Table 37–1. Size Criteria for Surgical Intervention for Asymptomatic Thoracic Aortic Aneurysm

It is well-known that patients with Marfan disease are prone to unpredictable dissection at an early size. For this reason, earlier intervention is recommended for patients with Marfan disease as indicated in Table 37–1.

For patients with a positive family history, but without Marfan disease, the same criteria are applied as for Marfan disease, because malignant early behavior of the aneurysm in these patients may be seen as well.

If the patient has a positive family history, or if an afflicted family member has suffered rupture, dissection, or death, preemptive surgical extirpation is carried out earlier than otherwise.

Studies of aortic anatomy increasingly recognize that patients with a bicuspid aortic valve also have inherently deficient aortas. About 5% of bicuspid patients will dissect the aorta during their lifetimes. Therefore, although controversial, lower intervention dimensions have often been applied for patients with bicuspid aortic valve, as for Marfan disease. Bicuspid aortopathy was considered, so to speak, as “Marfan’s light.” However, recent data indicate that bicuspid patients are not quite as vulnerable as initially thought, so traditional, non-Marfan criteria can be applied. The inherent propensity of the bicuspid aorta to dissect must also be taken into account in planning surgical management of the ascending aorta in a patient who requires aortic valve replacement for stenosis or regurgitation. In such circumstances, the ascending aorta should usually be resected if it is 4.5 cm or greater in diameter, thus preventing future dissection events. In fact, experienced aortic surgeons, if they are in the chest to perform aortic valve replacement, will often replace the ascending aorta even at 4 cm diameter—to avoid the need for a later operation.

3. What is the Yearly Rate of Rupture or Dissection for Thoracic Aortic Aneurysms?

The preceding data indicate the cumulative lifetime rates of dissection or rupture by the time the aorta reaches a certain size. Determining the yearly risk of complications from the natural history of thoracic aortic aneurysm is more challenging because it requires extremely robust data. Such data must produce enough hard endpoints to permit analysis within a year’s time for different size strata. Calculations of yearly rates of rupture or other complications based on size of the aorta have been produced. These yearly rates are expressed based simply on the size of the aorta (Figure 37–11).

Figure 37-11. Probabilities that rupture, dissection, or death will occur have been calculated for aortic aneurysms in the chest. The likelihood of these events jumps sharply for aneurysms that reach 6 cm or higher.

These data all point to a diameter of 6 cm as a very dangerous size threshold. At or above this size, the yearly risk for rupture is about 4%, the yearly risk of dissection is about 4%, and the risk of death is about 11%. (Death is often directly related to catastrophic complications from the aneurysm. In fact, recent studies from Japan, based on routine post-mortem CT scan “autopsies,” show that a staggering 8% of all sudden deaths are due to Type A aortic dissection.) The chance of any one of these phenomena occurring—rupture, dissection, or death—is 14%/year. As a mnemonic point of reference, a 6-cm aneurysm can be equated to about the diameter of a soft drink can. When a thoracic aortic aneurysm reaches the diameter of a soda can, it has certainly reached the point where it poses a major risk to the patient. All current societal guidelines recommend a general criterion for prophylactic surgery of 5.5 cm for the ascending aorta (for individuals of average body size).

These analyses should permit accurate decision-making when seeing a patient during an office visit and considering preemptive surgical extirpation of thoracic aneurysms. These data allow the physician to form a reasonable estimate of the individual patient’s risk of dissection, rupture, or death for each future year of life, if the aorta is not resected. The risk of rupture, dissection, or death based on aortic size is presented graphically in Figure 37–11.

The question arises whether the same surgical intervention criteria should apply for a small woman as for a large man. It is true that a larger individual can be “allowed” a larger aorta, generally speaking. Conversely, even a moderate-sized aneurysm can be quite threatening in an individual of small stature. For this reason, adverse event rates (rupture or dissection) based on aortic size corrected for body surface area (BSA) have been analyzed. By plotting the aneurysm size along the horizontal scale and the BSA along the vertical scale, each particular patient can be classified into low-, medium-, or high-risk categories—thus taking account of the aneurysm size in relation to the patient’s physical size.

