Acute Aortic Syndromes
Leon M. Ptaszek
Eric M. Isselbacher
Amy E. Spooner
Introduction
Representing the most lethal conditions affecting the aorta, acute aortic syndromes are associated with a high mortality rate if not recognized and treated promptly. Although the classical presentation of “aortic agony” is characterized by severe, sudden-onset pain in the chest or back [1], this presentation, although quite recognizable, occurs only in a minority of cases. As the initial manifestations of acute aortic syndromes are frequently variable, arriving at the appropriate diagnosis in a timely manner may be quite challenging. Prompt recognition of the acute aortic syndromes may be the difference between life and death for the afflicted patient. Frequently, the clinician must depend on subtle findings gleaned from history, detailed physical examination, and imaging in order to decide on an appropriate treatment plan. Here, we review the commonly encountered aortic syndromes, with a focus on aortic aneurysm rupture, as well as acute aortic dissection and acute aortic intramural hematoma (IMH). We focus primarily on the means by which these syndromes can be recognized and treated. Attention is also given to etiology and pathophysiology of the specific disease processes to the extent that evaluation of these processes is relevant to diagnostic and treatment strategies. Because patients with suspected acute aortic syndromes are frequently critically ill and require rapid disposition to treatment, we offer a unified evaluation and treatment algorithm. Each individual section serves as a guide to a syndrome-specific evaluation. Key features of a focused history and physical examination are emphasized. In addition, critical laboratory and imaging tests are reviewed.
Aortic Dissection
Definition and Classification
Dissection of the aortic wall involves longitudinal cleavage of the muscular media, leading to the formation of a second (or false) vessel lumen. The inciting event for a typical aortic dissection is thought to be a tear in the intima that leads to exposure of the underlying media, presumably weakened by medial degeneration. Once created, this cleavage front advances due to wall strain created by physiologic blood pressure. The cleavage front typically advances in the direction of blood flow, but dissection against the direction of flow is also observed [2].
There are multiple consequences of dissection. The native (or true) lumen is frequently compressed, leading to compromised downstream blood flow. The false lumen of the dissected aorta may also be less able to withstand physiologic blood pressure, due to changes in both its shape and its thinner external wall. The damaged aorta may therefore be more prone to rupture.
Aortic dissections are generally classified by location and extent. Dissections originate in the ascending aorta (65%) or in the descending aorta just distal to the origin of the left subclavian artery (20%). Dissection in the aortic arch (10%) and the abdominal aorta (5%) also occur [3]. Two classification systems for dissection location are in common use (Fig. 36.1). The DeBakey system includes three types of aortic dissection. Type I involves dissection of both the ascending and descending aorta, and/or the arch. Type II dissection involves only the ascending aorta proximal to the brachiocephalic artery, and type III involves only the descending aorta distal to the left subclavian artery [4]. The Stanford system includes two dissection types. All dissections involving the ascending aorta are included in type A: this includes types I and II in the DeBakey system. Stanford type B includes all dissections that do not involve the ascending aorta [5]. Classification of the location of a dissection carries prognostic and treatment importance. Surgery is indicated for dissection of the ascending aorta, whereas medical management is frequently the treatment of choice for descending dissection.
Chronicity of the dissection is defined as the time interval between onset of symptoms and evaluation. Dissections that are present for less than 2 weeks are defined as acute, whereas those that are present longer are defined as chronic [6]. It is noteworthy that the mortality associated with untreated ascending aortic dissection reaches 75% at 2 weeks [7].
Not all cases of aortic dissection are associated with an identified area of intimal tear. Several analyses have revealed that up to 13% of cases of apparent dissection turn out to be an IMH: a hemorrhage within the media that does not communicate with the intraluminal space [7,8,9]. In some cases, an atherosclerotic ulcer that penetrates from the intima beyond the internal elastic lamina is thought to precipitate intramural bleeding [10]. Classical aortic dissection and IMH are discussed separately later.
