CHAPTER 93: DISEASES OF THE AORTA

The aorta has complex biological and mechanical properties involving intrinsic relaxation and contraction that interact with left ventricular ejection to enhance hemodynamics. The major conductance vessel of the body, the aorta is an elastic artery with a trilaminar wall: the tunica intima, tunica media, and tunica adventitia (Fig. 93–1).^1,2 The innermost lining of the tunica intima is the endothelium, resting on a thin basal lamina. The subendothelial tissue is comprised of fibroblasts, collagen fibers, elastic fibers, and mucoid ground substance. An internal elastic membrane forms the outer lining of the tunica intima. The tunica media is approximately 1 mm thick, comprised of elastin, smooth muscle cells, collagen, and ground substance. The predominance of elastic fibers in the aortic wall and their arrangement as circumferential lamellae distinguish it from the smaller muscular arteries. A lamellar unit is made up of two concentric elastic lamellae containing smooth muscle cells, collagen, and ground substance.^3,4 The thoracic aorta incorporates 35 to 56 lamellar units and the abdominal aorta about 28 units.⁵ Surrounding the tunica media is the tunica adventitia, which is composed of loose connective tissue, including fibroblasts, relatively small amounts of collagen fibers, elastin, and ground substance. The adventitia strengthens the aorta and is essential to aortic surgeons for secure suturing of tissues. Within the tunica adventitia lie the nervi vasorum and vasa vasorum. The arteries arising along the course of the aorta give rise to the vasa vasorum, which develop into a capillary network supplying the adventitia and media of the thoracic aorta. The vasa vasorum do not supply the media of the abdominal aorta. Unlike the elastic fibers of the arterial wall, which are highly distensible, collagen is inelastic and provides the tensile strength required to prevent deformation and rupture of the aortic wall. Sophisticated biaxial strength testing of the human thoracic aorta has permitted new, detailed understanding of the tensile properties of the aorta (contributed by elastin and collagen) and permitted mechanical modeling of dissection and rupture events.⁶

The ascending aorta is approximately 3 cm in diameter, depending on age, sex, and body surface area. Among approximately 3500 individuals in the Multi-Ethnic Study of Atherosclerosis, the mean diameter of the ascending aorta was 3.2 ± 0.4 cm,^7,8 and in none of these normal individuals exceeded 5 cm. The diameter of the aortic arch is similar. Descending in the posterior mediastinum, the aorta tapers slightly to about 2 to 2.3 cm and the abdominal aorta narrows further to 1.7 to 1.9 cm in its distal portion. The aortas of males are larger than those of females. Aortic root dimension increases with age, height and weight, but the sex-based difference in aortic root dimension is not entirely explained by body surface area.⁵

The force of left ventricular (LV) ejection creates a pressure wave that traverses the aorta, producing radial expansion and contraction of the arterial walls.⁹ Potential energy derived from myocardial contraction and stored in the aortic wall during systole is transformed during diastolic recoil into kinetic energy, which drives blood into the peripheral vessels.¹⁰ The pressure wave is conducted at a velocity of approximately 5 m/s, increasing in amplitude along its course though the aorta.

Forward blood flow in the aorta begins when the aortic valve opens, and systolic pressure increases as the pressure wave courses along the length of the aorta. Flow velocity increases rapidly to a peak and then gradually decreases. With aortic valve closure, there is transient backward flow before forward flow resumes during diastole, particularly in the descending thoracic and abdominal aorta, albeit at considerably less than systolic velocity. The incisura in the arterial waveform in the proximal portion of the thoracic aorta gradually disappears and is usually absent in the abdominal aorta.¹¹

Each of the four components of the aortic wall—elastic tissue, collagen fibers, smooth muscle cells, and mucoid ground substance—change with age (Fig. 93–2).¹¹ Elastic fibers fragment, collagen content increases at the expense of smooth muscle cells, and glycosaminoglycans accumulate. As a result, the aorta becomes less distensible, reducing its capacity to absorb the forces derived from LV contraction.¹² The increased stiffness of the aortic wall with aging contributes to hypertension.¹³ Concurrently, weakening of the aortic wall leads to dilation of the lumen and elongation and uncoiling of the aortic arch, collectively producing ectasia. Accompanying these changes are alterations in aortic wall structure, fragmentation of elastic fibers, loss of smooth muscle cell nuclei (medionecrosis), accumulation of collagenous tissue, and deposition of basophilic ground substance.

When the aorta dilates due to disease of the medial layer, wall stress increases. This is accelerated by hypertension, particularly when pulsatile forces are high (see “Precipitation of Aortic Dissection” later in the chapter). When peripheral vascular resistance is high, the impact of the reflected pressure wave adds to that of antegrade pulsatile flow. The combination of inherited, degenerative, mechanical, and hemodynamic factors adversely affects the medial layer of the aortic wall, leading to dilation and setting the stage for the catastrophes of dissection or rupture.

The aorta can be affected by a variety of pathologic processes leading to aneurysm, dissection, or ischemic syndromes (Fig. 93–3A). The term aneurysm—derived from the Greek aneurysma, referring to dilation—is distinguished from ectasia, which refers to the modest generalized dilation and elongation of the aorta that occurs with aging.¹⁴ Although the size of the normal aorta varies with gender and body size, most agree that the maximum diameter of the thoracic aorta should not exceed 4 cm. For the abdominal aorta, which normally has a smaller diameter than the thoracic segments, it has been suggested that the term aneurysm be restricted to situations in which the diameter exceeds 3 cm.¹⁵ Another proposed definition depends on the affected segment having a diameter more than 1.5 to 2 times normal.¹⁶

Aneurysms may be classified according to morphology, location, or etiology. In true aneurysms, the wall of the aneurysm is composed of the normal histologic components of the aorta. A false aneurysm, on the other hand, represents a contained rupture in which the wall is a fibrous peel around a small perforation of the aorta, initially controlled by adherence of surrounding tissue and gradually enlarging over time.

A gross morphologic classification distinguishes true aneurysms as fusiform (most common) or saccular. A fusiform aneurysm is roughly cylindrical and affects the entire circumference of the aorta; a saccular aneurysm is an outpouching of only a portion of the aortic circumference. Frequently, a small neck provides continuity between the aortic lumen and the saccular aneurysm.

Aortic dissection refers to splitting of the medial layer of the aortic wall, permitting longitudinal propagation of blood between the components of the aortic wall (Fig. 93–3B). Aortic dissection is the most common cause of death due to human aortic disease.¹⁷ Atherosclerosis of the aorta can narrow the ostia of the great vessels or produce mobile atheromatous masses capable of embolism to the brain or other organs.

Several additional types of aortic pathology are illustrated in Fig. 93–4. Acute aortic transection, a consequence of trauma, involves localized disruption of an intrinsically normal aortic wall, which is resistant to propagating dissection. A ruptured aortic aneurysm is self-explanatory. Acute aortic dissection refers to separation of the layers of the aortic wall discussed below. Intramural hematoma represents a concentric, circumferentially oriented collection of thrombus in the aortic wall, without the discrete transluminal flap typical of aortic dissection. Penetrating aortic ulcer involves localized perforation of the medial layer of the aortic wall beneath an atherosclerotic plaque.

Aortic aneurysms may also be classified according to the segment involved as thoracic, thoracoabdominal, or abdominal. Aneurysm formation can involve a greater portion of the aorta (Fig. 93–5) than obstructive atherosclerotic disease, potentially affecting almost the entire vessel, while atherosclerotic obstruction tends to involve mainly the abdominal portion of the aorta when the iliac arteries are often affected as well. Dilation of the aorta may occur as a consequence of atherosclerosis, aging, infection, inflammation, trauma, congenital anomalies, medial degeneration, or combinations of pathologic states. These conditions may cause the aorta to thicken, thin, bulge, tear, rupture, narrow, dissect, or be altered by combinations of these conditions.

With normal aging, degenerative changes occur throughout the aorta, leading to a mild form of cystic medial necrosis, in which the normal lamellar structure of the media is locally destroyed, leaving lakes of amorphous material. Although essentially a normal physiologic process, cystic medial necrosis develops more rapidly in patients with bicuspid aortic valve, during pregnancy, and very markedly in people with Marfan syndrome. The mechanism by which the medial layer of the aorta is subject to this accelerated degeneration is a subject of active molecular genetic investigation. Severe elastic fiber degeneration, necrosis of muscle cells, and cystic spaces filled with mucoid material are most often encountered in the ascending aorta from just above the aortic valve to the brachiocephalic artery (see Fig. 93–5). Because of accompanying dilation of the aortic root, aortic regurgitation may be a secondary feature, although the valve leaflets themselves are histologically unaffected.

Cystic medial necrosis is the most common cause of ascending aortic aneurysm,¹⁷ and although this type of aortic pathology is typical of patients with the Marfan syndrome, it may also occur in the absence of other clinical stigmata.

Ascending Aortic Aneurysms

Ascending aortic aneurysms occur more frequently in men than in women, are typically fusiform, and usually extend into the aortic arch. Consequently, aneurysms of the aortic arch are often contiguous with aneurysms in the ascending aorta. Ascending aortic aneurysms are categorized according to the pattern of involvement of the aortic root (Fig. 93–6), with direct implications for surgical treatment. In the most common type, supracoronary aneurysm, the aortic annulus is normal in diameter, as is the short segment of aorta between the annulus and the coronary ostia. This type of aneurysm is treated by replacement with a tube graft, starting above the coronary arteries and extending cephalad to the end of the aneurysm. The second category, termed annuloaortic ectasia, is characterized by enlargement of both the aortic annulus and the proximal portion of the aorta. This type of aneurysm is typical of Marfan syndrome and related disorders characterized by cystic medial necrosis of the aortic wall.¹⁴ Because the annulus and proximal aorta are the most dilated portions, the aneurysmal ascending aorta has a “flask-like” shape, requiring replacement of the entire aortic root and valve, usually with a prefabricated composite graft that includes a prosthetic valve and aortic graft in an integrated unit. The third category, the tubular type of ascending aortic aneurysm, shares features of the other two configurations. In patients with tubular aortic aneurysms, the annulus and proximal aorta are mildly dilated, and the caliber of the ascending aorta is more uniform. When surgical repair is necessary, either a supracoronary tube graft or total aortic root replacement may be appropriate based on considerations that include the patient’s age. In younger individuals, composite grafting may confer greater protection against late dilation of the proximal portion of the aortic root, while in older patients more limited approaches may be advantageous.

Descending Thoracic Aortic Aneurysms

In contrast to the ascending aorta, the majority of aneurysms of the descending thoracic aorta are atherosclerotic.¹⁸ These are typically fusiform, often begin distal to the origin of the left subclavian artery, and may extend to the level of the abdominal aorta. Descending thoracic aneurysms may also develop in patients with aortic coarctation.

Abdominal Aortic Aneurysms

In more than 90% of cases, the superior margins of abdominal aneurysms are distal to the origins of the renal arteries. Atherosclerosis is present in the majority, although some authors suggest that atherosclerosis may be a secondary phenomenon in aneurysmal disease.

Thoracoabdominal Aortic Aneurysms

As the nomenclature indicates, thoracoabdominal aortic aneurysms have features of both thoracic and abdominal aortic aneurysms (Fig. 93–7). Although they constitute only approximately 3% of all aortic aneurysms, thoracoabdominal aneurysms are classified separately because of the diffuse and extensive nature of the aortic disease process (usually atherosclerosis) and special considerations for surgical repair, which may entail reimplantation of visceral arteries.¹⁹ Crawford and Coselli²⁰ delineated four types of thoracoabdominal aortic aneurysms according to the segment and extent of aorta involved, as described in Fig. 93–7.²⁰

The epidemiology of aortic aneurysms and the corresponding acute aortic syndromes present challenges because of confusion in terminology, referral bias, unknown incidence of asymptomatic cases, misdiagnosis of aortic rupture or dissection as myocardial infarction (MI), and limitations of administrative databases.^21,22,23 The Centers for Disease Control and Prevention list aneurysmal disease as the 19th most common cause of death and 15th most common in those older than 65 years.²²

The advent of routine postmortem computerized tomography in victims of out-of-hospital cardiac arrest brought to emergency departments (a growing trend in Europe and Japan) promises to improve understanding of the epidemiology of aortic dissection. These studies point to acute type A aortic dissection as the cause of as many as 8% of cases of out-of-hospital cardiac arrest, and suggest that type A aortic dissection is a more frequent cause of unexpected death than previously recognized.^24,25

Abdominal aortic aneurysm affects approximately 5% of individuals over the age of 65 years, and the prevalence is considerably higher in men than women.²³ The annual incidence of ruptured abdominal aortic aneurysm is approximately 10 per 100,000 population and is similar in men and women, though the age at diagnosis is a decade higher among women.^26,27 The rate increases to approximately 50 per 100,000 for men in the seventh decade of life, and each year more than one in 1000 men in the eighth decade experiences rupture of an abdominal aortic aneurysm.

