into the right atrial return of blood. As a consequence, cannulation of the coronary sinus with administration of retrograde cardioplegia will be ineffective at providing cardiac protection during cardiac arrest. Once the diagnosis is confirmed, the surgeon is informed to modify the conduct of cardiac protection during CPB. If TEE is unable to confirm the diagnosis (absence of venous access in the left arm to inject contrast), the surgeon can confirm its presence by direct inspection.

Framing

Incidental detection of an ASD or a patent foramen ovale (PFO) is a common occurrence. The clinical implications of an intracardiac defect are shunting, stroke, headaches, pulmonary hypertension, right ventricular dysfunction, and paradoxic embolization. If transseptal flow is present, it is generally from left to right, because left atrial pressure is generally greater throughout the cardiac cycle. Bidirectional flow is possible with transient increases of right atrial pressure that can be observed during normal respiratory maneuvers (i.e., Valsalva, coughing, and physical straining).⁴⁵ Routine physical activities associated with a Valsalva maneuver include heavy physical activity, lifting heavy objects, defecation, or vigorous coughing.⁴⁶ Although uncommon, right-to-left shunting can produce episodes of relative hypoxia and paradoxic embolization. TEE is an extremely sensitive technique to detect an ASD or a PFO.⁴⁷ ASD or PFO is generally well tolerated and often remains asymptomatic into adulthood; hence their detection often is incidental during a routine intraoperative TEE examination.

What type of ASD is present: primum, secundum, sinus venosus, or PFO? What is the size of the defect? Does the defect produce transseptal blood flow? What is the shunt fraction: Q_s/Q_t? Does the patient have any symptoms such as CHF, increased shortness of breath, history of stroke or transient ischemic attack, or refractory hypoxia? What is the pulmonary artery pressure and right ventricular function? Does the patient have any associated congenital abnormalities, such as a cleft MV with a primum ASD? Does the surgery as initially planned require the use of CPB? If so, was single or bicaval cannulation planned? Does the interatrial defect need to be closed?

Data Collection

The application of two-dimensional echocardiography, color-flow Doppler imaging, and contrast administration in the midesophageal four-chamber and midesophageal bicaval views provide for high sensitivity of detection and diagnosis of ASDs and PFOs. The clinical implication of these findings is determined by the type of pathology. The PFO may be detected in about 25% of adults; it occurs when the secundum septum fails to close or is stretched open because of elevated pressures in the LA (Figure 13-9). Ostium secundums, which account for 70% of ASDs, are located in the area of the foramen ovale. The cause of this lesion is attributed to poor growth of the secundum septum or excessive absorption of the primum septum. MV prolapse is present in up to 70% of patients with this abnormality and may be related to a change in the left ventricular geometry resulting from right ventricular volume overload. The primum ASD develops when the septum fails to fuse with the endocardial cushion at the base of the interatrial septum. A primum ASD commonly is associated with cleft anterior leaflet of the MV and mitral regurgitation. The tricuspid valve also may be abnormal. The final type of ASD is a sinus venosus defect, which comprises only 10% of ASDs. It commonly is associated with abnormal insertion of the pulmonary veins into the RA or SVC.

Figure 13-9 Detection of a patent foramen ovale in a patient undergoing cardiac surgery.

A 68-year-old patient was undergoing two-vessel off-pump coronary artery bypass grafting (OPCAB). In addition to coronary artery disease, the patient had a medical history of hypertension, non–insulin-dependent diabetes, and a stroke of unknown cause. The intraoperative transesophageal echocardiographic (TEE) examination that included an inspection of the interatrial septum in the midesophageal four-chamber and bicaval imaging planes showed a new finding of an aneurysmal interatrial septum with a patent foramen ovale (PFO, arrow). Color-flow Doppler showed minimal shunt flow from the left atrium (LA) to the right atrium (RA). The presence of the PFO, which was initially diagnosed by color-flow Doppler imaging (A), was confirmed by the transient flow of injected contrast (arrow) into the LA (B), after it was injected into the venous circulation. A provocative maneuver such as Valsalva transiently increases RA pressure more than that of the LA, increasing the sensitivity of PFO detection.⁴⁷ Because of the patient’s history of a cryptogenic stroke, the surgical plan and circulatory management were altered. The patient was fully heparinized and circulation supported by an extracorporeal pump, while the CABG was performed and the PFO was closed.

Discussion

In general, patients benefit from ASD closure. Although often asymptomatic, ASDs may present with atrial arrhythmias, heart murmur, abnormal ECG, dyspnea, cerebrovascular injury or stroke, or migraine headaches. Medical management and surgical closure of secundum ASDs were compared in randomized trials; surgical closure was associated with significantly decreased morbidity and mortality.⁴⁸ However, there are little data concerning treatment strategies and outcomes for the occult ASD or PFO detected intraoperatively. In the absence of recognized consensus guidelines, the decision to proceed with definitive closure should be based on the following factors: history of neurologic event without a definite cause, recurrent stroke while receiving anticoagulation, significant shunting through the defect, previous episode of hypoxia that may be related to intracardiac shunting, right ventricular dysfunction, or previous paradoxic embolism. Detection of a primum or sinus venosus defect requires a more involved surgical procedure, and associated anomalies must be addressed. Alternatives to operative closure are increasing as transvenous percutaneous closure devices become increasingly applicable. Transvenous catheter–based closure of an ASD by interventional cardiologists can be performed only on a secundum ASD or PFO (Figure 13-10). Approximately 30% of secundum ASDs are amenable to percutaneous closure.⁴⁹ The technique optimally requires a defect with limited size and a rim of tissue surrounding the defect of at least 5 mm to prevent obstruction of the coronary sinus or impingement of the AV (see Chapters 3 and 20).

Figure 13-10 Transvenous approach to closure of atrial septal defects (ASDs).

Patent foreman ovale and secundum ASDs can be closed by deploying a transvenous closure device as an alternative to surgical intervention. A, Self-expanding Amplatzer occlusion device (AGA Medical Corporation, Plymouth, MN) that is made from wires tightly woven into two interconnected disks. The device is deployed from the femoral vein and positioned using transesophageal echocardiography (TEE) or intravascular ultrasound. B, Color-flow Doppler image from an intravascular ultrasound that shows the deployment catheter as it enters through the right atrium (RA) and spans the secundum ASD. C, Color-flow Doppler ultrasound image showing the Amplatzer (arrowhead) in position across the interatrial septum, preventing shunt flow across the ASD.

A PFO is the most common congenital finding and usually is asymptomatic. However, a number of studies have found an increased prevalence of PFO and atrial septal aneurysms in patients having cryptogenic strokes (i.e., no identified cardioembolic or large vessel source).50–56 Atrial septal aneurysm is a congenital outpouching of the interatrial septum at the fossa ovalis and is strongly associated with PFOs.^50,51,53 The occurrence of atrial septal aneurysms detected by echocardiography is about 10% in the general population and up to 28% in patients with a history of stroke or transient ischemic episode.^52–54,57 A meta-analysis of these case-control studies found that PFO, atrial septal aneurysms, or both were significantly associated with ischemic stroke in patients younger than 55 years.⁵⁸ However, the relation between septal pathology and neurologic events in patients older than 55 is less clear. The data on treating patients with atrial septal abnormalities for either primary or secondary prevention of stroke are limited. The Patent Foramen Ovale in Cryptogenic Stroke Study and a meta-analysis by Oregera found that warfarin treatment was superior to other antiplatelet therapy and comparable with surgical PFO closure for the prevention of recurrent cerebral events.⁵⁹ Chronic anticoagulation is associated with complications. The decision regarding closure of an incidental PFO considers the presence of right ventricular dysfunction that may increase the risk for right-to-left transatrial blood flow and cryptogenic stroke. Clinical practices typically are based on individual surgical preferences. A survey of cardiothoracic surgeons in the United States noted a high degree of variability in management of intraoperatively discovered PFO.⁶⁰ During planned on-pump CABG surgery, 27.9% of responders stated they always closed intraoperatively discovered PFOs, whereas 10.3% did not. Only 11% of surgeons converted a planned off-pump procedure to an on-pump procedure to close the defect, but the rate of closure increased to 96% if the patient had a history of possible paradoxic embolism. A single-institution retrospective study of 2277 patients from 1995 to 2006 confirmed that incidental detection of PFO is common, and that repair was not associated with a survival benefit. The authors found that propensity modeling of the data detected an increased incidence of postoperative stroke in patients having PFO repair. Several case reports describe postoperative complications related to hypoxia and neurologic events in patients with persistent PFOs.^60–63

The benefit of aggressive management of a PFO in the absence of other interatrial septal abnormalities is less clear than ASD closure and is more controversial. Sometimes surgical intervention requires a significant alteration of the surgical plan and, therefore, may significantly increase operative risk. In cases in which PFO detection occurs during off-pump coronary artery bypass, surgical intervention necessitates marked changes in the conduct of both the operation and circulatory management. The off-pump coronary artery bypass patient with a small PFO typically will remain untreated. In the absence of recognized consensus guidelines, the decision to proceed with definitive closure should be based on the following factors: history of neurologic event without a definite cause, recurrent stroke while receiving anticoagulation, significant shunting through the defect, previous episode of hypoxia that may be related to intracardiac shunting, right ventricular dysfunction, or previous paradoxical embolism.