Classic Aortic Dissection
Epidemiology
Estimates of the incidence of aortic dissection range from 2 to 4 per 100,000 per year [11]. The highest incidence occurs in patients in their sixth and seventh decades of life. Incidence among men is double that for women [1,12]. Recent studies show that women tend to present later and with a more advanced disease state [12]. In addition, it has been shown that aortic dissection exhibits diurnal and seasonal rhythms. Dissections are most likely to occur in the morning or early afternoon, and more commonly in winter [13]. This seasonal difference does not appear to depend on climate [14].
Mortality rates associated with dissection are very high, and many patients do not survive to hospital admission. For those patients with aortic dissection who survive to admission, the early mortality rate is estimated to be as high at 1% per hour during the first day [7]. If left untreated, the associated mortality is estimated at 50% at 7 days and greater than 90% at 90 days [15]. Among patients who receive treatment, mortality during initial hospitalization ranges between 15% and 27.5%, as reported in several longitudinal studies [1,16,17].
Etiology and Pathophysiology
Any process that causes damage to the aortic tunica media, leading to medial degeneration, increases the risk for aneurysm or dissection. In the case of typical aortic dissection, the precipitating event is thought to be the creation of a tear in the intimal layer overlying a damaged area of the media. In the elderly patient with dissection, the presence of medial degeneration is correlated with the effects of aging, hypertension, and atherosclerotic disease [18,19,20]. Indeed, hypertension is found in 70% to 80% of patients with aortic dissection [1].
In the younger patient with aortic dissection, medial degeneration is still the culprit, but the constellation of correlated risk factors tends to differ [21]. Typically, young patients are more likely to have hereditary connective tissue disorders that compromise the integrity of the extracellular matrix in the tunica media, most notably Marfan syndrome, Ehlers–Danlos syndrome, bicuspid aortic valve, or familial thoracic aortic aneurysm syndrome (FTAAS) [1,6,21]. Young patients, defined in a recent study as being 40 years of age or younger, are also less likely to be hypertensive, and may have a larger aortic diameter on presentation. Paradoxically, mortality in this younger cohort does not appear to be lower than that in older patients [21]. All of these syndromes have been associated with breakdown of the fibrillin and collagen components of the extracellular matrix in the media, leading to medial degeneration. Aortic dissection risk is also increased in patients with Turner and Noonan syndromes [6]. Increased risk for dissection is found in a number of other conditions, including aortitis, especially in the context of giant cell arteritis and Takayasu arteritis [6,22,23]. Cocaine use has also been associated with dissection, ostensibly on the basis of increases in cardiac output, blood pressure, or as a consequence of direct vascular injury from cocaine itself (i.e., cocaine-induced vasculitis/endarteritis). In particular, crack cocaine has been identified as a potential precipitant of dissection [24,25].
As is the case for aortic aneurysm, the presence of certain structural abnormalities may be associated with an increased risk of dissection. In particular, a correlation has been described in patients with bicuspid aortic valve or, uncommonly, aortic coarctation. This association does not appear to be related to the hemodynamic effects of the abnormalities [26].
Notably, pregnancy is an independent risk factor for aortic dissection. The highest incidence of dissection is observed in the third trimester or early postpartum period. This risk is high particularly in pregnant women with a bicuspid aortic valve, Marfan, Ehlers–Danlos, or Turner syndrome [27,28]. In pregnant women with Turner syndrome, the risk of dissection or rupture exceeds 2%, and the risk of death is increased 100-fold [29]. Sporadic aortic dissections may occur in women without these predisposing conditions, possibly due to the elevated levels of relaxin and inhibin associated with pregnancy.
Iatrogenic injury to the aortic wall, sustained in the context of cardiac catheterization, intra-aortic balloon pump placement, or cardiac surgery, increases the risk of future aortic dissection [30,31,32]. Cardiac surgery involving the aortic valve appears to pose the greatest risk. Damage sustained by the aorta may take up to several years to develop into aneurysm and/or dissection [32,33].