The incidence of thoracic aortic aneurysm is also approximately 10 per 100,000 population and similar in women and men, but the age at diagnosis is a decade higher in women (70s). The yearly incidence of rupture or dissection of existing thoracic aortic aneurysms is approximately 7%, divided equally between these two types of events. A curious and as yet unexplained finding is that 79% of ruptures of thoracic aortic aneurysms occur in women.^27,28,29

Aortic aneurysm and its acute complications are recognized with increasing frequency. These trends during the past three decades have been confirmed by the Centers for Disease Control for the United States, Scotland, Sweden, the Netherlands, England, and Wales.^28,30 Clinical data suggest a true increase in incidence rather than more frequent diagnosis due to wider availability of aortic imaging.

Screening for Abdominal Aortic Aneurysm

The epidemiologic implications of screening for abdominal aortic aneurysms have been extensively studied.^23,31,32,33 Screening programs based on abdominal ultrasonography effectively detect aneurysms prior to rupture and reduce the frequency of aneurysm-related deaths. The cost of medical care per single aneurysm-related death prevented is about $10,000. One important benefit of screening is that a single negative ultrasonogram at 65 years of age suffices; patients without aneurysms at this age will not die of aneurysm rupture.^23,31,32,33

Cost-effective screening of the general population is focused on family history of aneurysmal disease, age, male sex, history of cigarette smoking, coronary artery disease, cerebrovascular disease, and hypercholesterolemia as factors linked with incremental risk of aneurysm development. Abdominal aortic aneurysm is five times more common in smokers, in whom 89% of all aneurysm ruptures occur.²³

The US Preventive Services Task Force recommends^31,33 one-time screening for abdominal aortic aneurysm by ultrasonography in men aged 65 to 75 years who have ever smoked. There is no recommendation for or against screening in men aged 65 to 75 years who have never smoked, and an explicit recommendation against routine screening for abdominal aortic aneurysm in women based on the relatively low yield. The Society for Vascular Surgery recommends one-time screening for all men older than 65 years (and at 55 years if family history is positive) and screening for women older than 65 years who have smoked or have a positive family history of aortic disease.³⁴

Clinical Markers of Aneurysm Disease

Aneurysms are often clinically silent until rupture or dissection occurs, and these catastrophes are most often fatal. Thus, detection of asymptomatic aneurysms is paramount. “Guilt by association” has been proposed as a strategic approach to aneurysm detection,³⁵ capitalizing on eight clinical correlates of thoracic aortic aneurysm: intracranial aneurysm, bovine aortic arch configuration, bicuspid aortic valve, renal cyst, abdominal aortic aneurysm, positive thumb-palm sign, family history, and temporal arteritis (Fig. 93–8). The association of intracranial or abdominal aneurysms with thoracic aortic aneurysms illustrates the coincidence of aneurysms in multiple vascular beds. Patients with thoracic aneurysms have an approximately 10% likelihood of concurrent intracranial aneurysm (higher for those with descending than ascending aneurysms),³⁶ probably as a consequence of common genetic and pathophysiologic mechanisms. Because of the prognostic significance and available catheter-based interventions to prevent rupture, routine screening for intracranial aneurysms may be warranted in patients with thoracic aortic aneurysms.

Bovine aortic arch and other arch anomalies like direct origin of the left vertebral artery from the aortic arch or aberrant right subclavian artery were previously considered benign variants, but these anomalies occur more frequently in patients with thoracic aortic aneurysm than in the general population.^37,38

The association of renal cysts and thoracic aortic aneurysm may reflect excess matrix metalloproteinase (MMP) activity.³⁹

The significance of family history of aortic aneurysmal disease, the thumb-palm sign, and arteritides are addressed below.

The emerging conclusion is that, if one of these “associates” or “markers” of thoracic aortic aneurysm is discovered in the history, physical exam, or imaging of a patient, the propensity toward a thoracic aortic aneurysm should be kept in mind.

Aortic Aneurysm as an Inherited Disease

Thoracic aortic aneurysms can be subdivided into two main categories: (1) syndromic, associated with abnormalities of other organ systems; and (2) nonsyndromic, with manifestations restricted to the aorta. Patients with syndromic thoracic aortic aneurysm may have Marfan syndrome, Loeys-Dietz syndrome, aneurysm-osteoarthritis syndrome, arterial tortuosity syndrome, Ehlers-Danlos syndrome, cutis laxa syndrome, or mutations affecting transforming growth factor beta (TGF-β). Nonsyndromic thoracic aortic aneurysm includes familial (affecting more than one family member) and sporadic forms.^40,41,42

Marfan Syndrome

This inherited disorder, described in 1896, is characterized by dolichostenomelia (long, thin extremities), ligamentous redundancy or laxity, ectopia lentis, ascending aortic dilation, and incompetence of the aortic or mitral valves (or both).⁴³ A large number of specific mutations (> 600) produce the Marfan phenotype, thus reducing the clinical usefulness of genetic testing in diagnosis.⁴⁴ Diagnosis is established largely on clinical grounds using the criteria summarized in Table 93–1.⁴⁵ The syndrome is linked to an autosomal dominant anomaly in the gene regulating synthesis of fibrillin type 1, a large glycoprotein that directs and orients elastin in the developing aorta.⁴⁶ Marfan syndrome was widely attributed to structural abnormalities and weaknesses of the aortic wall until dysregulation of TGF-β signaling was identified as a more reliable correlate.^47,48,49 TGF-β controls a panoply of cell functions. As fibrillin 1 binds TGF-β in an inactive complex, increased TGF-β signaling develops as a compensatory mechanism and plays a role in the pathogenesis of the degeneration of elastic fibers, accumulation of mucoid material within the medial layer of the aortic wall, and accelerated cystic medial necrosis that characterize this disease.

In patients with the Marfan syndrome, the aortic root tends to enlarge in fusiform fashion in association with aortic valve regurgitation, and about half of patients also have mitral insufficiency (Fig. 93–9). Aneurysms characteristically involve the sinuses of Valsalva and the tubular portion of the ascending aorta, producing annuloaortic ectasia (Fig. 93–10). Over time, abnormalities associated with Marfan syndrome typically affect the entire length of the aorta, although dissection most often involves the thoracic portion.⁵⁰

Ehlers-Danlos Syndrome

Ehlers-Danlos syndrome is an inherited connective tissue disorder, less frequent than Marfan syndrome, but also associated with aneurysm formation.^51,52 Although aneurysm formation and aortic dissection may occur in most types of Ehlers-Danlos syndrome, they are more frequent in the vascular type (type IV), which results from a mutation of the COL3A1 gene.⁵³

Loeys-Dietz Syndrome

Loeys-Dietz syndrome, which is caused by mutations in the TGFBR1 and TGFBR2 genes, is associated with aortic and arterial aneurysms that enlarge rapidly and are prone to rupture; in one series, the mean age at death was 26 years.^54,55

Bicuspid Aortic Valve

Patients with congenital bicuspid aortic valve have structural abnormalities of the ascending aorta predisposing to aneurysm and dissection.^56,57,58,59 Some have aortic coarctation as part of this syndrome as well. Aneurysmal enlargement of the aorta in patients with bicuspid aortic valve may occur independent of valvular stenosis or regurgitation and is not, therefore, predominantly a consequence of poststenotic dilation, although the aorta expands more rapidly when valvular dysfunction is present. Because bicuspid valves occur in ~2% of the population, the associated aortopathy is responsible for more cases of aortic dissection than the Marfan syndrome (which affects one in 10,000 people) (Table 93–2).

It is estimated that bicuspid valve disease accounts for more morbidity and mortality than all other congenital cardiac lesions combined.⁶⁰ Many authorities believe that this disease is so virulent that every individual with a bicuspid aortic valve will, given enough time, develop aortic stenosis, aortic insufficiency, or aortic aneurysm or dissection related to the bicuspid valve disease. Ascending aortic aneurysm in patients with bicuspid aortic valve is currently considered nearly as serious as ascending aortic aneurysm in patients with Marfan disease, and operation is often offered earlier than in nonbicuspid patients. Evidence now points to a congenital pattern of inheritance, and the fundamental embryology is being elucidated in animal models. The specific genetic mutations in humans with bicuspid valve have proven complex and multifactorial; abnormalities in the NOTCH1 gene have been documented, but only in some bicuspid patients.⁶¹ Evidence shows that excess MMP activity characterizes both the valve tissue and the aortic wall in bicuspid patients. The aorta grows more rapidly in bicuspid patients. Especially when there is associated valvular dysfunction, the current trend favors aggressive aortic resection in bicuspid patients with ascending aortic aneurysms (although specific criteria have varied in different editions of standard guidelines).

The histology and mechanical properties of the ascending aorta in patients with bicuspid aortic valves may predispose to dissection at smaller diameters than in patients with structurally normal (tricuspid) aortic valves.^62,63,64

Familial Thoracic Aortic Aneurysm

Many cases of thoracic aortic aneurysm once classified as atherosclerotic, hypertensive, or idiopathic are now traced to hereditary metabolic abnormalities affecting the aortic wall. There is a family history of aneurysmal disease in one of five patients with aortic aneurysm.⁶⁵ Probands with ascending aortic aneurysm are more likely to have family members with ascending aneurysm, and those with aneurysms involving the descending aorta are more likely to have family members with abdominal aneurysms.⁶⁶ Although there are several patterns of inheritance, autosomal dominance with reduced penetrance predominates (Fig. 93–11).

Genetic Testing for Thoracic Aortic Aneurysm

To date, as many as 21 genes have been linked with familial or syndromic thoracic aortic aneurysms and dissection (Fig. 93–12),⁶⁷ and discovery of additional candidate genes is likely. These encode molecules that regulate extracellular matrix (FBN1, FBN2, COL1A1, COL1A2, COL3A1), the cytoskeleton of smooth muscle cells (ACTA2, MYH11, MYLK), and the TGF-β signaling pathway (TGFβ2, TGFBR1, TGFBR2, SMAD3, SLC2A10). Most known mutations predispose to aortic aneurysm formation. However, there are certain genes that are either exclusively involved in aortic dissection (the MYLK gene⁶⁸), or cause aortic dissection at small aortic sizes (the ACTA2 gene^69,70). This near-exclusive dissection diathesis poses additional challenges in terms of counseling the patients regarding the most appropriate time for preventive surgical intervention, as aneurysm size is not useful in these small patient subgroups.

Certain genetic mutations influence the natural history of thoracic aortic disease, so the results of genetic testing may have implications for individualized management strategies in patients with thoracic aortic aneurysm. Genetic testing of patients with thoracic aortic disease can be accomplished using panels of genes or whole-exome sequencing. A potential benefit of routine whole-exome sequencing has been reported in patients with thoracic aortic aneurysm and dissection,⁶⁷ and genetic testing may lead to detection of asymptomatic aortic disease in the families of those with thoracic aortic disease, but the relative value of genetically based and imaging strategies for prevention of death related to aortic disease remains unsettled.