Framing

A relatively common clinical scenario for the echocardiographer is to assess the significance of previously unrecognized AV pathology. This discussion has pertinence for the echocardiographer faced with the new diagnosis of a bicuspid valve, AS, or AR.

What are the symptoms that brought the patient to medical attention? What is the patient’s baseline function? What is the anatomy of the AV? What is the severity of AR or AS? How do the intraoperative findings of AV disease differ from the preoperative assessment? Would surgical repair or replacement of the AV benefit the patient’s short- or long-term outcome? What is the planned procedure, and how would the risks be changed if the procedure was altered to address the new finding? Does another healthcare provider need to be involved in the decision whether to surgically address the valve? Is the pathology of the AV significant enough to require surgical intervention at this time?

Data Collection and Characterization of the Aortic Valve

Multiplane TEE permits an accurate assessment of AV area, valvular pathology, severity of regurgitation and stenosis, and detection of secondary cardiac changes. In the case of AS, the severity of valvular dysfunction is determined by measuring the transvalvular pressure gradient, calculating the AV area using the continuity equation, and by planimetry of the AV systolic orifice. Planimetry of the AV orifice with TEE is more closely correlated with the catheterization-determined valve area (using the Gorlin formula) than the value derived from TTE (r = 0.91 vs. 0.84).⁶⁴ The severity of AR by TEE generally is graded with color-flow Doppler imaging with measurement of the width of the regurgitant jet relative to the width of the LVOT. TEE is sensitive to even the most trivial amount of AR. Jet areas measured by TEE tend to be larger, and their severity is graded as greater compared with AR assessed by TTE.⁶⁵ Determining the clinical significance of AR typically requires assessment of more than just regurgitant grade, although severe 4+ AR is never left unaddressed. A more challenging decision is whether to surgically address lesser degrees of AR in patients requiring cardiac surgery for another reason (see Chapters 3, 19, and 21).

The cause and extent of AV disease can be delineated best by TEE, as shown in Figure 13-11. The relatively high resolution of the AV and associated structures in the near field of the midesophageal short- and longitudinal-axis views permits an accurate assessment of the severity and mechanism of valvular disease. The aortic leaflets should be inspected in the midesophageal long-axis view for the presence of vegetations, perforation, restriction, thickening/calcification, malcoaptation, and leaflet prolapse. The presence of subvalvular disease, such as a discrete fibrous subaortic membrane, also can be reliably excluded. The ascending aorta from the valve to the right pulmonary artery also should be viewed in long axis. This view is usually optimal for examining associated pathology of the aortic root and ascending aorta (e.g., aortoannular ectasia, bicuspid valve, type A aortic dissection).

Figure 13-11 Anatomy of the aortic root.

This schematic figure of the aortic valve in long axis shows the components of the aortic root, which include sinotubular junction (STJ), sinus of Valsalva (SoV), and the annulus of the aortic valve (AV-An). AscAo, ascending aorta; LVOT, left ventricular outflow tract.

AS is caused by calcification of the AV and rheumatic heart disease. Bicuspid AVs are at greater risk compared with the general population. AS produces a systolic pressure gradient between the LV and aorta. Secondary findings are dependent on where the patient’s condition is along the natural course of the disease. Secondary findings often contribute to the decision-making process because they infer the effects or consequences of the disease. AS is commonly associated with left ventricular hypertrophy and abnormal filling of the LV. The diastolic function is often impaired due to a thickened, noncompliant LV. Hence, MV and pulmonary vein blood flow velocities would demonstrate a blunted passive filling phase of the ventricle. Systolic function often is normal or hyperdynamic. The left ventricular chamber size is normal or small. However, longstanding AS results in progressive ventricular systolic dysfunction and heart failure. The LV becomes dilated with compromised contractile function. As the ventricle fails, CO decreases with a resultant decrease in trans-AV pressure gradient. Hence, the pressure gradient across an AV may be misleading as a measure for severity of AS.

Bicuspid Aortic Valve

A bicuspid AV is at increased risk for degeneration and calcification, leading to AR and calcific AS. Bicuspid AVs predispose for ascending aortic and arch aneurysm formation and are associated with increased incidences of aortic coarctation and ASD. The diagnosis of bicuspid AV prompts the echocardiographer to search for other commonly associated findings. The presence of isolated asymptomatic bicuspid AV without AS or AR does not mandate corrective surgery.

Bicuspid AVs are common in the general population, with an estimated incidence rate between 0.9% and 2.25%. The natural history of bicuspid valves is variable. Patients can go late into adulthood without AV or aortic disease. Bicuspid AVs can function without major hemodynamic abnormality well into the seventh decade of life.⁶⁶ In contrast, a significant percentage of patients experience progressive dilatation of the aortic root, producing AR, or premature calcification of the valve, producing AS.^67–71 For patients in their fourth decade of life, the predominant lesion at surgery is AR. With increasing age, the predominant lesion associated with bicuspid AVs becomes AS.⁷² The architectural makeup of the wall of the aorta is abnormal in patients with a bicuspid AV, predisposing for aneurysm formation.^73–76 The decision-making process in regard to surgical correction of a bicuspid AV must account for the size of the aortic root and the likelihood the patient will return for aortic root surgery in the not-too-distant future. The rate of ascending aortic dilatation in patients with a bicuspid AV may be as high as 0.9 mm/year.⁷⁷ Even though patients with a bicuspid AV are at increased risk for cardiac surgery at a later time in life, surgical intervention is not recommended for a bicuspid AV in the absence of aortic root aneurysm, AS, or AR. A normally functioning bicuspid AV may have a longer duration than a bioprosthetic artificial valve. Some centers in North America have adopted a more aggressive stance toward repairing a mildly dilated ascending aorta (≥4-cm diameter) because of the increased risk for acute aortic dissection and ruptures.⁷⁸

The majority of young patients with a bicuspid AV have a regurgitant lesion. If found as an incidental finding, valvular dysfunction is usually minimal and patients do not have a significantly increased risk for heart failure and death. Although AVR is associated with a low risk for mortality, the lifelong cumulative risk for valve replacement is not insignificant. Mechanical heart valves are associated with low but significant rates of valve thrombosis, thromboembolism, and hemorrhagic complications from lifelong anticoagulation. Biologic prosthetic valves are limited by degeneration of the bioprosthetic material. The use of a homograft or pulmonary autograft provides alternatives for patients who do not want to be or are at high risk for anticoagulation. Bicuspid AVs that can be rendered competent and nonstenotic may be “repaired” in the setting of surgery for aortic dissection or aneurysm. Reconstruction of an isolated, severely regurgitant bicuspid AV has been attempted and described^79–82 but is not in the mainstream of cardiac surgery in the United States because of the risk for recurrent AR. Surgical advances of AV repair have raised the hope of increased valve longevity that rivals or exceeds that of a biologic prosthesis (see Chapter 19).

Natural Course of Aortic Stenosis

The natural course of AS in the adult begins with a prolonged asymptomatic period associated with minimal mortality. Progression of the disease is manifested by a reduction in the valve area and an increase in the transvalvular systolic pressure gradient. The progression is quite variable, exhibiting a decrease in effective valve area ranging from approximately 0.1 to 0.3 cm²/year. AV calcification, as depicted by echocardiography, has been suggested to be an independent predictor of outcome. Patients with no or mild valvular calcification, compared with those with moderate or severe calcification, had significantly increased rates of event-free survival at 1 and 4 years (92% vs. 60% and 75% vs. 20%, respectively).⁸³ Decisions regarding valve replacement for mild or moderate AV disease in the setting of cardiac surgery for another cause are complicated by the variability in the natural progression of the disease. The pathogenesis of AS is an active process having many similarities to the progression of atherosclerosis.⁸⁴ AV calcification is not a random degenerative process but an actively regulated disease associated with hypercholesterolemia, inflammation, and osteoblast activity. More aggressive medical control of these processes might be expected to have a positive impact on outcome by retarding the degenerative process. Statin therapy has been proposed to slow the progression of disease.⁸⁵ The definitive evidence that such therapies attenuate the progress of AV stenosis remains unclear but should be forthcoming. If the progression of AV stenosis can be attenuated, the need for AVRs would decrease significantly (see Chapter 19).

Assessment of Mild and Moderate Aortic Stenosis

The intraoperative management of mild-to-moderate AS at the time of cardiac surgery remains controversial. A patient arrives in the operating room scheduled for a CABG but is discovered also to have mild or moderate AS that was unappreciated before surgery. The operative team must decide whether to surgically address the AV. The ACC/AHA Task Force recommends valve replacement at the time of coronary surgery if the asymptomatic patient has severe AS but acknowledges there are limited data to support intervention in the case of mild or moderate AS. It is in this exact scenario that the rate of progression of AS is of value, but it is rarely obtainable. A rapidly calcifying valve in a young patient that is becoming rapidly stenotic would sway the operative team to perform an AVR. A combined double cardiac procedure (CABG/AVR) increases the initial perioperative risk, as well as those risks associated with long-term prosthetic valve implantation. A delay in AVR and commitment to a second cardiac operation in the future subjects the patient to the risk for a redo sternotomy in the setting of patent coronary grafts and its associated morbidities. If the AV is not operated on during the initial presentation for CABG, the development of symptomatic AS may be quite delayed or may not happen.