Clinical Manifestations
There is no single physical examination finding that allows for positive identification of dissection: only imaging of the aorta verifies the diagnosis. Consequently, the initial evaluation and examination must incorporate a high index of suspicion and careful assessment.
The classic initial symptom of acute aortic dissection is severe chest or back pain. The severity of this pain is characteristically at its maximum at the point of inception. This is in sharp contrast with the typical crescendo onset of myocardial infarction pain [1]. The quality of the pain is often described as being “tearing” or “stabbing.” Acute pain is present in 85% of the patients described in the International Registry of Aortic Diseases (IRAD) and is present in up to 96% of patients described in other studies [1,6,36]. Of the patients in the IRAD registry, 90% described this discomfort as being the worst pain they ever experienced. Indeed, patients may be prone to writhing or pacing because of the pain. The initial location of the pain is correlated to the location of the dissection: of the patients in reported clinical series who presented with anterior chest or neck pain, 65% to 90% were found to have dissection of the ascending aorta. Interscapular or back may also represent dissection of the descending aorta [6]. On occasion, the patient may report a migration of the pain in association with extension of the dissection. In a series reported by Spittell et al., 17% of patients reported pain migration [6]. Recently, it was noted that aortic dissection may, in some instances, present with abdominal pain [37].
A common finding at the time of presentation is hypertension. Of the patients in the IRAD series, 36% of patients with type A dissection had elevated blood pressure, whereas 70% of patients with type B dissection had hypertension. Conversely, hypotension may also be a presenting feature of aortic dissection. This is a particularly ominous finding, as it likely represents developing shock. Hypotension is seen more frequently in patients with type A than type B dissection (25% vs. 4%, respectively) [1]. It is also noteworthy that patients with dissections who present with a “deadly triad” of hypotension/shock, an absence of pain, and evidence of branch vessel involvement exhibit a markedly higher mortality [38].
Evidence of heart failure, most notably pulmonary edema and hypotension, is found in up to 7% of patients with aortic dissection [1,6]. This finding is most frequently due to aortic regurgitation caused by a type A dissection [39]. However, in a recent report, a surprisingly high percentage of patients with heart failure at the time of dissection actually had a type B syndrome, with heart failure presumably due to myocardial ischemia or diastolic dysfunction with hypertension.
Syncope is present in up to 9% of patients with dissection. In these patients, syncope that is associated with focal neurologic signs is usually the result of occlusion of a branch vessel. Syncope in the absence of any other neurologic findings, present in up to 5% of dissection patients, likely represents aortic rupture into the pericardial space with tamponade. This finding portends rapid decline and requires emergent surgery.
Pericardial tamponade in the context of type A aortic dissection is a surgical emergency, as it represents a tenuously compensated rupture of the aorta. Unless the patient is in extremis, pericardiocentesis should not be performed, as the release of pressure in the pericardial space may precipitate a rise in blood pressure, recurrent hemorrhage into the pericardium, and cardiovascular collapse [40]. Dissection into the pleural space may also lead to hypotension and syncope, and similarly requires immediate surgical intervention.
A number of other vascular complications of aortic dissection may be apparent on initial evaluation. In up to 20% of the cases reported in the IRAD series, subjects presented with signs and symptoms consistent with occlusion of branch vessels. These occlusion events are typically the result of the extension of the dissection into a branch vessel (“static” occlusion), occlusion of the ostium of the vessel due to migration of the intimal flap (“dynamic” occlusion), or impaired flow in the true lumen due to distention of the false lumen. The spectrum of clinical findings associated with aortic side-branch involvement ranges from no signs and symptoms, to subtle findings, to florid manifestations, including severe ischemia of the affected territories. The mass effect of the dissection may lead to focal neurologic defects in rare cases. Involvement of a subclavian artery may lead to a difference in measured blood pressure between the two arms or pulse deficit. Impaired flow in the mesenteric arteries leads to signs and symptoms consistent with mesenteric ischemia. Dissections may also lead to occlusion of the renal arteries, leading to acute renal failure or renal infarction. Rarely, dissection leads to spinal artery occlusion with resultant paraparesis or paraplegia [1,6]. Lower limb ischemia may also occur in type B dissection [41].