Molecular Pathophysiology

After genetic mutations establish the propensity for aneurysm development, lytic MMP enzymes degrade the structural proteins of the aortic wall, leading to aneurysm formation and dissection (Fig. 93–13). These enzymes are normally regulated by tissue inhibitors of metalloproteinases (TIMPs), which antagonize the lytic action of the MMPs. An uncertain proportion of patients with aortic aneurysms manifest excessive MMP activity, but such activity has been documented in cases of abdominal aortic aneurysm^71,72 as well as in cases of ascending aortic aneurysm and aortic dissection.⁷³ Several dozen specific MMP enzymes and certain TIMPs have been implicated in the pathogenesis of aneurysm disease.⁷⁴

In addition to proteolysis of the extracellular matrix, the pathogenesis of aortic aneurysms involves inflammation, cytokine activity, and loss of smooth muscle cells. Together, these processes undermine the integrity of the aortic wall and set the stage for aneurysm formation, growth, and rupture. Inflammatory cells participate in the genesis of aneurysms.^74,75,76 Smooth muscle cells of the aortic wall are charged with repairing damaged proteins, and a substantial (~75%) reduction in the number of smooth muscle cells in the walls of abdominal aortic aneurysms impairs this function. Aneurysmal smooth muscle cells display a threefold increase in apoptosis and impaired growth capacity. Figure 93–14 illustrates the interaction of proteolysis, inflammation, cytokine activation, and deficient smooth muscle cell function in the pathophysiology of aneurysm formation.

Many patients with ascending aortic aneurysms, which are typically nonatherosclerotic, display relatively little evidence of systemic atherosclerotic disease. A study comparing patients with ascending aortic aneurysm associated with annuloaortic ectasia or aortic dissection with age- and gender-matched control subjects found significantly less arterial calcification as detected by computed tomography (CT) imaging in those with aneurysms.⁷⁷ Subsequent follow-up investigations showed that patients with ascending aortic aneurysm have lower intimal-medial carotid artery wall thickness (IMT) than control subjects.⁷⁸ A related study found a low incidence of MI among patients with ascending aortic aneurysm.⁷⁹ Among the possible explanations for these observations are that mutations responsible for ascending aortic aneurysm protect against atherosclerosis, that the MMPs that lead to aortic aneurysm development have anti-atherogenic properties, or that the biophysics of atherogenesis is influenced by changes in pressure or flow dynamics resulting from aneurysmal disease.⁸⁰

Aortic Aneurysms

Unfortunately, the vast majority of thoracic and abdominal aortic aneurysms are clinically silent, with rupture constituting the first manifestation.⁸¹ A minority of patients (5%–10%) experience symptoms, permitting earlier detection of the aneurysm. Aneurysms of any kind can produce pain arising from stretching of the aortic tissue or impingement on adjacent structures. Pain originating in the ascending aorta is usually felt retrosternally. Pain from the descending aorta is characteristically located between the scapulae. Pain in the lateral or posterior chest may occur when the aneurysm compresses surrounding structures or erodes into adjacent ribs or vertebrae. Pain from the abdominal aorta may occur in the abdomen, left flank, or lower part of the back. It is often difficult to distinguish aneurysm pain from that resulting from other causes, but patients may be able to distinguish deep visceral pain from superficial or musculoskeletal pain. The interscapular location of pain caused by descending aortic enlargement or dissection is less often caused by musculoskeletal disease.

Rupture of the aorta in any location produces acute symptoms, usually severe pain followed by loss of consciousness or death as a result of internal hemorrhage. Rupture of an abdominal aortic aneurysm usually produces the clinical picture of extreme distress as a result of an abdominal catastrophe. Despite surgical advances, mortality is still the most frequent outcome because abrupt circulatory collapse prevents timely intervention except in unusual circumstances. Patients frequently have severe abdominal or back pain, but the pattern varies considerably. The aneurysm may rupture into the retroperitoneum or into the peritoneal or pleural cavities, leading to tachycardia, hypotension, diaphoresis, pallor, or shock, depending on the extent of rupture and associated blood loss into the extravascular space. On occasion, rupture occurs directly into the duodenum, causing an aortoduodenal fistula and acute gastrointestinal bleeding. This possibility should be considered when gastrointestinal bleeding is evident along with signs of an aneurysm on physical examination. Rupture may also occur into the inferior vena cava or iliac veins, producing an arteriovenous fistula; this is suggested by rapid development of leg swelling or high-output heart failure in the presence of an abdominal aortic aneurysm. Rupture of a descending or thoracoabdominal aortic aneurysm produces similar physiologic derangements, with the associated pain typically located higher in the trunk, consistent with the anatomical location of the disease process.

Aside from pain, ascending aortic aneurysms may produce heart failure on the basis of incompetence of the aortic valve. As the aortic root enlarges, diastolic coaptation of the aortic valve leaflets progressively falters, causing regurgitation of blood through the resultant central gap. Aneurysms of the sinuses of Valsalva may rupture directly into the right ventricular cavity, right atrium, or pulmonary artery, causing heart failure associated with a continuous murmur. Ascending aortic aneurysms can distort or obstruct the trachea, producing respiratory symptoms. Compression of the superior vena cava may produce venous congestion in the head, neck, and upper extremities. Aortic arch aneurysms or descending thoracic aortic aneurysms may produce hoarseness or dysphagia (“dysphagia lusoria”) from distortion of the recurrent laryngeal nerve or direct impingement on the esophagus. Descending thoracic aortic aneurysms may cause hemoptysis from direct erosion into the lung parenchyma or bronchi or hematemesis from esophageal erosion. Because mural thrombosis is so common in atherosclerotic aneurysms, they may be the source of peripheral embolism of thrombus or atherosclerotic debris causing occlusion of distal vessels or clinical features of atheroembolism. Occasionally, patients with abdominal aortic aneurysms may become aware of prominent abdominal pulsation. Nausea and vomiting may occur if an aneurysm compresses the duodenum. Compression of the left iliac vein may cause left leg swelling, compression of the left ureter may cause hydronephrosis, compression of the testicular veins may cause varicocele, or compression of the bladder may cause urinary frequency or urgency.

Thoracic or abdominal aortic aneurysms are most commonly diagnosed not on the basis of symptoms, but incidentally by imaging procedures carried out for another reason. Occasionally, chest radiographs obtained to evaluate pulmonary symptoms or for general screening suggest an aortic aneurysm, which may then be confirmed by CT or magnetic resonance imaging (MRI).

Aortic Dissection

Aortic dissection typically produces sudden intense pain, often described as tearing or shearing in quality. Whereas ascending aortic dissection usually causes anterior, substernal chest pain, dissection of the descending aorta causes posterior, interscapular pain. Pain may migrate inferiorly to the flank or pelvis as the dissection propagates distally. Impending aortic rupture should be considered when pain subsides and later recurs.⁷⁶

Painless dissection occurs in as many as 4% to 15% of patients^81,82,83,84 and is commonly detected as an asymptomatic finding on elective CT scans obtained for another purpose. By convention, aortic dissection is considered acute when identified within 2 weeks of onset and chronic when symptoms or other markers of dissection have been present longer.

Clinically, the inciting event responsible for aortic dissection is an intimal tear that permits blood to pass from the true lumen into the middle or outer layer of the aortic media, forming a second or false lumen separated by an intimal flap.⁸⁵ In some cases, intramural hematoma precedes perforation of the intima, possibly related to rupture of a vasa vasorum.⁸⁶ The dissection may propagate distally (antegrade) or proximally (retrograde) to narrow or occlude the origin of any branch artery arising from the aorta. Antegrade propagation is more common, and the dissection usually has spiral morphology, leaving some aortic branches supplied by the true lumen and others in continuity with the false lumen. Because any branch of the aorta can be affected, dissection can present as disease of any organ system—stroke, paraplegia, abdominal visceral catastrophe, renal failure, or lower limb ischemia. For this reason, aortic dissection has been given the well-earned moniker of “the great masquerader.”

The intimal tear originates in the ascending aorta in 65% of cases, transverse arch in 10%, upper descending aorta just beyond the origin of the left subclavian artery in 20%, and more distally in 5% of cases. The false lumen may terminate at any point along the length of the aorta or in the iliac or femoral arteries, and there are sometimes multiple flaps and several sites of reentry. The false lumen can undergo retrograde dissection, thrombotic occlusion, pseudoaneurysm formation, compression, or rupture.

Aortic dissection occurs more often in men than in women, with a 2:1 to 5:1 preponderance,⁸¹ usually in the sixth or seventh decade of life. Systemic hypertension is the major predisposing risk factor. In the International Registry of Aortic Dissection (IRAD) database, 74.4% of 2952 patients with proximal aortic dissections and 80.9% of 1476 with distal dissections had a history of hypertension.⁸⁷ Other predisposing conditions include Marfan syndrome, Ehlers-Danlos syndrome, Loeys-Dietz syndrome, bicuspid aortic valve, aortic aneurysm, and annuloaortic ectasia associated with cystic medial necrosis. The most common causes of aortic dissection in patients younger than 40 years old are Marfan syndrome and pregnancy. Iatrogenic trauma resulting from intravascular catheterization is the cause in approximately 5% of dissections and may involve any segment of the aorta. Cocaine abuse is increasingly recognized as a factor predisposing to acute aortic dissection, possibly mediated by acute hypertension.^88,89

Aortic dissection is often fatal unless recognized early and aggressively treated. Because the presenting symptoms and signs are myriad and nonspecific, dissection may be initially overlooked in up to 40% of cases. The diagnosis is first apparent at postmortem examination in a disturbingly large fraction of cases. The mortality associated with untreated aortic dissection approaches 1% to 2% per hour during the first 48 hours, reaching 90% at 3 months; most deaths related to dissection occur within 14 days after onset.^2,90 Up to 20% of victims die before reaching hospital.⁹⁰ With expert care, however, survival rates of more than 70% have been reported for patients reaching the hospital with acute aortic dissection. Most require intensive medical therapy, either as the sole treatment or in conjunction with surgical intervention. Even if the patient survives aortic dissection, expansion of the dissected portion of the aorta (dissecting aortic aneurysm) contributes to mortality beyond the acute phase.

Dissection may occur in the ascending (type A) or descending (type B) segments of the thoracic aorta, defined by the location of the inciting intimal tear. Tears in the ascending aorta generally occur 2 to 3 cm above the coronary ostia, and in the descending aorta typically originate 1 to 2 cm beyond the origin of the left subclavian artery. The first type produces ascending (type A) dissection and the second descending (type B) dissection, but dissections originating in the ascending aorta commonly extend along the aortic arch to involve the descending and abdominal portions of the aorta as well.

Aortic dissection can cause death in four main ways (Fig. 93–15): (1) hemopericardium with cardiac tamponade caused by retrograde dissection; (2) acute aortic valve insufficiency caused by retrograde dissection; (3) hemothorax caused by rupture of a descending aortic dissection into the pleural space; and (4) tissue ischemia caused by occlusion of a branch artery. Figure 93–16 illustrates the mechanism of branch occlusion and benefits of spontaneous, catheter-based, or surgical fenestration.

Most patients with aortic dissection present with hypertension, but 3% to 18% present with shock, sometimes secondary to extension of dissection into the coronary arteries, acute MI, LV failure, acute severe aortic insufficiency, cardiac tamponade, or aortic rupture. Coronary perfusion may be compromised by retrograde dissection, compression by the false lumen, or hypotension. In one series, differential pulse volume and blood pressure between the right and left upper extremities was detected in 38% of patients with ascending aortic dissection. An abrupt loss of pulse may affect the carotid, subclavian, axillary, radial, ulnar, or femoral arteries, and acute limb ischemia has been reported in 20% of patients. Branch vessel occlusion results from compression by the distended false lumen of the true lumen of the branch vessel.

Approximately 15% to 20% of patients with aortic dissection develop neurologic deficits, with transient cerebral ischemia or stroke in up to 10% of patients resulting from extension of dissection into the carotid or vertebral arteries. In such cases, brain imaging and neurologic or neurosurgical consultation may be helpful to determine whether cerebral infarction has occurred; surgery is best avoided in such cases for fear of inducing intracerebral bleeding or otherwise extending the zone of infarction. When cerebral infarction is absent or incomplete, urgent operation is generally indicated because repair of the dissection may restore brain perfusion. Similarly, urgent surgical intervention is indicated when interruption of spinal circulation by a dissection of the descending aorta threatens to cause paraplegia.