A review of 1,344,100 patients in the national database of the Society of Thoracic Surgeons having CABG, CABG/AVR, or AVR alone culminated in a decision paradigm recommendation.⁸⁶ The study assumed rates of AV disease progression (pressure gradient of 5 mm Hg/yr), valve-related morbidity, and age-adjusted mortality rates that were obtained from published reports. The authors proposed three factors in the consideration of CABG or AVR/CABG: age (life expectancy), peak pressure gradient, and rate of progression of the AS (if known). Because the latter is difficult to discern, the analysis assumed an average rate of disease progression and recommended patients should undergo AVR/CABG when the pressure gradient exceeds 30 mm Hg. The threshold (AS pressure gradient) to perform both procedures is increased for patients older than 70 years because the reduced life expectancy diminishes the likelihood that they will become symptomatic from the AV disease. Whether to perform a concomitant AVR at the time of revascularization was also addressed by Rahimtoola,⁸⁷ who advocated a less aggressive approach. One problem with both studies is that they analyzed the transvalvular pressure gradient, which may be a misleading measure of the degree of stenosis of the AV because its value is dependent on CO. A low CO and flow rate will produce a low transvalvular pressure gradient, even in the setting of a severely stenotic AV. However, in the setting of preserved ventricular systolic function and mild or moderate AS, a pressure gradient is a useful metric. The variable rate of disease progression and the controversy regarding the indications for “prophylactic” AVR preclude a simple algorithm for dealing with this patient cohort. Increased age, lack of symptoms, minimal left ventricular hypertrophy, a valve area suggesting milder disease, and a pressure gradient less than 30 mm Hg would sway the decision not to replace the AV. In an asymptomatic young patient, a severely calcified valve, bicuspid valve, and left ventricular hypertrophy in the setting of moderate stenosis, as well as a pressure gradient greater than 30 mm Hg would suggest that an AVR might be beneficial in the long term. The considerations regarding the progression of AS and need for a subsequent, highly invasive redo surgery on older and presumably more ill patients must be reconsidered in the advent of the emerging technologies of the transcatheter AVR. Initial results seem promising; there is no evidence of early stenosis or prosthetic valve dysfunction. It often is useful to include the patient’s primary cardiologist and family in the decision-making process.

Assessment of Low-Pressure Gradient Aortic Stenosis

Patients with left ventricular dysfunction and decreased CO in the setting of AS often present with only modest transvalvular pressure gradients(< 30 mm Hg). Distinguishing patients with a low CO and severe AS from patients with mild-to-moderate AS can be challenging (Figure 13-12). The standard for assessing severity of AS is AV area, typically calculated using either a continuity method or by planimetry. Patients with low-gradient AS with severe left ventricular dysfunction who received an AVR had improved survival and functional status compared with patients who did not have a valve replacement.^88–90

Figure 13-12 Low-pressure gradient severe aortic stenosis.

A 76-year-old cachectic man was scheduled to undergo corrective surgery for severe mitral regurgitation (MR) and possibly clinically significant aortic stenosis (AS). The midesophageal short-axis view of the aortic valve (AV) showed a highly calcified trileaflet valve with restricted mobility (A). The measurement of AV area, 1.13 cm², which was obtained by planimetry, was believed to underestimate the severity of AS because of the shadowing artifacts related to the severity of calcification. The transgastric long axis of the left ventricle was obtained (B), and the velocity profiles of blood flow within the left ventricular (LV) outflow tract and the AV were measured. Although the patient had a diagnosis of severe AS, the maximal and mean pressure gradients were 33 and 21 mm Hg, respectively (C). The area of the AV was calculated to be 0.83 cm² using the continuity equation. The LV function was characterized by a severe dilated cardiomyopathy with an ejection fraction of 8%, LV end-systolic dimension (LVESD) of 7 cm, and LV end-diastolic dimension (LVEDD) of 8 cm (D). The diagnosis of low-pressure gradient AS was considered, and infusions of epinephrine and milrinone were started. Cardiac performance improved from 2.4 to 4.5 L/min, and the pressured gradients increased to 60 mm Hg, peak, and 41 mm Hg, mean (C). Although the calculated valve area that was recorded under conditions of inotrope support slightly increased to 0.9 cm², transesophageal echocardiography (TEE) clarified that the marked increase in the pressure gradient was consistent with a diagnosis of low-gradient AS and confirmed the presence of cardiac reserve.

A low-pressure gradient related to left ventricular dysfunction may not open the AV to its maximum capacity. Dobutamine challenge in a patient with low-pressure gradient AS can be useful in establishing true AV area. The ability to distinguish between true AV stenosis and a state of “pseudostenosis” relies on characteristic changes in hemodynamic and structural measurements in response to the augmented CO. The test is not usually performed in the operative setting but rather as a preoperative evaluation. The increase in calculated AV area is related to the increase in the CO and is attributed to partial reversal of primary cardiac dysfunction.^91–94 If dobutamine improves CO and increases AV area, it is likely the baseline calculations overestimated the severity of the AS. The dobutamine challenge is conducted as follows: patients with low-gradient AS receive intravenous dobutamine at 5 μg/kg/min with stepwise increases in dose.⁹² Patients may exhibit a significant increase in AV area (0.8 to 1.1 cm²) and a decline in valve resistance after dobutamine challenge. Patients with fixed, high-grade AS would demonstrate no change in valve area and an increase in valve resistance. The 2003 ACC/AHA/ASE Task Force gave a Class IIb recommendation (usefulness/efficacy is less well-established by evidence/opinion) for the use of dobutamine echocardiography in the evaluation of patients with low-gradient AS and ventricular dysfunction.⁹⁵ In addition to its role in distinguishing between true stenosis and pseudostenosis, low-dose dobutamine echocardiography is helpful in risk-stratifying of patients with severe true AS. Patients with augmented contractile function after dobutamine administration have an improved outcome after surgery.^96,97

Aortic Regurgitation: Natural Course and Management

Many elderly patients undergoing cardiac surgery will have some degree of AR. In most cases, the patients are asymptomatic and the severity is graded as trace to mild. The presence of AR has implications regarding the conduct of circulatory management, administration of cardioplegia, management of hemodynamics, and possible alteration of the surgical plan. The presence of AR obligates the echocardiographer to monitor for distention of the LV during CPB and cardiac arrest, as well as during the administration of antegrade cardioplegia. On release of the aortic cross-clamp, the heart may not instantaneously revert to an organized rhythm, precluding ejection of LV blood and predisposing for ventricular distention. The latter is particularly problematic in the setting of AR. The decision whether to surgically address the valvular dysfunction should not be based on left ventricular distention during CPB but considered in the context of the dataset that includes the natural history of AR in adults, established guidelines, and published outcomes, in addition to individual patient variables as previously described.

Chronic AR generally evolves in a slow and insidious manner with a very low morbidity during a long asymptomatic phase. Some patients with mild AR may remain asymptomatic for decades. Others exhibit progressive worsening of the regurgitant lesion and develop left ventricular systolic dysfunction, leading eventually to heart failure. Evaluation of left ventricular size and function is important because of the poor correlation between symptoms and severity of cardiac pathology. The nature of the transition between the compensated and uncompensated periods is poorly understood. Clinicians should be reluctant to consider valve replacement in an asymptomatic patient with preserved left ventricular function (Figure 13-13). Early surgery exposes the patient to perioperative mortality and morbidity, as well as to the long-term complications of a prosthetic valve. Guidelines for AV surgery in patients with AR were published in 1998 by the ACC/AHA Task Force and in 2002 by the Working Group on Valvular Heart Disease of the European Society of Cardiology.^98,99 AVR is clearly indicated for symptomatic patients with chronic AR and left ventricular dysfunction. However, the decision to perform valve replacement is less apparent for asymptomatic patients. The published recommendations suggest serial examinations to detect the progression of left ventricular dysfunction. In a strategy similar to that applied for patients with low-gradient AS, the induction of stress exercise testing may be helpful for assessing functional capacity and timing of surgery. However, intraoperative detection of occult AR does not provide for the luxury of serial examinations.

Figure 13-13 Detection of occult aortic regurgitation (AR) during nonvalvular cardiac surgery.

A 66-year-old woman was scheduled to undergo coronary artery bypass grafting using extracorporeal circulatory support. An intraoperative transesophageal echocardiography (TEE) examination diagnosed presence of a central AR jet that extended halfway into the left ventricle (LV). The severity of AR, which was graded as mild, was based on assessing the retrograde flow in the LV outflow tract (LVOT) using color-flow Doppler imaging in midesophageal long-axis plane. The severity of AR is determined by comparing the width of the AR jet, 0.59 cm, with the diameter of the LVOT, 2.12 cm; the ratio of 0.28 is graded as mild (A). The anatomy of the valve was that of a normal trileaflet structure with mildly thickened leaflets (A, B). No leaflet prolapse, vegetations, or annular/root dilation was present. In addition, no significant aortic stenosis was detected by Doppler or calcification of the leaflets. LV chamber size was normal, LV end-diastolic dimension (LVEDD) was 4.14 cm, and presence of LV hypertrophy was consistent with hypertension (C). Because the severity of AR was not clinically significant and it was not expected to markedly increase, surgical intervention to the aortic valve was deferred. However, the diagnosis of AR altered circulatory management by increasing the vigilance of the echocardiographer to monitor LV chamber size during antegrade cardioplegia and initiating supplemental retrograde cardioplegia. PWTD, pulse-wave tissue Doppler.