On occasion, type A dissection may extend proximally to the ostia of the coronary arteries, leading to myocardial infarction. Three percent of the patients in the IRAD series presented with dissection-related myocardial infarction, with attendant chest pain and biomarker elevation [1].
There is not yet a specific biomarker in common clinical use that allows the clinician to confirm the diagnosis. For example, the D-dimer is elevated in dissection, but has limited diagnostic utility [42,43]. Recent work has highlighted several specific biomarkers that are elevated in acute aortic dissection and may become diagnostically useful in the future. The most promising assay is an enzyme-linked immunosorbent assay (ELISA) for myosin heavy chain. The sensitivity and specificity of this test, when it is performed within 12 hours of the acute event, are 90% and 97%, respectively. The primary advantage of this test is its ability to distinguish dissection from other events, such as myocardial infarction. Assays for other compounds elevated in aortic dissection but not in other acute cardiac events, such as serum heart-type fatty acid–binding protein, elastin, and calponin, are also in development [44,45,46,47].
Imaging
Prompt imaging is critical in the evaluation of suspected aortic dissection. Multiple modalities are at the disposal of the clinician; however, the patient is best served by the modality that offers adequate image quality without delay or transport time. The specific technique of choice may differ among hospitals, as not all facilities have the same capabilities. Following is a discussion of the relative strengths and weaknesses of the commonly available imaging techniques in the diagnosis of aortic dissection. The decision regarding the optimal technique to be used in a specific context is left to the individual clinician. Frequently, multiple imaging modalities must be used in a single patient. In addition, a single patient may require serial studies if his/her signs or symptoms evolve [48].
In most hospital settings, a chest x-ray (CXR) is performed as a matter of course in the evaluation of chest pain. The CXR, which is noninvasive, inexpensive, and routinely performed at the bedside, offers much useful information. In the patient with an aortic dissection, the CXR may reveal an abnormal aortic silhouette [1,6]. Widening of the mediastinum is a variable finding, observed in 15% to 60% of cases. Another suggestive finding is separation of intimal calcium, if present, from the soft-tissue border of the aorta. In addition, extravasation of blood into the pericardial space may be visualized as expanded and blunted heart borders. Pleural effusions are also easily visualized on CXR. Although useful, the CXR cannot be considered a definitive study. Therefore, other modalities should be used, notably echocardiography, computerized tomography (CT) scanning, and magnetic resonance imaging (MRI) (Table 36.1).
Transthoracic echocardiography (TTE) is a readily available, noninvasive, and portable imaging modality that may be considered. A focused study can be performed within 15 minutes at the bedside. Dissected segments of aorta can be measured directly: this is typically restricted to the ascending aorta, as neither the aortic arch nor the descending aorta can be reliably visualized via an external approach. TTE is also a very reliable technique for the visualization of pericardial effusion. The intimal flap of aortic dissection may be seen as a “double” aortic wall. Direction of Doppler flow may also help the clinician distinguish between the “true” and “false” lumens of aortic dissection. It should be noted that sensitivity for type A dissection varies between 70% and 90%, and sensitivity for type B dissection is approximately 40% [49]. Given this suboptimal sensitivity, performing a TTE should not delay a more sensitive imaging study. Despite its convenience, TTE is limited in that it does not offer an unobstructed view of all portions of the aorta. Body habitus may also adversely affect the quality of TTE images.