Aortic rupture is the most common cause of death in patients with aortic dissection. The second most common cause of death in these patients is acute, severe aortic regurgitation, which has been reported in 44% of patients with dissection of the ascending aorta and is poorly tolerated hemodynamically, compared with chronic aortic insufficiency; because sudden volume overload allows no time for LV adaptation, cardiogenic shock typically ensues.

Variants of Aortic Dissection: Intramural Hematoma and Penetrating Aortic Ulcer

Intramural aortic hematoma differs from typical dissection in that there is no flap delineating the true and false lumens, and the hematoma is located circumferentially around the aortic lumen, rather than obliquely (Fig. 93–17).⁹¹ Whether the intramural hematoma arises from a small intimal tear that is not radiographically detected or from a rupture of a vasa vasorum within the aortic wall remains controversial. The clinical course is variable; the hematoma may persist, resorb (returning the aorta to a normal appearance), leave an aneurysm with the possibility of rupture, or later result in dissection.⁹²

Penetrating aortic ulcer involves disruption of the internal elastic lamina and erosion of the medial layer of the aortic wall, resulting in local penetration at the site of an atherosclerotic plaque. This lesion may mimic or result in aortic dissection, pseudoaneurysm formation, intramural hematoma, or rupture (Figs. 93–17, 93–18, and 93–19).⁹³

It is important to recognize that intramural hematoma and penetrating aortic ulcer are diseases associated with advanced age, characteristically occurring in patients older than those with type B aortic dissection. In addition, although branch vessel occlusion is commonly associated with aortic dissection, this does not occur as a consequence of penetrating aortic ulcer or intramural hematoma.⁹²

The management of patients with penetrating aortic ulcer or intramural hematoma is controversial.⁹² When the descending aorta is involved, most authorities advocate medical management with pharmacologic therapy aimed at plaque stabilization and reducing systolic arterial wall stress. Others take a more aggressive stance, operating on all patients except very old and infirm ones, based on observational follow-up of unoperated patients, many of whom die of aortic rupture within 1 to 2 years.

For intramural hematoma involving the ascending aorta, there is more unanimity favoring immediate surgical intervention, although the Japanese literature challenges the need for routine surgery even in this anatomic location.⁹⁴ Penetrating aortic ulcers usually involve the descending aorta distal to the origin of the left subclavian artery, and for those associated with persistent pain, stent grafting is the treatment of choice. Figure 93–18 illustrates a dramatic case in which a penetrating ulcer ruptured through the posterior wall of the ascending aorta, mimicking “cryptogenic” pericardial effusion until surgical exploration made the diagnosis clear. Aortic intramural hematomas and penetrating ulcers rarely improve, and late surgery is commonly needed. Patients with penetrating ulcers may benefit from initial surgical management, but whether this applies to intramural hematomas is less clear (Fig. 93–20).⁹² Endovascular therapy is an alternative to medical or open surgical management of intramural hematoma and penetrating aortic ulcer, and the appropriate role for endovascular therapy is currently under investigation.

Ascending aortic aneurysms are usually not detected by physical examination unless they produce the characteristic diastolic blowing murmur of aortic regurgitation, which is heard best at the upper right or left sternal border at end-expiration with the patient seated and leaning forward. Descending aortic aneurysms are seldom detectable on physical examination, except in rare instances when they can be palpated through an attenuated chest wall. In contrast, the abdominal aorta is easily palpable in most patients.

Dissection of the aorta is typically associated with physical signs, including the murmur of aortic regurgitation or decrement or loss of peripheral pulses (most commonly one of the femorals). Of course, a substantial difference in blood pressure between the two arms in the appropriate clinical setting should raise suspicion of acute or chronic aortic dissection as well.

For evaluation of patients with known or suspected aortic disease, imaging studies should be selected to identify aneurysm formation and delineate zones of dissection or rupture. The challenge in the emergency department is to identify the relatively few patients with aortic dissection among the larger number presenting with signs and symptoms suggestive of acute MI—requiring continuous awareness of the three life-threatening causes of acute chest pain (MI, pulmonary thromboembolism, and aortic dissection) and judicious use of CT and echocardiographic imaging for diagnosis. Plain chest radiography is useful for screening for thoracic aortic disease. In the IRAD database abnormal chest radiography results are common in aortic dissection patients. The general contour of the ascending aorta, the aortic knob, and the stripe of the descending aorta can usually be identified well and evaluated even on a plain chest x-ray.⁸⁷

Multiple three-dimensional imaging modalities are applicable to patients with aortic aneurysmal disease, including transesophageal echocardiography (TEE), CT, and MRI, each of which offers high sensitivity and specificity. Although CT and MRI show the three-dimensional structure of the aorta along its entire course, TEE may not fully delineate the aortic arch⁹⁵ or the abdominal aorta; conversely, TEE provides more detailed information about the pericardium, aortic valve, and LV function. Both the ascending and descending portions of the aorta are well visualized by TEE. The abdominal aorta can be examined by surface ultrasonography or by CT or MRI. The primary advantage of TEE is that it can be rapidly performed in the emergency department while the patient receives intensive medical support, but adequate sedation is required to avoid hypertension, which could extend dissection or instigate rupture.

The primary criterion for diagnosis of aortic dissection by CT is demonstration of two contrast-filled lumens separated by an intimal flap.⁹⁶ Inaccuracy may result from inadequate contrast opacification, nonvisualization of the intimal flap, artifacts extending across the aortic lumen that simulate an intimal flap, misinterpretation of adjacent vessels or a prominent sinus of Valsalva as a flap, atelectasis or pleural thickening, or thrombosis of the false lumen. Modern multidetector-row CT scanners offer rapid image acquisition, ECG gating, variable-section thickness, three-dimensional rendering, diminished helical artifacts, and smaller contrast requirements, overcoming many of these pitfalls. The sensitivity and specificity of CT and MRI for diagnosis of aortic dissection currently approach 100%, and TEE does not lag far behind. Contrast aortography is more invasive and does not provide the three-dimensional anatomic information afforded by CT, MRI, or TEE imaging. Intravascular ultrasonography using high-frequency transducers, though not widely available, may be a useful adjunct when other imaging studies are not definitive. It can provide an accurate determination of the location and extent of aortic dissection and assessment of branch vessels.

What Is the True Aortic Size?

It is important to know the true size of the aorta because key decisions regarding management of aortic aneurysms depend on the size of the aneurysm. However, determining the true size of the aorta in a given modality or with multiple or discrepant modalities is not a trivial matter.

Multiple specific sources of error or confusion in aortic measurement are encountered regularly in the clinical practice of aortic care.⁹⁷

Inherent Level of Resolution of Current Imaging Technologies

Although aortic size is commonly read to the millimeter, the technology does not satisfy that level of precision; one can have confidence in a measured change of 3 or 4 mm, provided careful review is made to ascertain that similar levels of the aorta are being measured.

Include or Exclude the Aortic Wall Itself in the Measurement?

There is no consensus on whether the aortic wall should be included or excluded in the aortic diameter, whether the test is made by echocardiography, CT, or MRI. This may make a difference of several millimeters in the calculated measurements. On a noncontrast CT image, for example, it is most comfortable to measure the entire aortic wall, including the lumen and wall; on the other hand, on a contrast image, it may be more comfortable to measure the luminal shadow alone. Echocardiographic images may be measured for the internal or external diameter of the aortic wall.

Limitations of Specific Imaging Modalities

Echocardiography

Echocardiography images of the ascending aorta are often beautifully crisp. However, a transthoracic echocardiogram (TTE) can only visualize the proximal several centimeters of the ascending aorta, perhaps to just above the sinotubular junction in a patient with good echo windows (Fig. 93–21). Thus, a TTE will miss an aneurysm of the mid-portion of the ascending aorta. Also, if the aneurysm is predominantly in the mid-ascending aorta, the TTE will “see” only the beginning of the aneurysm and will, consequently, underestimate its size. Even TEE is limited by the interposed tracheal air column and can be “blinded” to the upper portion of the ascending aorta (Fig. 93–22).

Computed Tomography Scan

CT scan with axial images cannot properly evaluate the very proximal portion of the ascending aorta. The entire portion of aorta between the annulus and the coronary arteries usually represents only a centimeter or two of distance. Axial imaging planes may miss an aneurysmal dilatation in this location. Also, it can be exceedingly difficult to be certain if a given axial cut is above or below the aortic valve—a key distinction if one is to take a diameter measurement based on that cut (Fig. 93–23). Furthermore, it is naïve to believe that the plane of the aortic valve is confined to the plane of the axial images (perpendicular to the longitudinal axis of the body). Even the normal annulus may be skewed. Furthermore, as the ascending aorta elongates, the aortic valve plane is forced into a more vertical orientation, rendering even more difficult the assessment of size on axial images (Fig. 93–24). In modern CT scanners, reconstructions are done not only in axial, but also in sagittal and coronal planes, but the resolution in the nonaxial reconstructions is often insufficient to permit precise assessment of the aortic shape or diameter. Also, the aorta changes in size, depending on the phase of the cardiac cycle in which an image is acquired, potentially affecting diameter measurement by as much as 17.5%.⁹⁸

Gated Computed Tomography

Gating of CT imaging—synchronizing image acquisition with the cardiac cycle, as determined by the electrocardiogram (ECG)—mitigates motion artifact in aortic CT scan images and substantially improves diagnostic accuracy (Fig. 93-25).

Centerline Method of Measurement

The centerline method of analysis of CT images of the aorta enhances precision in assessment of aortic wall morphology and measurement of aortic diameter (Fig. 93–26).⁹⁹ Commercial software programs automatically set the viewing plane perpendicular to the long axis of the aorta. It is important to bear in mind, however, that centerline methods may lead to smaller measurements of aortic diameter than traditional observer-based image analysis, and most criteria for selection of patients for aortic surgery have not been based on centerline methods. There is reason for concern, therefore, that the adoption of automated, computerized methods of measurement may underestimate the risk of aneurysmal dissection and rupture.

MRI is an inherently multiplane modality that provides axial, sagittal, and coronal images. Depending on resolution, accurate measurements may be feasible, potentially overcoming some of the limitations of echocardiography or CT imaging related to imaging windows, planar projection, and vascular calcification.

Geometric Complexity

The human aorta is geometrically complex. The ascending aorta is not truly or totally vertical. The aortic arch poses an additional challenge because an axial image through the aortic arch will produce an oblong, rather than a circular, contour; measuring the long axis of this oblong contour is misleading because this does not represent the true aortic diameter. The same holds true for any other portion of the aorta that takes a sharp curve, as often happens at the peridiaphragmatic portion.

Furthermore, in comparing measurements over time, one can easily be misled by diameters taken from even slightly different longitudinal levels of the aorta. Also, even in a given cross-section, the aorta will not be a true circle. The measured girth may vary depending on which particular geometric diameter is selected. This can be especially true even on good-quality transesophageal echocardiograms, which do not “see” all diameters of the aortic annulus and aortic root; asymmetric dilatation of one sinus, for example, may escape detection.

In some cases, it may be difficult to reconcile discrepant studies or to achieve an ultimate measurement that encompasses the separate “realities” of different imaging modalities.

In comparing serial imaging studies, it is important to avoid the temptation to compare with the last prior image. The aneurysmal aorta grows slowly, about 1 mm per year on average.¹⁰⁰ Thus, looking at closely spaced images may miss even inexorable growth. Rather, it is imperative to compare to the very first available image, even if it means searching or sending for early studies.¹⁰¹

It is important to remember as well that size is not the only important imaging criterion; shape matters as well, especially loss of the normal “waist” of the aorta at the sinotubular junction (as seen in Fig. 93–27). Loss of this normal indentation is an indication of intrinsic aortic disease that should raise attention and stimulate concern about future aortic events.