The intraoperative decision for asymptomatic patients should be based on the severity of AR and the presence of left ventricular dysfunction. Patients with a left ventricular end-systolic dimension less than 50 mm would be expected to remain asymptomatic for the next few years; the risk for development of symptoms or left ventricular dysfunction would range from 1% to 2% per year to as high as 6% per year.^100,101 However, the yearly rate of becoming symptomatic or developing significant left ventricular dysfunction increases to greater than 20% when the left ventricular end-systolic dimension exceeds 50 mm.^100,102

Framing

Once the decision to surgically intervene has been established, the echocardiographer can have a valuable role in assisting the surgeon in formulating and implementing the intended change of plans for surgical management.

What information would be required for the surgeon to determine whether surgical valve repair or valve replacement is more appropriate? Is the choice of surgical intervention limited by surgical experience or prosthesis availability? What type of valve prosthesis is most appropriate for this patient in these circumstances? What size of valve prosthesis can be inserted? Is pathology of the aortic root or ascending aorta present? Could valve replacement have unintended consequences that could be prevented or detected and treated? Does “pressure recovery” play an important role in estimating valve stenosis using a transvalvular gradient Doppler methodology?

Data Collection

The choice of specific intervention, whether to repair or to replace the valve, will depend on pathology of the diseased valve, experience of the surgical team, and discussion between the surgeon and the cardiologist, in addition to patient variables. The echocardiographer can best contribute to the surgeon’s understanding of the pathogenesis of AV disease by assessing function and structure in the short- and long-axis imaging planes. The surgeon is provided with a complete anatomic echocardiographic assessment of the AV and root, including dimensions of the annulus, sinus of Valsalva and sinotubular junction, leaflet anatomy, primary mechanism of regurgitation, and severity of thickening/calcification.

Discussion

The development of reparative procedures for the AV has lagged behind those implemented for the MV or for the development of prosthetic devices. Some of the major factors are the apparently irreversible changes produced by calcification of the cusps, the retraction of cusp tissue in AR, and the lack of durable biologic material suitable for supplementing the valvular apparatus. It is unlikely that surgical repair of stenotic and highly calcified valves will achieve much success within the near future. However, aortic valvular repair has been successfully applied for the resuspension of the AV to correct acute AR resulting from an aortic dissection. Yacoub, Cohn, and others reported improved short- and long-term results for minimizing residual AR and freedom from reoperation.^103–105

A diseased valve characterized by thickened, calcified leaflets and having restricted mobility is not a candidate for valve repair and would require valve replacement. Alternatively, remodeling of the aortic root and AV to improve the geometry of the tricuspid AV has been used successfully in cases of acute type A aortic dissection. In addition, a select cohort of patients having AR as the primary valvular diagnosis may be candidates for surgical repair. In the case of bicuspid AVs or isolated AR, successful repair of the AV was more likely when the anatomy demonstrated left coronary/right coronary cusp fusion and when the primary mechanism of AR was leaflet prolapse or commissural separation, compared with poor leaflet coaptation of a “central defect.”¹⁰³ Although a complete review of the technical aspects and controversy of AV-sparing surgeries are beyond the scope of this chapter, it should be noted that aortic dilatation has a negative prognostic implication for successful long-term outcome. A more detailed discussion of surgical repair of the AV can be found in Chapter 19 and in literature reviews.^104,105

Valve repair and replacement surgery for patients having symptomatic valve disease have been heralded as significant successes. Advances in surgical technique and perioperative care have enabled the morbidity and mortality of valve replacement to decrease despite the increasing proportion of high-risk patients. The choice of prosthetic heart valve replacement generally is based on patient age, size, and wishes¹⁰⁶ (Figure 13-14). A joint decision with the patient, the cardiologist, and the cardiac surgeon is indicated before surgery.

Figure 13-14 Algorithm for choice of prosthetic heart valve.

A/C, anticoagulation; AVR, aortic valve replacement; INR, international normalized ratio; MVR, mitral valve replacement.

(Redrawn from Rahimtoola SH: Choice of prosthetic heart valve for adult patients. J Am Coll Cardiol 41:893, 2003, Figure 8.)

Compared with nature’s own heart valves, prosthetic valves are relatively stenotic. Prosthetic valves are sized based on their external diameter, even though their transvalvular gradient reflects the internal diameter and orifice size. Each valve has an “effective orifice area” (EOA) that is based on in vitro studies performed by the manufacturer. The clinical importance of the disparity between prosthesis annular size and EOA becomes evident when relatively small prosthetic valves are inserted into comparatively large patients. Prosthetic valve/patient mismatch results when transvalvular flow through a prosthetic valve is limited and produces clinically significant stenosis (Figure 13-15). The increase in transvalvular blood-flow velocity necessary to accommodate and sustain an acceptable SV in a large patient with a prosthetic AV can produce surprisingly high transvalvular pressure gradients.

Figure 13-15 Detection of a significant transvalvular pressure gradient across a previously implanted prosthetic aortic valve (AV).

Patient who underwent a previous AV replacement (AVR) was now scheduled for coronary artery bypass grafting and possible AVR for aortic stenosis. The preoperative evaluation included an echocardiogram that demonstrated a small paravalvular leak and blood cultures that were negative. A, Color-flow Doppler image of the AV in midesophageal short-axis imaging plane that shows turbid flow from the orifice of the 19-mm Bjork–Shiley mechanical valve. The aortic annulus and other valves were inspected for evidence of active infection or abscess, but none was found. The clinical decision was to replace the AV or repair the paravalvular leak. The prosthetic valve appeared to be functioning normally aside from a small paravalvular leak. The intraoperative transesophageal echocardiographic (TEE) examination documented preserved left ventricular (LV) systolic function and significant LV hypertrophy with a posterior wall thickness of 1.6 cm. Examination of the mechanical prosthetic valve showed a well-seated valve with a small paravalvular leak at the fibromuscular continuity and the absence of any abscess or vegetations (B). Although the valve appeared to be functioning normally, the transvalvular pressure gradient was 40 mm Hg in the absence of any outflow tract obstruction (C, D). The presence of a significant gradient in a normally functioning valve and presence of significant LV hypertrophy raised the possibility of valve mismatch; the small paravalvular leak was determined not to significantly contribute to the transvalvular gradient. The patient was an active 64-year-old man who weighed 98 kg and had a body surface area of 2.2 mm². The presence of a significant transvalvular gradient across an apparently normally functioning valve in a large patient supported the diagnosis of prosthetic valve mismatch. The patient underwent a root replacement (No. 23) and had a transvalvular pressure gradient of 14 mm Hg. LVEDD, LV end-diastolic dimension; PWTD, pulse-wave tissue Doppler.

An objective of valve replacement surgery is to provide the patient with the largest EOA. Prosthetic valve mismatch is more likely to occur in older patients, those with increased body surface area (BSA), smaller prosthetic valve size, and preoperative diagnosis of valvular stenosis.¹⁰⁷ The incidence of AV mismatch increases with prosthetic annular size less than 21 mm. In larger patients, mismatch can occur in patients receiving prosthetic AVs greater than 21 mm because dominant factors in determining the pressure gradient are the SV and CO, which are related to body size and level of activity. The association of mismatch with stenotic native AVs is a result of calcified native valves, a calcified annulus, and a smaller annular size. The treatment of AS through AVR in the setting of preserved ventricular systolic function may result in dramatic increases in SV and CO. The resultant increase in transvalvular flow produces a large transprosthetic pressure gradient. In contrast, valvular insufficiency is usually associated with annular dilatation and decreased systolic function. Hence patients with annuloectasia often receive larger prosthetic valves, and functional AS after AVR rarely is a problem.

Increased postoperative transvalvular gradients have been implicated in increasing long-term morbidity and mortality.^108–110 The short-term clinical significance of prosthetic valve size mismatch is controversial and may be associated with immediate postoperative risk for cardiovascular complications and mortality. Controversy stems, in part, from studies with limited sample populations and the failure to normalize specific valvular EOA to a metric of patient size. Nonetheless, the concept of prosthetic valve mismatch adversely affecting outcome is generally accepted. The long-term risk for prosthetic valve mismatch is inability to restore or improve functional capacity. A consequence of an increased residual transvalvular pressure gradient is to prevent or attenuate regression of left ventricular hypertrophy.¹¹¹ Left ventricular hypertrophy is associated with diastolic dysfunction, decreased exercise capacity, and is a predictor of increased mortality. Significant differences in regression of left ventricular hypertrophy occur in patients who received prosthetic AVs larger than 21 mm (−21%) compared with patients with prosthetic valves that were less than 21 mm (−8%).^111,112 Hence it would be expected that a corresponding difference in functional capacity would parallel these differences in hypertrophy regression. A decreased indexed EOA (EOA/BSA = cm²/m²) was an independent predictor of long-term morbidity after aortic and MV replacement.¹¹³

In general, an indexed EOA for a prosthetic AV should be greater than 0.85 cm²/m².¹¹³ For a similar external rim size, the EOA can vary considerably between mechanical and bioprosthetic AVs and by manufacturer. Bioprosthetic valves tend to have a significantly larger EOA compared with mechanical valves. Biologic aortic roots have even larger EOAs. If the anatomy of the aortic root precludes implantation of a suitably sized prosthesis, the surgeon may perform a supra-annular implantation, allowing the placement of a larger valve. Alternatively, creation of a larger EOA may require enlargement of the aortic root, a biologic aortic root, a pulmonic valve autograft (Ross procedure), or implantation of a stentless valve (see Chapters 19 and 21).