Two important recent developments regarding aortic prediction have occurred. First, it has been found that ascending aortic length (measured from the aortic annulus to the leading margin of the innominate artery) predicts adverse aortic events a bit more accurately than diameter. Elongation, it appears, is another sign of a damaged, vulnerable aorta. Secondly, it has been found, based on CT scans obtained fortuitously just before aortic dissection, that the dissection process itself produces an instantaneous 8-mm enlargement of the aorta. So, the aorta is actually smaller at the moment of dissection than previously appreciated on post-dissection scans. This argues that we should consider shifting our general criterion for intervention to 5.0 cm instead of 5.5 cm (Figure 37–12.) The next iteration of societal guidelines may endorse this “left-shift” to smaller sizes for intervention.

Figure 37-12. Consideration is currently being given as to whether the general criterion for intervention for ascending aortic aneurysm (in individuals of normal body size) should be shifted-left on the aortic size chart—to 5.5 cm.

Very recent studies have shown that enlargement of the aortic root is more “malignant” than similar degree of enlargement of the ascending aorta itself.

C. Symptoms & Signs

Most thoracic aortic aneurysms are asymptomatic and are detected fortuitously during imaging of other thoracic structures. When they are symptomatic, deep visceral pain in the upper anterior chest or interscapular back can occur. The aorta can “talk” to the patient via ascending sympathetic fibers that reach pain centers in the brain. Aortic pain differs from angina pectoris because it is not necessarily precipitated by exertion nor relieved by rest or nitroglycerin. Often, it is rather constant and not influenced by body motion or position. All patients with chest pain should have a screening chest radiograph. Rupture of a thoracic aneurysm usually causes excruciating pain, accompanied by profound dyspnea as the chest fills with blood, and quickly results in shock. A large ascending aortic aneurysm occasionally may result in dysphagia or stridor due to esophageal or large airway obstruction. Rarely, a large aneurysm may cause bone pain due to pressure against thoracic skeletal structures.

D. Physical Examination

Physical examination is usually unremarkable. The presence of a murmur of aortic regurgitation should raise the suspicion of ascending aortic aneurysm, as should features suggestive of Marfan syndrome or related conditions. The thumb-palm sign, in which the tip of the thumb can be extended beyond the edge of the flat palm, can be a useful indicator of connective tissue laxity, raising the specter of associated aneurysm disease (Figure 37–13). Rarely, an abnormal pulsation will be felt due to a large aneurysm contacting the chest wall.

Figure 37-13. Patient with a positive thumb-palm sign for connective tissue disease. Being able to cross the thumb beyond the edge of the palm indicates that the long bones are excessive and the joints are lax. (Reproduced, with permission, from Elefteriades JA, et al. Guilt by association: paradigm for detecting a silent killer [thoracic aortic aneurysm]. Open Heart. 2015;2:e000169. With permission from BMJ Publishing Group Ltd.)

E. Diagnostic Studies

The remarkable strides made in recent decades in three-dimensional body imaging have dramatically advanced the diagnosis and treatment of thoracic aortic aneurysm. Echocardiography (especially transesophageal), computed tomography (CT), and magnetic resonance imaging (MRI) scans all yield images that clarify the presence, location, size, and extent of aneurysmal disease. An example of the precise imaging afforded by MRI is indicated for a specific, very extensive aneurysm in Figure 37–14. Proper interpretation of CT and MRI scans is essential; specifically, axial sections for aortic measurements must be selected carefully so that the aorta is vertical at the chosen site (oblique measurements will be exaggerated by an oblong rather than circular aortic contour). The “centerline method” applies automatic computerized assessment of diameter perpendicular to the long axis of the aorta, thus avoiding obliquity issues.

Figure 37-14. Magnetic resonance scan of a massively dilated aorta, which extends from the aortic valve to the iliac bifurcation. Note that the heart is compressed to a small shadow crushed between the elongated aorta and the diaphragm. This aneurysm was successfully resected in two stages.

It is easy to make errors in measurement of aorta on CT or MRI imaging. Potential sources of errors are indicated in Figure 37–15.

Figure 37-15. Potential sources of error in measurement of the aorta on CT images. These errors plague everyday practice and impede accurate assessment of growth over time. These sources of error include: systole vs. diastole, with or without aortic wall, cursor on or just outside of aortic wall, obliquity of the measuring line relative to the aorta, and sinus-to-commissure or sinus-to-sinus for the aortic root (sinuses of Valsalva). (Reproduced with permission from Elefteriades JA, Mukherjee SK, Mojibian H. Discrepancies in Measurement of the Thoracic Aorta: JACC Review Topic of the Week. J Am Coll Cardiol. 2020;76(2):201217.)