Biomarkers in Aortic Disease

Thoracic aortic aneurysm is a virulent, potentially lethal disease. Also, it is a predominantly silent disease. This combination of circumstances cries out for discovery of biomarkers for this disease—blood tests that can detect aneurysm in the general population, monitor the progress of an aneurysm, and predict complications in patients known to be affected (Table 93–3).^102,103

D-dimer is a byproduct of fibrin degradation. D-dimer is a useful biomarker in acute aortic dissection. It is 99% sensitive in the detection of acute aortic dissection.¹⁰⁴ If the D-dimer is not elevated, the patient does not have aortic dissection. However, D-dimer is extremely nonspecific, being elevated in patients with pulmonary embolism and coronary thrombosis, essentially in any state in which thrombosis and thrombolysis proceed. The extent of D-dimer elevation reflects the longitudinal extent of aortic dissection, predicts the mortality of patients with aortic dissection, and differentiates dissection from MI. However, because D-dimer elevation occurs after the dissection, it is not useful as a predictor.

MMPs are known to be intimately involved in the pathogenesis of aortic aneurysm and dissection. Their use in the monitoring of aortic disease is in the earliest stages. It has been shown that MMP elevation signals recurrent blood flow in an aneurysm after endovascular therapy.^105,106

Inflammation and collagen degradation are also known to be involved in the pathogenesis of aortic aneurysms. Accordingly, inflammatory markers and indicators of collagen turnover are being investigated as potential biomarkers in aortic diseases. CD4+CD28- T cells and elastin peptide are being investigated, in those two categories, respectively. Smooth muscle heavy chain myosin has also shown some promise.

Measurement of aneurysm-related RNA signatures shows promise as a marker of aneurysm activity that may prove helpful for detection and monitoring.¹⁰⁷ Ideally, this RNA panel would detect aneurysms and predict rupture and dissection (Fig. 93–28). Another potential tool employs advanced glycation end products (AGE) and their soluble receptors (sRAGE), which may be involved in the pathogenesis of thoracic aortic aneurysm. Low levels of sRAGE and high levels of AGEs seem related to the risk of thoracic aortic aneurysm development and progression.¹⁰⁸ Other promising biomarkers under investigation, alone or in combination, include D-dimer, C-reactive protein, neutrophil/lymphocyte ratio, DAO (diamine oxidase), IL-6 (interleukin 6), TNF-α (tumor necrosis factor alpha), α-SMA (alpha smooth muscle actin), smMHC (smooth muscle myosin heavy chain), and ncRNAs (noncoding RNAs).^{109,110,111,112} However, most of these markers do not predict dissection, but rather, detect dissection after it has occurred.

The imaging studies discussed above provide specific information that may establish or exclude the diagnosis of aortic aneurysm or dissection. Aortic dissection may masquerade as other acute cardiovascular, pulmonary, musculoskeletal, neurologic, or gastrointestinal disorders, because it can produce symptoms related to almost any organ. The diagnosis is most strongly suggested by migratory chest and back pain of less than 24 hours in duration arising in a patient with a history of hypertension. Prompt diagnosis is essential because the clinical course may evolve rapidly to a catastrophic outcome unless appropriate therapy is instituted. In particular, aortic dissection should be considered in all patients presenting with chest pain without another obvious cause, and the thoracic aorta should be imaged routinely in such cases. The IRAD database has for over two decades described the presentation and differential diagnosis of aortic dissection and emphasized its protean potential clinical features.^87,113

Other conditions frequently confused with aortic dissection are musculoskeletal chest pain, mediastinal tumors, pericarditis, pleuritis, pneumothorax, pulmonary embolism, cholecystitis, ureteral colic, appendicitis, mesenteric ischemia, pyelonephritis, stroke, transient ischemic attack, and primary limb ischemia. Especially troublesome are patients with abdominal symptoms and signs who display no apparent abdominal cause; it is important to consider aortic dissection in this situation. Given the extensive differential diagnosis, prompt objective diagnostic testing is necessary when the possibility of aortic dissection is raised. Aortic dissection and ruptured thoracic or abdominal aneurysm should be considered in the differential diagnosis of chest, abdominal, or back pain.

• Obtain appropriate imaging to exclude acute aortic pathology (triple rule-out CT scan): 256-row multidetector CT angiography can effectively exclude the major forms of life-threatening thoracic pathology, including coronary atherosclerosis, pulmonary thromboembolism, and aortic aneurysm or dissection (including intramural hematoma and penetrating aortic ulcer), but the threshold at which this technology can be most cost effectively applied has not yet been established.¹¹⁴

• Use appropriate biomarkers indicating activation of the coagulation system: The D-dimer assay not only helps evaluate patients with suspected acute pulmonary embolism but, when results are negative, also effectively excludes acute aortic dissection. Thrombosis of the false lumen in cases of aortic dissection typically raises D-dimer levels substantially in the peripheral blood.¹¹⁵

• Examine the aorta carefully whenever a CT scan is obtained in search of other thoracic or abdominal pathology.

Studies of the natural history of thoracic aortic aneurysms using large databases involving thousands of patient-years of observation have established that thoracic aortic aneurysm is associated with approximately 50% mortality over 5 years, but not all deaths in these cohorts are directly attributable to aneurysm. On average, aneurysms of the aorta expand at a rate of about 0.1-0.2 cm annually, the descending aorta enlarging somewhat more rapidly than the ascending aorta. Such aneurysms usually develop over decades, so imaging is seldom necessary more often than once a year, and often only every 2 or 3 years, depending on the initial diameter and other factors. Rapid expansion of a thoracic aneurysm in the absence of aortic dissection is rare and usually reflects measurement error. One common source of error is oblique measurement of the aorta at the arch or near the diaphragm, where an ectatic aorta may take a sharp turn (Fig. 93–29). Current CT and magnetic resonance (MR) technology can correct for most of these problems in measurement. More important than frequent imaging is to compare the most recent images with the earliest image available for a given patient to avoid underestimation of total growth over time.^97,101

Indications for Surgery

Asymptomatic, Intact Thoracic Aortic Aneurysms

When viewed in terms of cumulative lifetime risk, the natural history of thoracic aortic aneurysms is characterized by an abrupt increment in the incidence of dissection or rupture at a maximum diameter of 6 cm for the ascending aorta; these events tend to occur at a larger dimension at the level of the descending aorta (Fig. 93–30). Yearly risks of rupture, dissection, or death reflect a stepwise increment in risk as the aorta expands, rising most dramatically to 14.1% at a dimension of 6 cm.^116,117,118 Based on this observation, patients without overwhelming comorbidities should undergo resection of thoracic aortic aneurysm before the maximum diameter reaches 5.5 to 6.0 cm (Table 93–4). For patients with Marfan syndrome or a family history of this disease, a criterion of 5 cm is usually applied because these patients are more prone to rupture or dissection.^116,117,118

There are limited data regarding the aortic diameter at which the risk of dissection is high enough to warrant operative intervention in patients who do not fulfill criteria for aortic valve replacement (AVR) because of severe aortic stenosis or aortic regurgitation. Specifically, it is not clear whether patients with bicuspid aortic valves should undergo surgical intervention for aortic repair at diameters smaller than recommended for patients with tricuspid aortic valves and ascending aortic aneurysms. In general, aortic dissection develops in patients with bicuspid aortic valves at younger ages than in those with tricuspid aortic valve,^119,120,121 but in some studies, the diameters of the aortic root or ascending aorta in patients presenting with acute type A dissection were larger in those with bicuspid than tricuspid valves.^119,120

Lacking evidence from prospective trials, surgical repair of aortic root or ascending aortic aneurysms in patients with bicuspid aortic valves is generally recommended at diameters ≥ 5.5 cm.¹²² Surgical intervention may be considered when the aortic diameter reaches ≥ 5.0 cm, when aortic expansion occurs at a rate in excess of 0.5 cm annually, or in those with a family history of aortic dissection.^122,123 It remains controversial whether the anatomical location at which enlargement is greatest (ie, at the level of the sinuses of Valsalva or the tubular portion of the ascending aorta) or body surface area should be considered in determining the threshold for surgical intervention.

Asymptomatic, Intact Thoracoabdominal Aortic Aneurysms

No randomized trials have addressed the size at which suprarenal, juxtarenal, or type IV thoracoabdominal aortic aneurysms should be repaired to prevent rupture. Because of the high risk of postoperative death, renal insufficiency, and other surgical complications, however, elective intervention is generally recommended for these aneurysms at slightly larger diameter than for infrarenal abdominal aortic aneurysms. In the absence of contraindications to surgery or comorbidities associated with a life expectancy less than 2 years, intervention is generally indicated in patients with suprarenal or type IV thoracoabdominal aortic aneurysms larger than 5.5 to 6.0 cm.

As for aortic aneurysms confined to the chest or abdomen, the dimensional criteria for surgical intervention apply only to asymptomatic aneurysms. Symptomatic aortic aneurysms at any level should be resected regardless of size. The development of symptoms frequently portends rupture and mandates surgical or endovascular repair.

Asymptomatic, Intact Abdominal Aortic Aneurysms

Multiple studies have addressed the appropriate size for resection of abdominal aortic aneurysms, including randomized trials in the United Kingdom and the United States^124,125 and meta-analyses (Table 93–5). Small aneurysms (< 4 cm in diameter) pose a low risk of rupture and should be managed with periodic surveillance for enlargement and symptoms. Abdominal aortic aneurysms between 4 and 5 cm in diameter are associated with a 1% per year risk of rupture, and decisions regarding surveillance or surgery should be individualized based on age, familial features, and an assessment of surgical risk. Aneurysms 5.5 cm in diameter or larger carry a substantial risk of rupture and should be repaired. As for thoracic aneurysms, the occurrence of symptoms supersedes diameter as a basis for intervention. In general, patients with infrarenal or juxtarenal abdominal aortic aneurysms measuring 4.0 to 5.4 cm in diameter should be monitored by ultrasonography or CT scans every 6 months to detect expansion. Repair may be beneficial in selected patients with aneurysms of 5.0 to 5.4 cm in diameter at this level. For patients with aneurysms smaller than 4.0 cm in diameter, monitoring by ultrasonography examination yearly is reasonable. Intervention is not recommended for patients with asymptomatic infrarenal or juxtarenal aneurysms smaller than 5.0 cm in diameter in men or 4.5 cm in women. A meta-analysis found that these criteria should be employed as well for endovascular treatment of abdominal aortic aneurysms, and that similar size indications for intervention should apply to both endovascular and surgical therapy.¹²⁶

Some Dissections Can Occur at Small Sizes

We have known for some years that dissections do occasionally occur at small aortic sizes. This can be seen in Fig. 93–31. A paper from IRAD noted that a good number of the aortic dissections in their registry occurred at sizes below our recommended intervention criterion of 5.5 cm¹²⁷ (Fig. 93–32). Insightfully, the IRAD authors recognized the dangers of amending size criteria for surgical intervention downward, and they did not recommend any such change for fear of promoting surgical harm in operating on small aortas.

The key to reconciling our observational studies that small aortas pose only low risk with the IRAD observation that a substantial number of dissections occur at small sizes lies in recognition of the “at-risk” denominator pool of patients. The observational studies (enumerated above in discussing evidence-based criteria for aortic resection) present danger rates for patients under observation with small aortas at Yale University; indeed, their risk of aortic events, although not zero, is low—too low to justify surgical intervention. For the IRAD patients with dissection, however, the dissectors were drawn from the general population at large. Whether the distribution of aortic sizes is normal or paranormal, the number of at-risk patients increases dramatically as one moves toward normal from the largest aortic sizes in the distribution curve (Fig. 93–33). In fact, it is likely that there are millions of patients in the United States with aortas between 4 and 5 cm. One could certainly cause harm by operating on all of them prophylactically. In fact, if one were to divide the numerator of observed dissections in IRAD by a denominator of millions, the observed yearly percentage rate of dissection in the small sizes would be very small.