Framing

Valvular diseases do not often occur in isolation. Other anatomic structures may be influenced, either by the same pathophysiology or as a secondary consequence of the primary valvular lesion. Mitral insufficiency is a common finding, occurring in about two thirds of the patients having significant AS.¹¹⁴ Patients presenting for AVR commonly have mitral regurgitation and pose the question of whether to repair or replace the MV in addition to the AV (Figure 13-16). Such an undertaking is not without risk. Patients undergoing double-valve replacement have increased mortality compared with isolated AVR.

Figure 13-16 Management of concurrent mitral regurgitation (MR) in a patient undergoing aortic valve replacement (AVR) for aortic stenosis (AS).

The patient is a 70-year-old who underwent an AVR for critical AS. The patient had history of hypertension, moderate MR, congestive failure, and episodes of shortness of breath. Left ventricular (LV) function was moderately depressed and the chamber size was dilated (LV end-diastolic dimension [LVEDD] = 6.36 cm) (A). B, The patient had a central jet of MR that was graded as mild; the annulus was mildly dilated, and the pathology of the leaflets and apparatus was only mildly thickened. The pulmonary vein blood flow showed minimal blunting of the systolic component, consistent with increased left atrial pressures but not clinically significant MR. The discrepancy in the severity of MR between the preoperative study and intraoperative transesophageal echocardiogram (TEE) reflects the effect of general anesthesia on loading conditions. After reviewing the prebypass TEE examination and the patient’s history, it was decided to perform only an AVR. A bioprosthetic valve was chosen by the surgeon because of the patient’s age, thus eliminating the requirement for anticoagulation. Because the MR was believed to be more functional, the replacement of the severely stenotic AV, the major pathologic lesion, was anticipated to decrease MR over time. The choice of implanting a No. 23 pericardial prosthetic valve was based on the annular size that was measured before initiation of bypass. The postbypass TEE examination documented that the gradient across the AV decreased from 74 to 18 mm Hg and cardiac function (C) and MR (D) improved after AVR. CABG, coronary artery bypass grafting; CPB, cardiopulmonary bypass.

Should moderate mitral regurgitation be surgically addressed in patients undergoing AVR for AS? Does the severity of mitral regurgitation regress after AVR? Does the presence of AR in patients with severe AS alter the expected prognosis of the regression of mitral regurgitation after AVR? Can it be predicted which cohort of patients with AS would benefit from concomitant MV repair? In which patient cohort would the mitral regurgitation be expected to regress with the unloading effects of replacing a stenotic AV?

Data Collection

The most important data for these decisions are the grade of mitral regurgitation, AV area, anatomy of the MV, and cause of mitral regurgitation (i.e., rheumatic disease, ischemic, myxomatous degeneration). Patients are likely to have the signs and symptoms of AS and mitral regurgitation and pulmonary hypertension. The grade of mitral regurgitation is often not evident until a TEE is performed and may be underestimated by a preoperative transthoracic examination. An enlarged LA suggests the mitral regurgitation is not acute.

Discussion

Mitral regurgitation is a maladaptive consequence of increasing AS. More significant mitral regurgitation is associated with greater trans-AV pressure gradients, as well as more progressive dilatation and worsening systolic function.^115,116 If MV anatomy is markedly abnormal and the severity of regurgitation is severe, the decision is relatively obvious: MV repair or replacement at the time of AVR. If the MV is anatomically normal (no leaflet prolapse, no perforation, no rheumatic changes) and the regurgitation is trace or mild, the decision is also relatively obvious: Correct the AS and do not surgically address the MV. In the setting of rheumatic AS, it is common to have other valves involved in the disease process. Hence mitral calcification, thickening, and leaflet fusion are common in patients with rheumatic AS. The MV is rarely replaced based on anatomic abnormality alone in the absence of significant stenosis or regurgitation.

The more controversial decision is whether to surgically address the MV that has mild-to-moderate regurgitation when a patient presents for AVR for AS. There is neither an absolute answer nor consensus. The uncertainty of the decision is based on the unpredictability of the residual mitral regurgitation after the AV is replaced. If the grade of mitral regurgitation of moderate or severe were expected to be reduced to mild, the operative team might elect not to operate on the MV. A confounding factor is the grade of mitral regurgitation during general anesthesia may be underestimated and not reflect the grade during the awake or exercising state. Furthermore, the lasting effects of replacing an AV in the setting of mitral regurgitation are not established immediately after CPB. The heart undergoes significant changes after AVR for AS, including changes in pressure-volume relations of the LV, transmitral pressure gradients, and left ventricular remodeling.

Mitral regurgitation with an anatomically normal MV is referred to as “functional mitral regurgitation.” Increased left ventricular afterload, driving pressure, and left ventricular remodeling have been implicated in the development of functional mitral regurgitation in patients with severe AS. The natural history and clinical impact of mild-to-moderate mitral regurgitation in patients undergoing AVR for AS suggest that concurrent MV surgery may not be warranted in all cases. In the cohort of patients with functional mitral regurgitation and severe AS, the decrease in mitral regurgitation after AVR alone was variable; between 39% and 90% of patients having moderate mitral regurgitation experienced a regression in its severity after AVR.^117–122 In general, patients with functional mitral regurgitation of moderate grade had significant improvement (decrease) in the mitral regurgitation after AVR, suggesting that leaving the MV alone in this setting is not unreasonable. Conversely, a significant number of patients, between 8% and 64%, had either no change or an increase in mitral regurgitation severity. In a study of 196 patients who had concomitant mitral regurgitation, Moazami et al¹²¹ noted a decrease in 3-year survival for those patients whose moderate-to-severe mitral regurgitation persisted after AVR and recommended surgical intervention for this cohort of patients.

In general, functional mitral regurgitation of moderate severity usually will regress at least one grade after AVR for severe AS (see Figure 13-16). Mild-to-moderate mitral regurgitation that improves after AVR for AS seems to be a lasting phenomenon.¹¹⁷ Unfortunately, there has been little control for the severity of concurrent AR, which is a common finding associated with AS. Although the physiologic load of patients with combined AS and AR is different, the particular pathology of the AV did not affect regression of mitral regurgitation after AVR.¹²¹ Harris et al¹²⁰ ascribed the decrease in the severity of mitral regurgitation after AVR to several anatomic changes: decreases in mitral annular area, left atrial size, and left ventricular length. Together with the decrease in driving pressure that occurs with resolution of AS, all three of these anatomic changes alter the architecture of the MV and ventricle and contribute to decreasing functional mitral regurgitation. The greater the decrease in systolic area and increase in FAC of the LV after AVR, the greater is the reduction in postoperative mitral regurgitation.¹²⁰

It is difficult to define the negative predictors for the regression of mitral regurgitation after AVR. The presence of valvular abnormalities, such as rheumatic leaflet thickening, calcification or prolapse (i.e., myxomatous degeneration), and chordal pathology that could contribute to a “nonfunctional” cause of mitral regurgitation would most likely persist. Mitral annular calcification is a potential complicating factor and its impact is unclear. Annular calcification would be expected to impede the reduction in MV area that normally occurs during systole. This reduction appears to be important in the mechanism of mitral regurgitation reduction after AVR for AS. Likewise, mitral annular calcification renders mitral surgery more difficult and the implantation of a mitral prosthesis more problematic, likely increasing the risk for paravalvular leak and AV groove disruption. Many patients with severe AS have a calcified AV with concomitant MV annular calcification. Mitral annular calcifications may be predictive of fixed mitral regurgitation after AVR.¹²³ An enlarged left atrium (>5 m) has been reported to be a significant predicator of persistent mitral regurgitation after AVR.¹²⁴ Other factors to consider include presence of CHF, CAD, pulmonary hypertension, and atrial fibrillation. Ischemic mitral regurgitation may improve after revascularization as a more favorable balance of myocardial oxygen supply/demand is produced after AVR.¹¹⁸ “Functional” mitral regurgitation of moderate-to-severe grade that is not operated on during AVR may remain unchanged after AVR and require reinstitution of CPB for MV surgery. In the absence of more predictable criteria to define this group, patients with moderate-to-severe mitral regurgitation caused by intrinsic valvular pathology should be considered for surgical intervention. Reports suggest that AVR combined with MV repair can be accomplished with excellent long-term results and minimal effect on mortality.¹¹⁸

Framing

The most disabling complication after cardiac surgery is stroke. Major focal and nonfocal neurologic deficits, cognitive decline, and coma after surgery are common. The pathogenesis of cerebral damage is multifactorial, with embolism considered a major contributor. Other factors include hypotension, low flow, reperfusion injury, and inflammation. Embolic events are strongly associated with the severity of atherosclerotic disease, characterized by plaque thickness of greater than 4 mm, ulcerated plaques, and mobile protruding plaques in the aorta.^125,126 The severity of atherosclerosis of the descending aorta, as determined by TEE, is a significant risk factor and an independent predictor of adverse cardiac and neurologic outcome in patients undergoing CABG.¹²⁷ Surgical manipulation of the thoracic aorta may liberate debris from diseased aortic tissue. The process of microembolization has been detected by transcranial Doppler during aortic cannulation, application and removal of the aortic cross-clamp, commencement of CPB, and initiation of ventricular ejection. The clinical consequence of distal embolization is dependent on the number, composition (e.g., air bubbles, fat particles, platelet aggregates, and calcium deposits), size, and location of the emboli (see Chapters 12, 16, 18, 19, 21, 28, and 36).