A far more accurate ultrasound study for suspected aortic dissection is transesophageal echocardiography (TEE). By virtue of the close proximity of the aorta to the ultrasound probe in the esophagus, this technique offers clear views of most portions of the thoracic aorta and affords excellent information regarding aortic valve function. TEE may be useful to guide surgical intervention for type A aortic dissection. TEE, like TTE, is portable and can be performed easily at the bedside, which makes it the procedure of choice for evaluation of critically ill or medically unstable patients who may be at higher risk during transportation for radiographic examinations.
In aortic dissection, TEE is superior to TTE in visualization of the intimal flap; sensitivity varies between 90% and 100%, and specificity is approximately 90%. Color Doppler imaging may identify the blood flow between the true and false lumens. Perhaps the most important procedural drawback regarding TEE is the need for conscious sedation, which may be difficult to administer in a patient who is hemodynamically unstable.
CT scanning allows for a full view of the entire aorta. Consequently, the sensitivity (90% to 100%) and specificity (90%) for visualization of the intimal flap in aortic dissection are comparable to TEE [49]. Specific CT techniques, such as spiral CT, also allow for facile three-dimensional reconstruction. The “double barrel” produced by dissection can be quite distinct. In classic aortic dissection, an intimal flap can be seen, separating a true and false lumen. Pericardial and pleural effusions may be easily visualized, but blood flow and tamponade physiology cannot be assessed directly. A diagnostic CT scan requires intravenous contrast, and care must be taken to address the risks of allergic reaction and contrast nephropathy. Many patients presenting with the acute aortic syndromes may also have renal insufficiency or failure; however, in the critically ill patient in whom aneurysm rupture is suspected, definitive diagnosis and treatment of the aortic process should take priority.
Table 36.1 Imaging Modalities for Patients with Suspected Acute Aortic Syndromes | ||||||||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
CT scanning and MRI share several of the same advantages, such as high image resolution and the ability to scan the entire aorta. Overall, the sensitivity and specificity of intimal flap detection by MRI are nearly 100% [49]. MRI does not require the use of IV contrast, which represents an advantage over CT scanning; however, MRI is more expensive and not as readily available or as rapidly performed as CT scanning. The primary limitation of MRI is lack of availability: not all hospitals have MR scanners available for emergent use. Even when available, issues of transporting a potentially unstable patient are still present. MRI is also contraindicated in patients in whom vascular clips, implantable cardioverter-defibrillators (ICDs) or pacemakers are present.
In the past, retrograde aortography was considered the gold-standard technique for aortic imaging. Because aortography is
an invasive test that requires the assembly of a catheterization laboratory team and the use of IV contrast and ionizing radiation, it is typically reserved for those cases where diagnostic uncertainty remains after one or more other imaging studies have been obtained. The ability of aortography to detect aortic dissection depends on the presence of blood flow between the true and false lumens; therefore, in cases where blood flow between these chambers is limited, the aortogram may be nondiagnostic. Overall, among patients with classic aortic dissection, the sensitivity and specificity for intimal flap visualization are 80% to 90% and 90% to 95%, respectively [49]. Aortography is still the study of choice for visualization of aortic branch vessels, which may not be visualized with other imaging modalities as well. In addition, aortography is particularly useful if endovascular treatment is contemplated.
an invasive test that requires the assembly of a catheterization laboratory team and the use of IV contrast and ionizing radiation, it is typically reserved for those cases where diagnostic uncertainty remains after one or more other imaging studies have been obtained. The ability of aortography to detect aortic dissection depends on the presence of blood flow between the true and false lumens; therefore, in cases where blood flow between these chambers is limited, the aortogram may be nondiagnostic. Overall, among patients with classic aortic dissection, the sensitivity and specificity for intimal flap visualization are 80% to 90% and 90% to 95%, respectively [49]. Aortography is still the study of choice for visualization of aortic branch vessels, which may not be visualized with other imaging modalities as well. In addition, aortography is particularly useful if endovascular treatment is contemplated.