Analysis of the normal distribution of aortic diameters in the MESA⁷ and IRAD database in relation to the risk of aortic dissection¹²⁷ revealed that although dissection can occur at smaller diameters, patients with aortas ≥ 4.5 cm are at 6000-fold greater risk of dissection than those with aortas ≤ 3.4 cm (Fig. 93–34).⁸ These studies support the current criteria for surgical intervention.

This posture is supported by longitudinal follow-up of at-risk patients triaged to medical management based on the size paradigms described above (no symptoms and aortic size < 5.5 cm). Not a single patient triaged to medical therapy died of rupture during follow-up (Fig. 93–35). On the other hand, among patients triaged algorithmically to the surgical arm who were too ill for surgery or who refused surgery, the rate of aortic events leading to death was indeed very high, validating surgical therapy for patients meeting the algorithm criterion.⁹⁷

MMPs and other pathophysiologic factors leading to degeneration of the aortic wall play an important role in aneurysm formation.⁷³ Mechanical studies of human aortas in vivo have shown loss of elasticity when expanded to a diameter of 6 cm, beyond which systolic forces cannot be dissipated by further aortic distension and instead accelerate disruption of the integrity of the aortic wall.¹²⁸ In addition to diameter, blood pressure plays a major role in determining wall tension. Without aortic enlargement, hypertension alone cannot generate sufficient wall stress to overcome the tensile strength of aortic tissue.

There are case reports of young, ostensibly healthy, athletic men experiencing aortic dissection during isometric weight training or other extreme exertion.¹²⁹ All had unknown moderate enlargement of the aorta, and all experienced acute onset of dissection pain during effort. Aortic dissection proved fatal in one-third of these cases. The magnitude of hypertension during weight lifting contributes excess wall stress that exceeds the tolerable aortic load limit, resulting in acute dissection. Nearly three-fourths of patients report either extreme emotion or exertion at the onset of symptomatic dissection, again suggesting a link with hemodynamic forces. Based on these considerations, some authors recommend that individuals embarking on programs involving heavy weight lifting or extreme physical training first undergo echocardiography to exclude ascending aortic aneurysm.

Based on these observations, the following sequence of events is postulated to cause aortic dissection (Fig. 93–36)⁹⁷:

• Genetic predisposition sets the stage for development of aortic aneurysm or dissection.

• Medial degeneration. Genetic predisposition results in activation of mechanisms of inflammation, injury to the medial layer of the aorta, loss of smooth muscle cells, and histologic damage by cytokines. As a result, the aortic wall is damaged and weakened.

• Aneurysm formation results from progressive dilation, increasing mechanical stress on the aortic wall.

• Hypertensive episode. At a moment of extreme exertion or emotion, an abrupt rise in blood pressure increases aortic wall stress beyond the tensile strength of aortic tissue.

• Aortic dissection. The aorta dissects (and/or ruptures).

These steps hinge on the assumption that dissection originates from an intimal tear, which in turn leads to splitting of the medial layer and creation of the false lumen. However, Humphrey et al.¹³⁰ proposed that a pooling of glycosaminoglycans and proteoglycans within the medial layer of the aorta may have a delaminating effect due to localized loss of tensile strength, local stress concentration, and increased swelling pressures—all of which promote dissection.^130,131 Dissection may start with an initial delamination process within the media, which propagates proximally or distally to create an intimal tear. Pooling of glycosaminoglycans and proteoglycans may be related to the dysregulation of TGF-β^131,132,133 based on the hypothesis that aortic smooth muscle cells increase production of glycosaminoglycans and proteoglycans in response to increased TGF-β activity.^130,131

The modern era in the treatment of aneurysms and dissections was ushered in by the pioneering innovations of Cooley and DeBakey, which were first reported in 1952.¹³⁴

Medical Treatment of Chronic Aneurysms

Patients with aortic aneurysms are commonly treated with β-adrenergic antagonist drugs to decrease systolic arterial wall stress, but the effectiveness of this strategy has been validated only in one long-term study of 70 patients with clinical features of Marfan syndrome.^135,136 The value of β-adrenergic antagonist drugs in patients with and without Marfan syndrome is controversial.^{137,138,139,140,141} Among the concerns, in addition to the potential for side effects, is evidence that β-blockers decrease the elasticity of the aortic wall. The overall effectiveness of β-blockers for thoracic aortic aneurysm has recently been reviewed.^139,140,141 As can be seen in Table 93–6, the evidence that β-blockers delay aortic dilatation in Marfan patients is quite equivocal and there is, further, simply no evidence of effectiveness in preventing the vital clinical end-points of aortic dissection or death (see Table 93–6).¹⁴¹

Angiotensin receptor blocking drugs (ARBs) represent a potential alternative to β-blockers in patients with Marfan syndrome,^142,143,144 based on groundbreaking work by Dietz and colleagues, which showed benefit of losartan in a mouse model and in infants with Marfan disease.^142,143,144 The concept was developed that ARBs (acting on the TGF-β pathway) can attenuate progression of aortic disease in Marfan patients. However, a randomized clinical trial comparing the losartan to atenolol showed no difference in rates of progressive aortic root dilation among children and young adults with Marfan syndrome,¹⁴⁵ but the issue remains unresolved in part because of concerns about dose selection.

The antibiotic agent doxycycline, an MMP inhibitor, has shown promise in large-scale clinical trials in patients with abdominal aortic aneurysms.⁷⁴ Other drugs that have undergone animal or clinical testing, including cyclooxygenase (COX) 2 inhibitory anti-inflammatory agents, hydroxymethylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins), the immunosuppressive agent rapamycin, and the angiotensin receptor inhibitor losartan,^146,147 have not been associated with consistent clinical benefit. A retrospective study in large numbers of patients with thoracic aortic aneurysm suggested substantial benefit in meaningful clinical end points from statin therapy for thoracic aortic aneurysm, most pronounced for descending and thoracoabdominal aneurysms.¹⁴⁸

Observational data from IRAD raise the possibility the β-blockers might be beneficial in some patients with type A aortic dissection and calcium-channel blockers might be useful in those with type B aortic dissection, but additional studies are needed to verify these preliminary findings before these agents can be recommended for routine clinical use in these situations.^149,150

Medical Treatment of Acute Aortic Syndromes (Dissection, Rupture, and Impending Rupture)

Aortic dissection propagates more vigorously when either blood pressure or the force of cardiac contraction is elevated.^2,151,152 Accordingly, control of blood pressure is an important aspect of treatment in patients with acute aortic syndromes, including dissection, rupture, or impending rupture. Intravenous nitroglycerin or sodium nitroprusside are commonly used for this purpose. Lowering blood pressure by administration of vasodilator drugs alone may increase the sheer stress on the aortic wall; however, this can be ameliorated by decreasing the force of cardiac contraction pharmacologically. The objective is to reduce the rate of rise of arterial pressure (dP/dt), reflected in attenuation of the upslope of the aortic pulse wave,¹⁵¹ by administering a short-acting β-blocking drug such as esmolol or the β- and α-adrenergic antagonist labetalol by intravenous infusion. When β-blocking drugs are contraindicated, the nondihydropyridine calcium channel blockers diltiazem and verapamil are reasonable alternatives. Table 93–7 presents specific drug options for the medical management of acute aortic syndrome.

Surgical Therapy

Specific surgical procedures have been developed for resection and graft replacement of aneurysms involving the ascending aorta, transverse (arch), descending thoracic, thoracoabdominal, and abdominal aortic segments. The Dacron grafts are durable and do not require long-term anticoagulant therapy. The risks of such surgery include bleeding, stroke, paralysis, MI, renal failure, death, and other complications; safety is greatest when surgery is performed electively before aortic rupture or dissection. Due to advances in surgical techniques (specifically myocardial and brain protection), graft materials (impervious), cardiopulmonary bypass, and anesthetic and postoperative care, these operations can be performed electively with outcomes at experienced centers rivaling those of coronary artery bypass and valve replacement surgery.¹⁵³

Ascending aortic (type A) dissection requires urgent surgery to avoid death as a consequence of intrapericardial rupture, aortic regurgitation, or MI. Repair involves reapproximation of the aortic wall between layers of Teflon felt. At experienced centers, survival is approximately 85%.¹⁵⁴ Patients with descending aortic (type B) dissections face a better prognosis with initial medical management (“anti-impulse” therapy with β-blockers and vasodilator drugs), withholding surgery unless specific complications develop.¹⁵⁵ Medical management of uncomplicated type B aortic dissections offers long-term survival comparable to that of the general population, whereas patients with an initially complicated course face unfavorable long-term outcomes (Fig. 93–37).¹⁵⁵

After corrective surgery for type A aortic dissection or stabilization on medication therapy for type B dissection, close observation is necessary, with serial aortic imaging during the first month and periodically thereafter. Some patients eventually develop enlargement of the dissected aorta requiring resection,¹⁵⁶ and in these cases, the dimensional criteria for surgical intervention are the same as those used for intact aneurysms.

Aortic Root

The morphology of the aortic root dictates the surgical procedure required for correction of aneurysmal pathology (see Fig. 93–6). For supracoronary aneurysms, tube graft replacement is typically used unless stenosis or insufficiency of the aortic valve requires concomitant valve replacement. In patients with annuloaortic ectasia, composite graft replacement of the aortic root and the aortic valve with a prefabricated unit is preferred. This more complex procedure requires meticulous reimplantation of the coronary arteries with “buttons” of surrounding aortic tissue. Yacoub and associates¹⁵⁷ and David and coworkers¹⁵⁸ have advocated techniques for “valve-sparing” aortic root replacement. Accumulating data suggest that patient survival and aortic valve function are generally well maintained following valve-sparing procedures in appropriately selected cases,¹⁵⁹ but caution is warranted because recurrent or progressive aortic insufficiency can occur.¹⁶⁰

Aortic Arch

Surgery for aneurysms of the aortic arch requires methods for cerebral protection during attachment of the cerebral arteries (great vessels) to the aortic graft. For attachment of the great vessels, Carrel patches (with a rim of aorta carrying the innominate, left carotid, and left subclavian arteries) are commonly used or, alternatively, prefabricated branched grafts that permit individual anastomosis of each great vessel. At 64.4°F (18°C), the temperature commonly used, the metabolic rate is below 15% of normal, permitting 30 to 45 minutes for manipulation of the aortic arch. When longer periods of circulatory arrest are necessary, direct cerebral perfusion can be provided through cannulae placed individually in the great vessels. An “elephant trunk” placed in the descending aorta facilitates subsequent procedures to address associated disease in the descending or thoracoabdominal aorta (Fig. 93–38).

Thoracoabdominal Aorta

The key issues in surgery of the thoracoabdominal aorta involve protection of the lower body organs and spinal cord during the period of aortic cross-clamping and attachment of the visceral arteries (superior mesenteric artery, celiac axis, and renal arteries). The arterial supply of the spinal cord is segmental, and viability of spinal cord cells is highly dependent on the artery of Adamkiewicz or arteries arising from the low intercostal or lumbar territory (T8-L2), which are excluded (at least transiently, and at times permanently) during thoracoabdominal aortic surgery. Intraoperative perfusion of the lower body by blood aspirated from the left atrium helps prevent paraplegia,^161,162 one of the most worrisome complications of this procedure. Other protective techniques include mild systemic hypothermia, aspiration of cerebrospinal fluid to decrease ambient pressure on the spinal cord, and reimplantation of the intercostal arteries.