In this era of specialized three-dimensional imaging, it is important not to forget the chest radiograph, which can often yield significant information about the thoracic aorta. An example is provided in Figure 37–16. Ascending aortic aneurysm presents as a bulge beyond the right hilar border. Arch aneurysm produces enlargement of the aortic knob. Descending thoracic aneurysm is often easily seen as a deviation of the stripe of the descending aorta, which normally runs parallel to and just left of the vertebral column.

Figure 37-16. An exemplary chest radiograph indicating that significant information about the aorta can be gleaned from this simple test. Note the bulge of the ascending aorta to the right of the upper mediastinal border. This young patient with Marfan disease suffered dissection at an ascending aortic dimension of 4.8 cm.

Many if not most thoracic aortic aneurysms are found incidentally on three-dimensional imaging studies done for another, unrelated purpose. Most thoracic aneurysms remain undiagnosed in the general population. Recently, a paradigm called “guilt by association” has been proposed for uncovering hidden aneurysms in the general public based on the close association of thoracic aortic aneurysm with a group of related conditions (Figure 37–17). These related conditions include various medical conditions (intracranial aneurysm, abdominal aortic aneurysm, temporal arteritis, autoimmune disorder, renal cysts), certain aortic anatomic variants (bovine aortic arch, direct origin of left vertebral artery from aortic arch, bicuspid aortic valve), and family history of aneurysm disease. The presence of one of these associated conditions warrants imaging to rule out concurrent silent aortic aneurysm.

Figure 37-17. Paradigm of guilt by association for detection of silent thoracic aortic aneurysms. (Reproduced with permission from Elefteriades JA, et al. Guilt by association: paradigm for detecting a silent killer [thoracic aortic aneurysm]. Open Heart. 2015;2:e000169. With permission from BMJ Publishing Group Ltd.)

Table 37–2. Current Risks of Thoracic Aortic Surgery

B. Indications & Contraindications

By considering the rates of natural rupture, dissection, and death from the thoracic aneurysm itself versus the risks of operation, the physician can make an informed recommendation about elective, preemptive surgery. Once patients and their families are provided the natural history and surgical risk data, they often have strong opinions of their own. Some patients are reluctant to undergo major surgery, with its significant attendant risks, for an asymptomatic problem. Most patients, however, seem to feel they will never be comfortable until the threatening aneurysm is resected.

One more very important general point needs to be considered. Once the aorta has dissected, the prognosis is forevermore adversely affected (Figure 37–18). Patients who required emergency surgery not only had a higher rate of early mortality, but their survival curve was dramatically poorer. Patients who elected for planned, nonemergent procedures showed a survival rate very similar to that of a normal population. The poor long-term outlook for patients who required emergency surgery is due largely to the fact that, even after surgical replacement of portions of the aorta, the remainder of this vital organ will always remain dissected. Because the aortic wall was deficient to start with, at half-thickness, after dissection, it is rendered even more vulnerable to subsequent enlargement and rupture.

Figure 37-18. Long-term survival rates based on treatment. Medically treated patients had, of course, smaller and less symptomatic aneurysms. Note particularly that patients having emergent surgery not only manifested a higher likelihood of perioperative mortality, but also had a poorer long-term outlook. On the other hand, patients who received elective surgery showed excellent survival rates, comparable to an age- and sex-matched normal population. These data argue for elective, prophylactic extirpation of the aneurysmal aorta, before rupture or dissection can occur.

C. Surgical Techniques

As discussed earlier, the type of operation for the ascending aorta is based on the pattern of aneurysmal pathology. For many patients, a supracoronary tube graft suffices (Figure 37–19A). For others, a composite graft, including both a valve and a graft, with obligate coronary artery reimplantation, is appropriate (Figure 37–19B). Newer valve-sparing aortic replacement procedures (eg, David procedure) have been developed and validated in large clinical series. Although these valve-preserving operations are becoming increasingly popular, the long-term fate of the preserved valve remains somewhat controversial.

Figure 37-19. (A) Supracoronary tube graft replacement and (B) composite graft replacement. (A: Reproduced with permission from Cooley DA, Wukasch DC. Techniques in Vascular Surgery. Philadelphia: WB Saunders; 1979.)