Is the patient at increased risk for postoperative neurologic dysfunction? Would epiaortic ultrasound or TEE improve detection of significant atheromatous disease? What modifications to the conduct of the general anesthesia, monitoring, CPB, or cardiac surgery could be implemented in a risk-reduction effort? Is epiaortic scanning necessary?

Data Collection

TEE imaging of the anterior wall of the ascending aorta is limited by far-field imaging resolution and its juxtaposition to air in the open chest. Imaging of the mid and distal ascending aorta by TEE is limited by the interposition of the airway with the esophagus and aorta. Epiaortic ultrasound can provide high-definition imaging of these otherwise hidden portions of the aorta. Epiaortic scanning offers higher sensitivity for detection of atheroma of the ascending aorta, as compared with TEE, especially in the mid and distal segments.^3,128 The descending aorta is immediately adjacent to the esophagus and easily imaged using conventional TEE. A common practice is the interrogation of the descending thoracic aorta for high-grade atheroma. In the absence of atheromatous disease in the descending aorta, the ascending aorta and locus of aortic cannulation are significantly less likely to have high-grade disease. If the descending aorta contains high-grade or mobile atheroma, it may be prudent to examine the ascending aorta with epiaortic scanning for potential sites for aortic cannulation and clamping (Figure 13-17).

Figure 13-17 Detection of atheromatous plaque by epiaortic ultrasound.

A 67-year-old patient was scheduled to undergo coronary artery bypass grafting × 3. The transesophageal echocardiographic (TEE) examination was performed to evaluate the mitral valve for mitral regurgitation that might necessitate surgical intervention. Routine examination of the descending thoracic aorta detected severe atherosclerotic disease characterized by several ulcerated areas and plaques with a thickness of 4 mm. Because the severity of atherosclerotic disease in the descending aorta predicts significant disease in the ascending aorta,¹²⁷ examination of the ascending aorta was attempted. TEE examination of the ascending aorta at the site of aortic cross-clamp and cannulation (white rectangle) was limited by poor resolution in the far field and interposition of the trachea (A). A handheld probe that was placed in a sterile sleeve was used to examine the aorta (B). The bottom of the sleeve was filled with saline as an offset to better image the anterior surface of the aorta (Ao). Several calcified plaques (arrows) were detected. Based on findings of the epiaortic scanning, the surgeon modified the site of cannulation and used a single cross-clamp technique to avoid a second clamping of the aorta during the proximal anastomosis.

Discussion

Historically, surgeons palpate the ascending aorta to determine the extent of intraluminal atherosclerotic disease in an effort to choose an appropriate location for cannulation, cross-clamping, and proximal graft anastomosis. However, palpation is a notoriously poor predictor of atheromatous disease.^129–131 The occurrence of atheroma in the ascending aorta of cardiac surgery patients may be as high as 60% to 90%.^130,132 Advanced age, hypertension, and diabetes are risk factors for atheromatous disease of the aorta and stroke after cardiac surgery.^3,133 The information provided by ultrasound imaging of the aorta defines the location and severity of atheromatous disease, which can guide the surgeon to more strategically choose the cannulation, cross-clamp, and anastomotic sites.

Possible modifications to the conduct of cardiac surgery and CPB include (1) altering the site of cannulation, (2) aborting CPB alto-gether and performing the CABG surgery as an off-bypass procedure, (3) performing the surgery on a fibrillating heart without aortic cross-clamping, (4) performing aortic atherectomy or replacement of the ascending aorta, or (5) attempting a “no-touch” technique using deep hypothermic circulatory arrest. Epiaortic scanning combined with modification of surgery techniques in patients undergoing CABG decreased the number of emboli as detected by transcranial Doppler and significantly reduced neurologic behavioral changes at 1 week and 1 month after surgery.¹³⁴ Modifications in the methods used in CABG revascularization, including the avoidance of proximal aortic graft anastomoses after detection of atheroma by echocardiography, resulted in a lower incidence of late neurologic complications.²

The application of epiaortic scanning for all patients undergoing cardiac surgery is controversial. Epiaortic scanning takes time and expertise in its interpretation and poses the potential risk for wound contamination. There are insufficient data to support its use in all cardiac patients, although high-risk patients (advanced age, diabetes, hypertension) are the most likely to benefit.^135,136 As the profile of patients undergoing cardiac surgery ages, they have a greater incidence of atherosclerosis of the aorta and the presence of concomitant risk factors for postoperative complications. Echocardiographers working in the operating room need the skill set to interrogate the aorta by handheld ultrasound and the ability to guide the surgeon in performing the examination. It could be potentially life-saving in high-risk patients. A diagnostic scan of the ascending aorta and arch to evaluate the location of possible cannulation and clamp sites can be performed in several minutes. Standardized approaches to a comprehensive organized intraoperative epiaortic and epicardial examination are useful guides.¹³⁷

Framing

The incidence of postoperative neurologic dysfunction is greater in patients undergoing open cardiac procedures. The sources of microemboli include atheroma, calcified debris, tissue, and entrapped air. The particulate matter can embolize to the coronary circulation (see Figure 13-5), causing acute right ventricular or left ventricular dysfunction; to the cerebral circulation, producing postoperative neurologic deficits; and/or to distal vital organs.^138–140

Which patients are at risk for embolization from an intracardiac or aortic source? What are the echocardiographic characteristics that define this high-risk population? What techniques can be applied to detect intracardiac or aortic sources of emboli? Are any anesthetic or surgical interventions useful in decreasing the risk for embolic stroke during cardiac surgery?

Data Collection

TEE data collection is focused on intracardiac air, thrombus, calcium, or other particulate matter that is at high risk for embolization (e.g., endocarditis). Cardiac ultrasound is highly sensitive in the detection of intracardiac air. The American Society of Anesthesiologists/Society of Cardiovascular Anesthesiologists Task Force suggested the detection of air emboli during cardiotomy, heart transplant operations, and upright neurosurgical procedures to be classified as IIa indication for perioperative TEE. Classification IIa suggests a condition in which the existing evidence/opinion was in favor of the application of the technology. Entrapped air appears as bright, highly reflective particles that produce a shadowing artifact when the air coalesces. When performing de-airing maneuvers before release of the aortic cross-clamp and separation from CPB, the patients are placed in Trendelenburg position. In addition, an ascending aortic vent can be placed to aspirate the debris/air that was not removed by the left ventricular vent.

Discussion

The surgical team can perform several maneuvers to reduce sources of embolization that originate with open-chamber cardiac procedures. Placement of a drainage cannula in the LV via the right superior pulmonary vein can aspirate much of the debris that is suspended in solution. A small vent cannula in the anterior portion of the proximal ascending aorta often is inserted to aspirate gas bubbles or debris as they are ejected by left ventricular contraction. The heart is agitated by the surgeon in an attempt to eject the left ventricular debris or air while the “protective” aortic root vent continues to aspirate. The agitation often results in the release of pockets of air that may reside at the left atrial appendage, interatrial septum, and apex of the LV or are entrapped in the chordal structures or ventricular muscles of the heart. There is no consensus regarding when to terminate the de-airing process and separate from CPB. Not until some degree of pulmonary blood flow is restored is it possible to eliminate most of the retained intracardiac air. The aortic root vent often is left on continual aspiration even after CPB, with judicious volume replenishment through other sources. The aortic root vent should be discontinued before the administration of protamine. Although there is a paucity of evidence to support improved outcome, TEE is commonly used in the detection of retained air and the monitoring of the de-airing procedures. The effectiveness of de-airing can be tracked qualitatively. The relation between degree of “echo contrast” and actual quantity of air is unknown.

Framing

The utility of TEE has expanded beyond simply examining cardiac performance and pathology. TEE is well suited as a quick, highly sensitive, and specific tool to detect the presence of aortic pathology and some of its life-threatening sequelae. The detection and differentiation of true aneurysms and pseudoaneurysms, aortic dissections and transections, intraluminal atherosclerotic disease, and abnormalities of the aortic root are often the challenge of the intraoperative team (see Chapter 21).

What are the limitations of TEE for evaluation of the thoracic aorta? Is TEE the most appropriate diagnostic test to evaluate aortic pathology in the elective and emergency patient? Does the acuity of the patient preclude other imaging modalities, such as magnetic resonance imaging (MRI) or computed tomographic (CT) scanning? What are the quantitative measures that impact surgical management of aortic dissection, aortic aneurysms, aortic root disease, and aortic transection? What information can the echocardiographer provide the surgeon to assist in formulating a management strategy? What degree of aortic dilatation is sufficient to necessitate replacement? What is the risk of not replacing an aneurysmal aorta?