The technical challenge of attaching all the important branch vessels of the abdominal aorta has been met by technical advances, most specifically by the implementation of reimplantation of the branch arteries on pedicles of surrounding aorta, often by the inclusion technique (in which the pedicle is not mobilized completely from its bed). Current rates of death or paraplegia for this type of surgery at experienced centers are below 10%, reflecting technical improvements in graft technology, surgical techniques, perfusion, anesthesia, and perioperative care.¹⁶³

Concomitant Coronary Artery Disease

The most important risk factor for cardiac events and death in patients undergoing abdominal aneurysm surgery is coronary artery disease. Because operative mortality is related mainly to myocardial ischemia, it has been suggested that coronary revascularization be performed in selected patients before abdominal aortic aneurysm resection.^164,165,166 No well-designed studies are available, however, comparing the outcome of serial myocardial revascularization and abdominal or thoracic aortic aneurysm repair with aneurysm surgery alone. Even so, an effort should be made to identify preoperatively patients at highest cardiac risk using noninvasive diagnostic methods. Findings of extensive myocardial ischemia may prompt angiography to define the coronary pathoanatomy and LV function. Thereafter, decisions regarding coronary revascularization must be based on symptoms, angiographic findings, and other elements of risk. One anticipates a survival advantage imparted by surgical revascularization in patients with left main coronary artery disease or stenosis greater than 70% involving each of the three major coronary arteries in patients with substantial zones of myocardial ischemia. In those with severe discrete proximal stenosis of a first-order coronary artery and symptomatic or extensive ischemia or reversible LV dysfunction amenable to catheter-based myocardial revascularization, clinical decision making must balance the long-term advantages of drug-eluting intracoronary stents against the need for potent platelet inhibitor therapy that may be difficult to administer in the days before and after aortic aneurysmectomy. Furthermore, in patients soon to undergo aortic reconstructive surgery, coronary balloon angioplasty or bare-metal stenting may be preferred, even though the potential for coronary restenosis may later require a second revascularization procedure after aortic surgery has been completed.

Diagnostic cardiac catheterization and angiography are routinely performed as part of the preoperative evaluation of patients undergoing surgical repair of ascending aortic aneurysms to evaluate the morphology of the ascending aorta, function of the aortic valve, and coronary anatomy. Based on the angiographic findings, myocardial revascularization can be readily incorporated into the reparative surgical procedure if warranted. For descending aortic aneurysms, if severe proximal coronary artery disease is found, then options include staged percutaneous or surgical coronary revascularization followed after an interval of several weeks by resection of the aneurysm. Results of such aggressive revascularization via a staged approach were recently reported with favorable outcomes.¹⁶⁷

Endovascular Therapy for Aortic Aneurysms

In 1991, Parodi and colleagues¹⁶⁸ first reported the technique of transfemoral catheter-based repair of infrarenal abdominal aortic aneurysms as an alternative to open repair for patients at high risk of complications with conventional surgical treatment. Over the ensuing years, a variety of stent grafts and delivery systems have been introduced, and endovascular repair has become a valuable alternative for elderly patients and those with advanced cardiopulmonary disease.

Most stent graft devices involve a modular construction consisting of a metallic exoskeleton surrounding an intimal fabric graft to maintain linear stability and prevent kinking.

Stent-graft therapy has a valid role in the management of patients with ruptured aortic aneurysm,¹⁶⁹ traumatic transection of the aorta,¹⁷⁰ and penetrating aortic ulcers.¹⁷¹ The role of endograft therapy in typical, fusiform degenerative aneurysms is more controversial. An important limiting complication of stent-graft therapy is the propensity for endoleak,¹⁷² in which blood flow continues into the excluded aneurysm sac after stent graft deployment. Endoleaks may lead to rupture of treated aneurysms.¹⁷³ Type I endoleaks are caused by incompetent proximal or distal attachments, producing high intrasac pressures that can lead to rupture and require repair using intraluminal extension cuffs or open surgery. Type II endoleaks resulting from retrograde flow from branch vessels (eg, intercostal, lumbar, or inferior mesenteric arteries) occur in as many as 40% of patients after endograft implantation. More than half of these seal spontaneously. Generally, type II endoleaks are followed with surveillance imaging and as long as the aortic sac does not enlarge, no treatment is necessary. However, if the aortic sac enlarges, type II endoleaks can usually be corrected by selective arterial embolization. Type III endoleaks are caused by fabric defects, tears, or disruption of modular graft components. These carry the same potential for aneurysm rupture as type I endoleaks and should be promptly repaired. Type IV endoleaks, the least common variety, result from graft porosity and diffuse leakage through interstices and usually develop within 30 days of implantation. Type IV endoleaks do not occur with the device technology that is currently used. Types I and III endoleaks seem the most dangerous and prompt repair is generally recommended.

Other complications of stent-graft repair of abdominal aortic aneurysms include occlusion of the iliac limbs of bifurcation endografts and migration from the proximal attachment as a result of progressive aortic expansion, which can be detected in at least one in five cases. Technical improvements in graft design and deployment techniques are reducing implant complications, but because of the potential for endoleak, graft migration, or graft limb occlusion, periodic follow-up imaging at intervals of 6 to 12 months is recommended after endovascular stent-graft repair.^172,174

The effectiveness of endograft therapy in preventing aneurysm growth, rupture, and aneurysm-related mortality compared with open surgical repair remains controversial.¹⁷⁵ Although often ultimately lethal, thoracic aortic aneurysm is an indolent disease, and even without therapy years may elapse between diagnosis of a small- or moderate-size aneurysm and aneurysm-related death. The EUROpean collaborators on Stent-graft Techniques for abdominal aortic Aneurysm Repair (EUROSTAR) survey of endograft repair of abdominal aortic aneurysms exposed cases of late death and rupture even after initially successful intervention. Information regarding late outcomes after endograft repair of abdominal aortic aneurysms is more comprehensive than for thoracic aortic aneurysm, but it appears that the risk of endoleak is ongoing over at least 5 years (Fig 93–39). Endoleak predicted rupture, surgical intervention, or death in 13%, 14%, and 27% of patients, respectively, by 5 years postprocedure in patients originally presenting with large aneurysms (Fig. 93–40). These concerns about long-term effectiveness supported the original view that endograft therapy should be reserved for patients in whom comorbidity precludes direct open surgical repair and emphasize the need for lifelong surveillance.^{176,177,178,179} However, more recent data support an even wider application of endograft therapy, although long-term survival is poor and has not improved much over the years.^180,181 Randomized trials of thoracic endografting versus open surgical repair will be required to provide conclusive evidence regarding the relative merits. Some midterm randomized studies are now coming to fruition:

1. EVAR-2 (Endovascular Aneurysm Repair Trials). EVAR-2 compared endovascular aneurysm repair (EVAR) with no specific therapy (“medical therapy”). The key finding from EVAR-2¹⁸² (presented in Fig. 93–41) shows no benefit from stent grafting of abdominal aortic aneurysm compared with no therapy. The all-cause mortality curves for EVAR and medical therapy are superimposable, as are the curves for aneurysm-related mortality. There is no evidence of benefit from EVAR in this trial.

2. DREAM (Dutch Randomized Endovascular Aneurysm Management). The DREAM trial compared EVAR for abdominal aortic aneurysm with traditional surgical therapy. The midterm follow-up in the DREAM trial found that at 2 years, the survival curves cross, and from that point on, stent treated patients have poorer survival than surgically treated patients. That is, EVAR shows an early advantage because surgery has some mortality, but this advantage is lost because EVAR did not show durable benefit.

3. INSTEAD (INvestigation of STEnt grafts in patients with type B Aortic Dissection). INSTEAD investigated stent therapy for patients doing well beyond 2 weeks after uncomplicated type B aortic dissection. The hope was that “tacking down” the dissection flap would lead to later benefit. Contrary to expectations, INSTEAD found early mortality and complications subsequent to stent therapy. There was no survival advantage over medical therapy alone.^183,184

4. INSTEAD XL. Later follow-up (up to 5 years) from the INSTEAD trial has recently been reported, under the moniker “INSTEAD XL.”¹⁸⁵ A retrospective “landmark analysis” was applied. This showed lower all-cause and aortic-related mortality and better aortic morphology in the stented group. Methodologic concerns have been raised (including the “landmark” analysis, which ignores all events occurring before the “landmark” time, and also the study’s being “underpowered” for mortality assessments).^186,187,188

5. IRAD. An analysis of the IRAD data also suggested better survival for type B dissection patients treated by TEVAR, but interpretation is difficult due to potential referral and selection biases.^189,190

More medium-term follow-up is needed to assess the durability of stent therapy. In the meantime, open surgical repair remains (debatably) the current standard in terms of both efficacy and durability, and it is essential that the same strict criteria for open surgery be applied in the decision to perform the less invasive endovascular therapies.

Pathologic Anatomy

Atherosclerosis of the aorta is common in Western society. The usual risk factors are tobacco smoking, diabetes mellitus, hypertension, hypercholesterolemia, obesity, and a sedentary lifestyle, and the contribution of elevated levels of plasma homocysteine and C-reactive protein has more recently been suggested.¹⁹¹ The pathogenesis of atherosclerosis is discussed in Chapter 32.

Atherosclerosis most commonly develops in the infrarenal aorta and may be asymptomatic or produce intermittent claudication, critical limb ischemia, or atheromatous embolism. Atherosclerosis typically also may involve the aortic arch and the origins of the brachiocephalic, carotid, and subclavian vessels, and the descending thoracic aorta. In this era, we usually detect atherosclerosis and calcification on CT scans done for other reasons, or on CT scans done deliberately as screening studies for calcium score. The thoracic aorta is so large that it is distinctly uncommon for calcified atherosclerotic lesions to cause stenosis of the aorta itself. Branch vessels (innominate, left carotid, and left subclavian arteries) are commonly involved and may produce brain or arm symptoms.

Atheromatous Embolism

Embolism of cholesterol-laden atheromatous material and thrombus from the surface of the aorta occurs commonly in patients with severe aortic atherosclerosis.¹⁹² Atheroembolism may be spontaneous, although it more frequently occurs after surgical or arteriographic manipulation, such as catheter-based coronary or peripheral interventions. In a retrospective study of 71 autopsies, the incidence of cholesterol embolism in patients who had undergone arteriography before death was 27% compared with 4.3% in an age- and disease-matched control group that did not undergo angiography.¹⁹³

Whether or not anticoagulant and thrombolytic drugs can exacerbate atheroembolism associated with the “blue-toe syndrome” is controversial.^194,195,196 Few cases of this complication were reported in clinical trials of anticoagulation in high-risk patients with atrial fibrillation despite the frequent finding of morphologically complex aortic plaque in this population.¹⁹⁷

Patients with atheromatous embolism typically have a history of angina pectoris, MI, transient ischemic attack, stroke, intermittent claudication, or peripheral gangrene. Clinical signs and symptoms are variable, depending on the amount, size, and location of origin of the atheromatous material as well as on the tissue affected. Whereas macroembolism may present catastrophically as an acute ischemic limb, patients with microembolism may have milder localized signs or a clinical picture suggesting systemic illness, including fever, weight loss, anorexia, myalgia, headache, nausea, vomiting, or diarrhea. Occasionally, the presentation may suggest vasculitis, infective endocarditis, or malignancy. Cutaneous manifestations are the most frequent findings¹⁹⁸ and include cyanotic toes, gangrenous digits, livedo reticularis, or nodules (Fig. 93–42).¹⁹⁹ When atheroembolism affects both lower extremities, the source is generally the aorta, but when only one extremity is involved, it may be difficult to determine whether the origin is the diseased ipsilateral iliofemoral artery or a more proximal or distal site.

Atheroembolism originating from the suprarenal aorta may involve the kidneys, producing occlusion of multiple small arteries and segmental ischemic atrophy. This small-vessel occlusive disease may cause accelerated hypertension, microscopic hematuria, or renal failure. Pathologically, biconvex cholesterol crystals occlude the interlobular and afferent arterioles (150-200 μm in diameter). A foreign-body reaction leads to small vessel occlusion, reducing glomerular filtration rate, activating the renin–angiotensin–aldosterone system, and accelerating hypertension. Various patterns of renal insufficiency may develop and progress over weeks or months to irreversible renal failure, requiring dialysis.²⁰⁰ The differential diagnosis includes renal artery stenosis, renal artery thrombosis, infective endocarditis, vasculitis such as polyarteritis nodosa, and other causes of acute renal failure. No single laboratory test is diagnostic because acceleration of the erythrocyte sedimentation rate, leukocytosis with eosinophilia, and anemia are common in many systemic illnesses.²⁰⁰ Blood urea nitrogen and creatinine elevations may be early manifestations of renal involvement, and the urine sediment may be abnormal. Elevated serum amylase or hepatic transaminase levels may indicate pancreatic or hepatic involvement, and creatine phosphokinase and aldolase arise from affected muscle. Renal biopsy is rarely required, but may reveal pathognomonic needle-shaped cholesterol clefts within small vessels. Atheroembolic renal disease carries a poor prognosis, with a mortality of 81% (179 of 221 patients) in one series; the most common causes of death were cardiac, renal, or multiorgan failure.²⁰⁰

Atherosclerosis of the Aortic Arch

Atherosclerosis of the thoracic aorta is a strong predictor of initial and recurrent stroke, coronary events, and death.²⁰¹ The thickness and morphology (protrusion, ulceration, or mobility) of atheromatous plaque correlate with the prevalence of stroke. Whether this association has a direct atheroembolic mechanism or reflects associated cerebrovascular pathology has not been conclusively determined.