The main debate regarding the conduct of ascending aortic operations and those on the aortic arch concerns the optimal means of protecting brain function during the time that anastomoses in the vicinity of the aortic arch are performed. Deep hypothermic circulatory arrest—a state of suspended animation, which is generally safe for 30–45 minutes or longer—is preferred by many surgeons because of its simplicity and effectiveness. In a study on 500 patients, the effectiveness of this remarkable technique as a sole means of brain preservation was confirmed. Retrograde cerebral perfusion—via the superior vena cava—has its advocates, although the actual amount of effective brain perfusion achieved by this means has been questioned. Direct perfusion of the head vessels—usually via a cannula in the innominate artery or cannulas in both the innominate and the left carotid artery—also has its supporters, despite its added complexity. Direct perfusion is gaining in popularity, and it does provide a margin of protection, especially for very complex arch reconstructions or for surgical teams relatively inexperienced with arch replacement. However, very recent evidence has demonstrated a sobering 100% rate of cerebral embolic events (often subclinical) with antegrade perfusion. No technique has been demonstrated as conclusively superior over the others.

For descending and thoracoabdominal operations, the technique of left atrial to femoral artery bypass has become extremely popular. This method takes strain off the heart by diverting blood away from the left ventricle. This approach mitigates the effect of high aortic cross-clamping on cardiac afterload. It also perfuses the lower body, especially the extremely vulnerable spinal cord. Despite decades of concerted attention, paraplegia from descending and thoracoabdominal aortic replacement continues to be a major clinical problem. The cause is multifactorial, with clamp time, air and particulate embolism, and disconnection of critical intercostal branches all playing a role. Besides the benefits of left atrial to femoral artery perfusion, most authorities feel that routine spinal fluid drainage and deliberate maintenance of a strong postoperative blood pressure (to encourage collateral blood flow) are also effective adjuncts against the complication of postoperative paraplegia.

D. Specific Clinical Scenarios & Issues

1. Patient with Pain, but Aneurysm Smaller Than Criteria

The answer to whether such an aorta should be replaced is a resounding yes. The dimensional criteria are specifically intended for asymptomatic patients. Any and all symptomatic aneurysms need to be resected because symptoms are a precursor to rupture. Aneurysm pain represents stretching or irritation of the aortic adventitia, the adjacent chest wall, the mediastinal pleura, or some other structure impinged on by the expanding aneurysm. Even an aorta smaller than the criterion can rupture or dissect. The size criteria are “fuzzy” lines, not sharp demarcations. A patient with pain but a relatively small aneurysm is of great concern, and preemptive resection is needed. In one case, a patient complained of pain typical of an ascending aortic aneurysm. The aorta was 5.0 cm. Because the medical team thought this was too small for resection, they underestimated the symptoms at presentation. The aorta subsequently ruptured, and the patient died within 48 hours. This point cannot be overemphasized: The size criteria are explicitly intended only for asymptomatic patients; all symptomatic aneurysms need to be resected.

2. Differentiating Aneurysm Pain from Musculoskeletal Pain

This very important point is not always easy to determine, even in the most experienced hands. The patient usually has a good sense of whether the pain is originating from muscles and joints. The clinician usually gets an additional understanding by asking the following questions:

1. Is the pain influenced by motion or position? (If so, it is probably musculoskeletal.)

2. Do you have a history of lumbosacral spine disease or chronic low back pain? (If so, the symptoms may not be aortic in origin.)

3. Do you feel the pain in the interscapular back? (An affirmative answer indicates an almost certain relationship to thoracic aortic aneurysm.) The interscapular region corresponds to the upper descending aorta.

Presume that the pain is aortic in origin if no other cause can be conclusively established. This is the only approach that can prevent rupture.

3. Appropriate Interval for Serial Aortic Imaging

Patients with a thoracic aortic aneurysm should be monitored indefinitely. Stable, asymptomatic patients can undergo imaging about once every 2 years, remembering that the aneurysmal aorta grows at a relatively slow 1 mm/year. In case of new onset of symptoms, imaging should be done promptly, regardless of the interval from the prior scan. For new patients, for whom only one size data point is available, imaging should be done at shorter intervals until the behavior of aorta is understood. Imaging may be done every 3–6 months for new patients with moderately large aortas. Remember to compare the present scan with the patient’s first scan, not with the last prior scan. That is the way to detect growth. Many patients have suffered because scans were only compared with the last prior scan, and major growth went undetected.