Data Collection

Data acquired in the assessment of the patient with aortic disease include a targeted history for evidence of malperfusion or vital organ injury (e.g., stroke) and examination of blood pressure and pulse in each of the extremities, carotid arteries, and abdominal aorta. A history of chest or back pain is common with aortic dissection or aneurysm, with tearing of the aorta and its adventitia. Increasing shortness of breath may occur with AR, heart failure, or a left pleural effusion. Patients may present with hoarseness related to recurrent laryngeal nerve damage causing vocal cord dysfunction. The preoperative blood work may detect renal insufficiency with an increased blood creatinine concentration. The chest radiograph often shows a widened mediastinum or calcification of the thoracic aorta. The TEE data are aimed at dimensional and anatomic assessment of the aortic root, ascending arch, and descending aorta. The echocardiographer measures and reports the size of the aortic annulus, sinus of Valsalva, sinotubular junction, and ascending aorta (see Figure 13-11). Left ventricular hypertrophy may suggest long-lasting hypertension. A dilated LV may accompany chronic AR. Dilatation of the aortic root is common in the setting of AS and AR. Patients with a congenital predisposition for aortic aneurysm, such as Marfan syndrome or bicuspid AV, often develop a dilated and aneurysmal ascending aorta. Epiaortic scanning increases the sensitivity of detecting abnormalities of the ascending aorta because it is in the far field from a TEE window and may be hidden by the interposition of the trachea and left mainstem bronchus, which obstruct the distal half of the ascending aorta.¹²⁸

Discussion

The echocardiographer may determine the definitive diagnosis and the extent of the disease. The information contributes to the decision of surgical management.

Aneurysm rupture occurred at a stunningly high frequency of 32% to 68% in patients with thoracic aortic aneurysms managed by conservative medical treatment and accounted for 32% to 47% of deaths. The one-, three-, and five-year survival of unoperated thoracic aneurysms was 65%, 36%, and 20%, respectively.^141–143 The risk for aortic rupture is a function of the size and rate of growth of the aneurysm. The risk for rupture for aneurysms less than 40, 40 to 59, and greater than 60 mm was 0%, 16%, and 31%, respectively.¹⁴⁴ The diameter of aneurysms tends to grow at a rate of approximately 5 mm/year. However, the rate of aneurysm expansion is variable. In general, the change in diameter of thoracic aortic aneurysms is a function of the initial diameter and increases more quickly in larger aneurysms.

Surgical intervention is a consideration for patients with an aneurysm of the aorta that is greater than 5 cm, accelerated rate of growth of the aneurysm, or the patient has a collagen vascular disease, such as Marfan syndrome. The incidence and progression of disease in the ascending aorta and arch after previous sternotomy have not been reported. The scarring and connective tissue growth around the mediastinal structures (aortic root, ascending aorta, and arch) after previous heart surgery would not be expected to alter the rate of aneurysm enlargement. However, it may change the pattern of rupture, likely decreasing the risk for frank rupture into a risk for a contained rupture. No consensus exists regarding management of cardiac patients having an incidental finding of an ascending aorta greater than 4 cm in diameter. The decision to pursue surgical intervention at the time of surgery depends on the pathogenesis of the disease, presence of symptoms, age of the patient, progression, and other operative and patient factors. A young patient with a 4.5-cm ascending aorta that is thinned and is associated with a bicuspid AV is more likely to undergo a root replacement and ascending aorta graft compared with an elderly patient with a similar aortic diameter who is undergoing CABG and whose aortic size has remained stable over the previous decade. Surgical replacement of the ascending aorta and arch is complex, requiring specialized surgical training. Significant modification of the initial surgical procedure may increase the risk for perioperative complications and potentially jeopardize the success of the originally planned operation. Extensive ascending aortic surgery involving the arch vessels requires hypothermic circulatory arrest followed by controlled exsanguination to allow for visualization of the operative field.

Disease of the AV often is accompanied by abnormalities of the thoracic aorta. Aortic root dilatation may require ascending aortic replacement. The decision is based on size of the aortic root diameter and presence of abnormalities of the aortic wall, such as thrombus, dissection, or focal aneurysm. There are no strict guidelines that dictate practice. The surgeon might choose to replace the entire root or just the AV and ascending aorta, sparing the sinuses of Valsalva. The Wheat procedure is a replacement of the AV and ascending aorta, while preserving the native sinuses of Valsalva and coronary ostia. Alternatively, in the setting of a dilated or aneurysmal sinus, the surgeon may elect to replace the entire aortic root. Choice of implanting a biologic root or composite AVR is a surgical decision based on many of the factors presented in Table 13-1. Implantation of a root or composite graft necessitates reattachment of the coronary arteries, increasing the risk for postoperative bleeding and coronary ischemia.¹⁴⁵ Despite a history of normal coronary arteries, the close scrutiny of regional wall motion after bypass is aimed at detecting newly created abnormalities of coronary blood flow. Air into the coronary circulation or compromise of coronary blood flow through the newly constructed coronary buttons may produce severe myocardial dysfunction and SWMAs within the specific coronary vascular distribution. Kinking of the coronaries or stenosis at the ostia often produce transmural ischemia that is not subtle. Enlargement of the sinus segment may facilitate coronary anastomoses. A prosthetic graft with an enlarged area to mimic the sinus of Valsalva has been developed. Enlargement of the sinus segment may facilitate coronary anastomoses and, thus, may decrease the risk for such complications, but this remains to be proved.

No consensus guidelines exist on the specific dimensions of the aortic root beyond which a root replacement is deemed necessary. However, distortion of the root producing severe AR or disruption in its integrity requires surgical attention. The echocardiographer reports the size of the aortic annulus, sinus of Valsalva, sinotubular junction, and ascending aorta (see Figure 13-11). These measurements are obtained using the AV long-axis view. The short- and longitudinal-axis views from TEE can define the severity and mechanism of AR, as well as the success of AV repair. Patients scheduled for aorta surgery with an intrinsically normal AV with AR caused by a correctable aortic lesion (incomplete leaflet closure, leaflet prolapse, or dissection flap prolapse) can undergo AV repair.^104,146 In contrast, valvular pathology that includes connective tissue degeneration (Marfan or bicuspid valve), aortitis, or severe calcification would most likely require valve replacement. Surgical management is dependent on the skill set of the surgeon and individual patient factors.

Framing

The unstable patient with suspected acute aortic disease or injury is often the most challenging of TEE cases. Few more crucially important decisions are posed to the intraoperative echocardiographer than to quickly and accurately diagnose the nature and extent of acute aortic injury. Hypotension and respiratory distress may prevent a complete and comprehensive evaluation before surgery (Figure 13-18). History often is unobtainable. The echocardiographer becomes a detective. Clues are quickly gathered from the available clinical presentation, history, and associated physical findings. The TEE is often the only modality used to establish the diagnosis and define the surgical plan.

Figure 13-18 Acute aortic syndrome as the cause of hemodynamic compromise.

A 62-year-old previously healthy, unrestrained driver had a motor vehicle accident. On arrival to the emergency department, the patient was hypotensive (blood pressure = 90/45) and tachycardic (heart rate = 120). He described an episode of loss of consciousness that was associated with severe chest pain but could not recall if the syncopal episode preceded the accident. The chest radiograph was significant for several fractured ribs, widened mediastinum, and a pleural effusion. The patient became progressively more unstable and was transferred to the operating room to perform diagnostic transesophageal echocardiography (TEE) and definitive surgical procedure if necessary. The echocardiographer performed a quick transthoracic echocardiographic examination that confirmed the presence of pericardial effusion with findings that were consistent for tamponade. After fluid resuscitation and induction of anesthesia, a TEE examination was performed. The midesophageal four-chamber view showed presence of a pericardial effusion (PE) that compromised right atrial filling (A). The midesophageal long-axis view of the aortic valve showed a type A dissection that was characterized by intimal flaps within the aortic root and that extended distally into the descending thoracic aorta. B, The annulus of the aortic valve was of normal size, but the sinus and root were markedly enlarged (diameter of sinotubular junction = 4.22 cm). The dissection extended into the noncoronary and right coronary sinus segments, narrowing blood flow at the coronary ostia (C, arrow). Although the electrocardiogram did not show acute ischemia, the right ventricular function and inferior wall of the left ventricle were mildly hypokinetic. Although an effaced aortic root, ascending aortic aneurysm, and acute dissection in this age group are suggestive of congenital bicuspid valve, the short-axis view of the aortic valve (D) showed a trileaflet valve with a coaptation defect with aortic insufficiency at the noncoronary cusp. The surgeon resuspended the aortic valve and replaced the ascending aorta and hemiarch with a tube graft. The valve repair was successful with only +1 aortic insufficiency and cardiac return to normal after surgery. RCA, right coronary artery.

Data Collection

Because the diagnosis and cause for instability are not established, the entire mediastinum, including the left pleural space, is interrogated before definitive therapy is initiated. Rarely is there not enough time to do a complete TEE examination. The operative team can often proceed with confidence in the management of these critically ill patients with only TEE to guide the treatment. The primary event in aortic dissection is a tear and separation of the aortic intima. It is uncertain whether the inciting event is a primary rupture of the intima with secondary dissection of the media or hemorrhage within the media and subsequent rupture of the overlying intima. Systolic ejection forces blood into the aortic media through a tear that leads to the separation of the intima from the surrounding media, creating a false lumen. Blood flow may exist in both the false and true lumens through communicating fenestrations. Aortic dissections are classified by one of two anatomic schemes (the DeBakey and Stanford classifications). Transection is diagnosed through the detection of para-aortic hematoma near the isthmus and a “step-up” in the internal media wall (see Chapter 21).