Management
The primary goal of treatment in a patient with aortic dissection is to minimize the effects of the dissection while rapidly evaluating the patient’s candidacy for surgical repair, if indicated (Figure 36.5). Initial medical management while waiting for possible surgery should focus on management of pain, decrease of blood pressure to a minimum acceptable level, and decrease in the force of left ventricular contraction (dP/dt). In general, long-acting agents are not favored, as such agents are difficult to titrate rapidly. Early observation should occur in an intensive care setting, with an arterial line in place. For patients presenting with evidence of heart failure, pulmonary artery catheter placement may be considered, but is usually not necessary.
Pain management is titrated aggressively in patients with dissection. The goals of pain treatment are patient comfort and decrease in adrenergic tone. Narcotic analgesics are effective in rapid reduction of symptom severity, especially when administered in intravenous form. Long-acting oral formulations of narcotics are not recommended.
Blood pressure and dP/dt can be simultaneously decreased with a beta-blocker. Noncardioselective agents such as propranolol, labetalol, and esmolol have been used extensively in this context. Beta-blockers should be considered even in patients who are not hypertensive at presentation, as the reduction in dP/dt is thought to be beneficial in reducing the advancement of dissection. The goal heart rate is 60 beats per minute, and the goal systolic blood pressure is no higher than 120 mm Hg. In the event that a patient’s blood pressure is still elevated even after a goal heart rate has been reached with β-blockade, nitroprusside may also be administered as a constant intravenous infusion; however, intravenous nitroprusside should not be used without concomitant β-blockade, given the possibility of an increase in heart rate and dP/dt accompanying its potent vasodilatory effects.
Table 36.2 Commonly Used Medications with Routes/Doses | ||||||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
In the event that a beta-blocker cannot be used, due to contraindications such as bronchospasm, nondihydropyridine calcium channel blockers are the second-line agents. Verapamil and diltiazem, both of which have vasodilator and negative inotropic/chronotropic effects, may be used. Some patients have hypertension that is resistant to blockade of both β-adrenergic receptors and calcium channels. In this case, dosing of an intravenous angiotensin converting enzyme inhibitor, such as enalaprilat, may be indicated.
Hypotension may be seen in conjunction with dissection. It should be noted that the mode of blood pressure measurement should be scrutinized before changing a treatment plan; “pseudohypotension” may occur if dissection propagates into the limb in which blood pressure is being measured. In such cases, it is recommended that hypotension be verified by measurement of blood pressure in other limbs prior to discontinuation of beta-blockers or calcium channel blockers (Tables 36.2 and 36.3).
Surgical Intervention
The primary concept that relates to the optimal choice of therapy has not changed for nearly 30 years. In most cases, the location of the dissection determines whether the patient should undergo immediate surgery. Type A dissection is treated with surgery in virtually all cases, as the outcomes associated with surgical repair are superior to outcomes with medical management: 26% versus 50% mortality at 30 days in the IRAD series [1]. The one relative contraindication to attempted surgical repair of type A dissection is stroke in evolution, due to high risk of hemorrhagic transformation of the stroke during surgery [50]. In aggregate, survival of patients with acute type A dissection who are treated with surgical repair has improved over the last 25 years [51]. Aortic dissection repair is
complex surgery, and each patient’s medical comorbidities need to be addressed in detail before surgery as time allows. In the past, patients older than 80 were thought to have an operative survival rate too low to justify attempted repair. A recent multicenter study reported acceptable outcomes in aortic dissection repair performed in selected octogenarians. Although this study raises the possibility of aortic dissection repair in this age group, this approach remains controversial and each patient must be approached individually [52].
complex surgery, and each patient’s medical comorbidities need to be addressed in detail before surgery as time allows. In the past, patients older than 80 were thought to have an operative survival rate too low to justify attempted repair. A recent multicenter study reported acceptable outcomes in aortic dissection repair performed in selected octogenarians. Although this study raises the possibility of aortic dissection repair in this age group, this approach remains controversial and each patient must be approached individually [52].