4. Choice of Imaging Modality for Serial Follow-Up

Three quality imaging techniques are currently available: echocardiography, CT scan, and MRI. If echocardiography is chosen, it is important to remember that a standard transthoracic echocardiogram cannot see the distal ascending aorta, the aortic arch, or the descending aorta with conclusive accuracy because of intervening air-containing bronchi and lung tissue. Supplement such studies with a periodic CT scan or MRI, which can visualize the entire aorta. The choice between CT and MRI may depend on ease of availability and radiologic expertise in a particular environment. Both modalities can image the entire aorta extremely well. Elevated creatinine or contrast allergy may contraindicate CT and instead favor MRI. The need to evaluate complex aortic lesions in multiple imaging planes would also favor MRI (although very recently, concerns have been raised about the risk to the kidneys of gadolinium contrast agents used for MRI scanning). Of course, indwelling metallic foreign objects, such as pacemakers or metal artifacts from previous surgery, may make CT the necessary choice instead of MRI. The advantages and limitations of echocardiography are shown in Figure 37–20. Both echocardiography and CT are required for complete evaluation of the entire aorta.

Figure 37-20. A and B: Limited distance above the aortic valve (AV) for which the ascending aorta (Ao) can be seen on transthoracic echocardiography. Schematic (A) and actual echocardiographic (B) image. AMVL, anterior mitral valve leaflet; IVS, interventricular septum; LA, left atrium; LV, left ventricle; LVPW, left ventricular posterior wall. C and D: Limited distance above the aortic valve for which the ascending aorta can be seen on transesophageal echocardiography. The tracheal air column interferes with visualization of the upper ascending aorta. Schematic (C) and actual echocardiographic image (D). RV, right ventricle. (A and C: Reproduced with permission from Mr. Rob Flewell. B and D: Reproduced with permission from Elefteriades JA, et al. Thoracic aortic aneurysm clinically pertinent controversies and uncertainties. J Am Coll Cardiol. 2010;55:841. Copyright © 2010 by the American College of Cardiology Foundation.)

5. Evaluation of Family Members

The data on familial inheritance has become strong enough that the treating physician is obligated to recommend that family members be evaluated. Physicians of family members should be made aware that aneurysm disease has been diagnosed in the family. A CT scan is recommended for adult males and for females beyond childbearing age. For children and for females of childbearing age, echocardiography of the ascending aorta and abdominal aorta is recommended. Investigators hope to identify humoral markers or genetic aberrations that can be used for familial screening of the aneurysm trait in the very near future. Routine genetic testing via Whole Exome Sequencing, while somewhat controversial and inconsistently approved by insurers, is becoming more and more widespread. We now offer and encourage screening of patients and family members by WES for nearly all of our patients. (See Figure 37–4, which presents, in pie diagram form, the most recent specific genetic findings from WES for thoracic aortic aneurysm.)

6. Activity Restrictions

Continuation of any and all aerobic activities, including running, swimming, and bicycling, is recommended. Serious weight lifters, at peaks of exertion, can elevate systolic arterial pressure to 300 mm Hg or higher. This type of instantaneous hypertension is, of course, not prudent for aneurysm patients. Weight lifters should limit themselves to one-half their body weight. The evidence for effort-induced aortic dissection is mounting. Participation in contact sports or those that might produce an abrupt physical impact, such as tackle football, snow skiing, water skiing, and horseback riding, is proscribed.

7. Role of Stent Grafting

A word of caution is appropriate concerning stent grafts. Stent therapy has become routine for many patients (especially those with abdominal or descending aortic aneurysms). It is certainly less invasive and more easily tolerated than open surgical techniques. Specialists, however, must avoid “irrational exuberance.” Owing to the high need for subsequent conventional surgery after abdominal aneurysm stent placement, the large, multicenter Eurostar study questioned the very efficacy and advisability of stent grafting. Endoleak, stent dislodgement, and aneurysm expansion or rupture were disturbingly widespread in medium-term follow-up. It should be remembered that stents were originally designed to keep tissue from encroaching on the vessel lumen, not to keep the vessel from expanding. One noted authority believes that the aneurysmal aorta essentially “ignores” the stent graft, dilating regardless of the stent, at its own pace (personal communication, Dr. L. Svensson). Also remember that the natural history of the thoracic aorta is that aneurysms grow slowly and that hard endpoints (rupture, dissection, and death) take years to be realized. For this reason, short-term stent studies are nearly meaningless. Long-term studies are needed. This newer modality of treatment should be approached with enthusiasm tempered by caution. Its advent should not at this point influence the decision about whether or not to intervene for a specific aneurysm; criteria (see earlier in chapter) should be met before any intervention, including stent therapy, is undertaken. Stent therapy appears especially well suited for patients with traumatic rupture of the descending aorta, in which setting a stent can be acutely life-saving. Those cautions having been stated, it is important to recognize that stent therapy is proliferating rapidly and that incremental improvements in the stent hardware are being made on a regular basis.