Discussion

Acute dissections (Stanford type A or DeBakey type I or type II) involving the ascending aorta or arch are considered acute surgical emergencies. In contrast, dissections confined to the descending aorta (distal to the left subclavian artery; Stanford type B or DeBakey type III) are treated medically unless the patient demonstrates proximal extension, hemorrhage, or malperfusion. From the International Registry of Acute Aortic Dissection, 73% of the 384 patients with type B dissections were managed medically; in-hospital mortality rate was 10%.¹⁴⁹ The long-term survival rate after applying medical therapy was approximately 60% to 80% at 4 to 5 years^149–152 and approximately 40% to 45% at 10 years.^151,152 Survival was best in patients with noncommunicating and retrograde dissections.¹⁵³ From the International Registry of Acute Aortic Dissection, in-hospital mortality rate for surgical patients was significantly greater (32%).¹⁴⁹ The increased rate of mortality for surgically treated patients likely was influenced by selecting a cohort of patients with more advanced disease and complicated course (malperfusion, leakage, extension). The overall reported short- and long-term outcomes were similar for medically treated patients with type B dissections.¹⁵² Of 142 patients with type B aortic dissections, there was a trend toward lower mortality with medical therapy compared with surgical treatment at 1 year (15% vs. 33%). Both groups had similar survival rates at 5 and 10 years (60% and 35%).¹⁵²

Endovascular stent grafting for descending thoracic aortic dissections is gaining momentum as a less invasive alternative to surgery in stable patients with type B dissections.^154,155 The stent graft is positioned to cover the intimal flap and seal the entry site of the dissection, resulting in thrombosis of the false lumen. In one nonrandomized evaluation of 24 consecutive patients with a subacute or chronic thoracic type B dissection, there was no morbidity (paraplegia, stroke, embolization, malperfusion syndrome, or infection) or mortality with stent grafting.¹⁵⁴ In contrast, surgical intervention was associated with a 33% mortality rate and a 42% incidence rate of adverse events within a 12-month period. Other studies are promising but less positive.¹⁵⁶ In a series of 19 patients with acute dissections (15 type B and 4 type A dissections), the major morbidity rate was 21% (small-bowel and renal infarction and lower extremity ischemia) and 30-day mortality rate was 16%. However, there were no additional deaths or instances of aneurysm/aortic rupture during the subsequent 13-month follow-up period.

Ascending aortic dissections (involving the aortic root, ascending aorta, or arch) are acute surgical emergencies because of the high risk for a life-threatening complication such as AR, cardiac tamponade, myocardial infarction, rupture, and stroke. The mortality rate is as high as 1% to 2% per hour early after symptom onset.¹⁵⁷ Neither acute myocardial ischemia nor cerebral infarction should contraindicate urgent intervention. Although patients with stroke in progress may be at increased risk for hemorrhagic cerebral infarction because of intraoperative anticoagulation, leading to hemorrhagic stroke, the authors have seen several patients who experienced dramatic neurologic recovery. Operative mortality for ascending aortic dissections at experienced centers varied from 7% to 36%, well below the greater than 50% mortality with medical therapy.^{150,157–164}

Traumatic aortic rupture is a life-threatening vascular injury that often results in lethal hemorrhage. In a multicenter trial of 274 patients, the overall mortality rate reached 31%, with 63% of deaths attributable to aortic rupture.¹⁶⁵ Aortic transection and rupture usually occurred at the aortic isthmus (between the left subclavian and the first intercostal arteries) and resulted from shear forces generated by unrestrained frontal collisions.¹⁶⁶ Although aortography had been considered the gold standard for the diagnosis of transection, TEE and contrast-enhanced spiral CT and MRI are currently favored, especially for patients with renal insufficiency.^167–171 Intravascular ultrasonography has been proposed as a potential diagnostic tool for the identification of limited aortic injuries.¹⁷² Traumatic aortic rupture needs to be distinguished from an aortic dissection. Imaging of a dissected aorta typically reveals true and false lumens at multiple levels. The focal aortic injury of aortic transection is quite localized and may be overlooked when performing a cursory examination. A second potential diagnostic problem is that protuberant atherosclerotic changes of the aorta may be difficult to differentiate from partial aortic tears. The thick and irregular intraluminal flap, which corresponds to disruption of both intimal and medial aortic layers, can be imaged in both the short- and long-axis planes in the vicinity of the isthmus. In the longitudinal view, the medial flap is nearly perpendicular to the aortic wall because traumatic lesions are usually confined within a few centimeters distal to the left subclavian. The formation of a localized contained rupture of the false aneurysm is common.^168,173 Color-flow Doppler imaging and spectral Doppler can be used to detect turbulence associated with nonlaminar flow at the aortic defect and the presence of a pressure gradient. Traditional treatment includes immediate surgical invention using a right lateral decubitus approach and resection of the aorta with insertion of a tube graft. Deployment of endovascular stent grafts has been successful. Two series that included a total of 16 patients having aortic transection reported successful repair with no mortality or serious morbidity.^174,175 However, the application of this device under such conditions poses a high risk for left subclavian malperfusion and paraplegia.¹⁷⁶ The decision regarding appropriate management and time course of therapy will depend on the technical availability and expertise within the institution and the forthcoming results of clinical trials that use newer, less invasive technologies.

Framing

The presence of an IMH is a subtle finding that may be missed when evaluating the thoracic aorta for a dissection or rupture. IMH is distinct from a classic aortic dissection but has been considered by some as being a variant.¹⁷⁷

What are the characteristics of an IMH? What are the common findings and symptoms at presentation? What, if any, is the clinical significance of this pathologic finding? Does location influence the choice of medical or surgical approach? What are the prognostic indicators of progression?

Data Collection

Aortic IMH is often considered a variant of an aortic dissection.¹⁷⁷ Although many of the patients may present with chest or back pain, many of the associated sequelae, such as pericardial or pleural effusions, stroke, and myocardial infarction, are absent.¹⁷⁹–¹⁸⁰ It is characterized by the absence of a detectable intimal tear. The hematoma probably is produced by a hemorrhage of the vaso vasorum into the aortic wall. In a meta-analysis, acute aortic IMH was most often associated with long-standing hypertension; traumatic cause was also a significant factor for 6% of cases.¹⁷⁸ In the meta-analysis of 143 cases, 81% were diagnosed by CT, and the remaining patients by MRI, TEE, or both.¹⁷⁸ The sensitivity and specificity of TEE to detect an IMH were reported to be 100% and 91%, respectively.¹⁸⁰ The diagnosis is characterized by the absence of a dissecting flap or intimal disruption with specific regional thickening of the aortic wall of greater than 7 mm in a crescent or circular shape. On MRI or CT, this crescent-shaped or circular area in the aortic wall demonstrates high attenuation that does not enhance with contrast.¹⁸¹ Although IMHs can be detected in any portion of the aorta, the location appears to be a factor in its cause and prognosis. The ascending aorta, which was involved in 33% to 57% of cases, appeared to represent the early stage of a classic dissection,^{178,179,181–183} whereas traumatic IMHs typically involved the descending thoracic aorta.¹⁸²

Discussion

Acute management is similar to that of a classic aortic dissection. In the meta-analysis of 143 patients with aortic IMH, lesions of the ascending aorta were associated with a lower early mortality with surgical intervention than medical treatment (14% vs. 36%); lesions of the descending aorta had a similar mortality with medical or surgical therapy (14% vs. 20%).¹⁷⁸ The force of left ventricular systolic blood pressure, the driving force for disease progression, is reduced by the administration of β-blockers. Patients with an IMH in the descending thoracic aorta should be medically treated using β-blockers. Management of patients having IMH of the ascending aorta is controversial. A comprehensive evidence-based approach is not practical at present. Research publications have been limited to case reports, single-institutional series, multicenter registries, and meta-analyses. The experience seems to vary between the Asian centers (Japan and Korean), which have more success with medical management, and the Western centers (European and American), which have advocated a more aggressive surgical management. In a recent single-institutional study from Korea of 357 patients having acute aortic syndrome, 101 patients had IMH, 16% of whom were unstable receiving surgery and the rest were medically treated.¹⁸³ Of the patients who initially received medical management, the aortic pathology progressed in 37% of patients, lending to delayed surgery or death. Prognostic factors of impending deterioration include IMH greater than 15 mm or increased aortic diameter. The difference in response to management strategies also may reflect population genetics, comorbid diseases (atherosclerosis, hypertension, environmental factors), or experience in accurately diagnosing this variant. The periaortic thickness of IMH, which can be small and subtle, may be underdiagnosed in the West.¹⁸⁴ The risk for sudden deterioration and death related to pursuing medical management must be balanced by the potential morbidity/mortality of surgery. In a center having low surgical mortality, a surgical approach might be favored. A high-risk patient at a surgical center of lesser experience may consider aggressive medical management.

In conclusion, the intraoperative echocardiographer is confronted with a broad array of diseases that require on-site decision making in the perioperative setting. The key ingredients to sound decision making are a broad fund of knowledge (database), a systematic approach with attention to all vantage points and frames, and identifying, addressing, and prioritizing the pertinent questions.

“Plans are nothing … Planning is everything.”

—Dwight Eisenhower