Atheromatous embolism arising from the aortic arch or the carotid and vertebral arteries may cause stroke, transient ischemic attack, amaurosis fugax, blindness, headache, confusion, organic brain syndromes, dizziness, or spinal cord infarction. Retinal artery occlusion may be identified by Hollenhorst plaque visible on ophthalmoscopic examination as yellow, highly refractile atheromatous material at an arteriolar bifurcation.

Treatment

Because treatment of atheroembolism seldom reverses damage, emphasis is on prevention of subsequent ischemic events. When the source of embolism can be confirmed, it is often feasible to isolate or replace a discrete segment of the aorta by surgery, angioplasty and stent, or stent-graft insertion. Treatment should include symptomatic care of affected ischemic tissue and risk factor modification to prevent progression of atheromatous disease and promote plaque stabilization. If embolism affects the lower extremities, this involves local care of ischemic ulcers; when gangrene is present, amputation may be required. The role of sympathectomy is controversial, but this may be helpful when pain is intractable. In cases of renal atheroembolism, dialysis should be performed as necessary and blood pressure controlled pharmacologically.

Optimum antithrombotic therapy (anticoagulants, platelet inhibitors, or a combination of both) for atheromatous embolism has not been defined, but platelet inhibitor drugs generally lessen the risk of cardiovascular ischemic events. Anticoagulation therapy is far from protective, however, with a recurrence rate of cerebral events of 26% in patients with plaques more than 4 mm in thickness despite antiplatelet or warfarin therapy. Case reports of improvement with lipid-lowering therapy are supported by observations in nonrandomized series and evidence in coronary disease that HMG-CoA-reductase inhibitor (“statin”) drugs improve plaque stabilization.^202,203

The role of surgical therapy for patients with localized atheromata in the aortic arch and cerebral ischemia is controversial because it is seldom possible to determine conclusively whether the cause of symptoms is the aorta or associated cerebrovascular disease.^204,205

Aortoiliac Occlusive Disease

Atherosclerotic occlusive disease of the infrarenal aorta and iliac arteries may occur with or without atherosclerosis of the infrainguinal vessels.^206,207 When isolated, aortoiliac atherosclerosis typically occurs in younger individuals who smoke cigarettes. Almost half the cases are women, many of whom angiographically exhibit the “hypoplastic aortic syndrome,” with small-caliber aortic, iliac, and femoropopliteal arteries. The disease in these cases is usually confined to the aortic bifurcation. Disease localized to the distal aorta and common iliac arteries (type I) rarely produces limb-threatening ischemia because of extensive collateral vessels. The classic presentation is Leriche syndrome, a clinical triad of intermittent claudication involving the low back, buttocks, hip, or thigh, which is often mistaken for degenerative joint disease of the low back or hips, impotency (which occurs in 30%-50% of men with aortoiliac occlusive disease), or “global atrophy” of the lower extremities, reflecting the chronicity of low-grade ischemia.²⁰⁸ The femoral pulses are often weak or absent, but the ankle-brachial index may be normal at rest. A decline in ankle systolic pressure after exercise confirms hemodynamically significant stenosis.

Management

Treatment of patients with occlusive atherosclerosis of the aorta should include measures directed at improving symptoms (eg, claudication or limb ischemia) and thus quality of life, as well as reduction of overall cardiovascular risk. The latter involves the same measures as management of other manifestations of systemic atherosclerosis or peripheral artery disease (see Chaps. 32 and 96).²⁰⁹ Catheter-based interventions to relieve aortic obstruction should be considered in lifestyle-interfering claudication or critical limb ischemia.

Patients with aortoiliac occlusive disease should be instructed on a supervised exercise program. The Claudication: Exercise Versus Endoluminal Revascularization (CLEVER) trial was a randomized prospective trial comparing optimal medical therapy (OMT) alone, supervised exercise plus OMT, and stenting plus OMT. At 6 months and 18 months, those assigned to exercise responded as well as those who received stents.^210,211 If the patient is not satisfied with the results of an exercise program, revascularization should be performed. In the past, decisions on the type of treatment the patient was offered was based on the location and extent of obstruction as assessed by The Transatlantic Inter-Societal Consensus (TASC) II classification.²¹² However, the European Society of Cardiology and American College of Cardiology/American Heart Association guidelines now recommend an endovascular first approach in patients with aortoiliac occlusive disease, independent of the TASC classification.^213,214 The BelgianeItalian tRial Vascular Iliac StentS In the treatMent of TASC A, B, C, & D iliac lesiOns (BRAVISSIMO) reported a 24-month primary patency rate of 87.9% and a technical success of 100% in 325 patients with aortoiliac lesions.²¹⁵ Neither TASC category nor lesion length was predictive of restenosis. At the current time, open surgery is rarely performed for aortoiliac occlusive disease.

Etiology

Sudden occlusion of the terminal aorta may result from a large “saddle” embolus,²¹⁶ trauma, dissection, or in situ thrombosis superimposed on aneurysmal or atherosclerotic disease. Most emboli large enough to occlude the terminal aorta (saddle embolism) originate in the heart in patients with mitral stenosis, atrial fibrillation, acute anterior MI, infective endocarditis, or paradoxical embolism through a right-to-left intracardiac shunt from a peripheral venous source. When thrombotic occlusion of the aorta develops at a point of atherosclerotic narrowing, collateral perfusion is usually sufficient to prevent acute limb ischemia. Acute aortic occlusion related to thrombosis of an abdominal aortic aneurysm is considerably less common than thrombosis of popliteal aneurysms.

Clinical Features

Unlike gradually progressive obstruction, abrupt total or near-total interruption of flow through the terminal aorta or common iliac arteries poses an immediate threat to life and limb. Although the clinical picture varies depending on the collaterals, the full-blown syndrome is characterized by an abrupt onset of pain, typically severe, in the lumbar area, buttocks, perineum, abdomen, and legs. Diffuse cyanosis may be present from the umbilicus to the feet, and the lower limbs may be pale and cold. Numbness, paresthesia, and paralysis dominate the picture. Pulses are absent in the lower limbs and, unless circulation is restored promptly, muscle necrosis may produce myoglobinuria, renal failure, acidosis, hyperkalemia, and death.

Management

In contrast to chronic aortoiliac occlusion, acute aortic occlusion calls for immediate revascularization. The optimum procedure depends on the cause and the strategy for prevention of recurrent embolism. Transfemoral catheter-based embolectomy can extract even large amounts of embolic material from the distal aorta. Even after circulation has been restored, however, mortality is high, related to the underlying disease.

Inflammation of the aortic wall may occur in noninfectious diseases, such as Takayasu disease, giant-cell arteritis, IgG-related diseases, the spondyloarthropathies, Behçet syndrome, relapsing polychondritis, Cogan syndrome, rheumatoid arthritis, systemic lupus erythematosus, sarcoidosis, idiopathic retroperitoneal fibrosis, as isolated aortitis, and other disorders.^217,218 Only the most common of these uncommon entities are discussed in detail.

A possible role for infectious agents—including bacteria, viruses, and mycobacterium—in instigating arteritides of various sorts has been postulated.²¹⁹

Takayasu Disease

The prototypical nonspecific aortitis, Takayasu arteritis was named for the Japanese ophthalmologist who first called attention to the funduscopic findings.^220,221 Because of its predilection for the brachiocephalic vessels, this arteritis has been labeled pulseless disease and aortic arch syndrome. The classic form occurs with greatest frequency in Asian countries, but patients with a similar nonspecific aortitis are encountered worldwide. The etiology is unknown; no infectious agent has been identified, and identification of endothelial antibodies in 18 of 19 patients with this disease supports an autoimmune mechanism. This finding is nonspecific.

Recently, a genetic underpinning for Takayasu disease has been discovered, with mutations in the HLA-B and IL12B genes.^222,223 Continued progress in genetic understanding promises to clarify whether Takayasu and other, nonspecific arteritidies are one and the same or different diseases.

Histopathology

Histologic examination during active stages of the disease discloses a granulomatous arteritis similar to giant-cell arteritis and to the aortitis associated with the seronegative spondyloarthropathies and Cogan syndrome. In later stages, medial degeneration, fibrous scarring, intimal proliferation, and thrombosis result in narrowing of the vessel, yet there remains a lack of adequate histopathologic criteria for the differential diagnosis of noninfectious arteritides, including Takayasu disease and giant-cell aortitis.²²⁴ Aneurysm formation is less common than stenosis, but aneurysm rupture is an important cause of death in patients with Takayasu arteritis. Angiographically, the left subclavian artery is narrowed in approximately 90% of patients. The right subclavian artery, left carotid artery, and brachiocephalic trunk follow closely in frequency of stenosis. Thoracic aortic lesions occur in 66% of patients, the abdominal aorta is involved in 50%, and aortoiliac involvement is seen in approximately 12%. Pulmonary arteritis occurs in about half of patients and may be associated with pulmonary hypertension.

Clinical Features

In 70% to 80% of patients, clinical manifestations of the illness appear during the second or third decade of life, but onset in childhood and in middle life has been reported. Women are affected eight to nine times more often than men.²²⁵

During the early or “prepulseless” phase, symptoms include fever, night sweats, malaise, nausea, vomiting, weight loss, rash, arthralgia, and Raynaud phenomenon. Splenomegaly may occur, and laboratory findings may include acceleration of the erythrocyte sedimentation rate, elevated levels of C-reactive protein, anemia, and plasma protein abnormalities. However, it should be recognized that a minority of patients may not experience any of the signs or symptoms listed above and may present with discrepancy of arm blood pressures, absent pulse(s), the presence of a supraclavicular or cervical bruit, or incidental findings on imaging performed for another reason.

When arterial obstruction develops, upper-extremity claudication may occur as a consequence of subclavian artery stenosis. Stroke, transient cerebral ischemia, dizziness, or syncope usually indicate stenosis of the brachiocephalic arteries or subclavian steal. The retinopathy that first drew the attention of Takayasu is believed to result from retinal ischemia. Hypertension, observed in more than 50% of the cases, may be severe and occurs due to stenosis of the aorta proximal to the renal arteries or involvement of the renal arteries themselves.²²⁶

Cardiac manifestations result from severe hypertension, dilatation of the aortic root producing valvular insufficiency, or coronary artery stenosis (Fig. 93–43). Angina pectoris, MI, and heart failure have been reported. Clinical pericarditis has been observed infrequently, but healed pericarditis is often encountered at necropsy. Involvement of the visceral arteries may result in splanchnic ischemia, and aortoiliac obstruction may produce intermittent claudication in the lower limbs.

Patients in whom severe aortitis is evident at the time of diagnosis face a 25% to 30% risk of ischemic events or death over the next 5 years. Those without ischemic complications at presentation tend to fare better over 5 to 10 years. Severe hypertension and cardiac involvement are associated with a shortened life expectancy.

Diagnosis

The American College of Rheumatology has identified six major criteria for the diagnosis of Takayasu arteritis.²²⁷ Onset of illness by 40 years of age avoids overlap with giant-cell arteritis. Other criteria include upper-extremity claudication; diminished brachial pulses; greater than 10 mm Hg difference between systolic blood pressure in the arms; subclavian or aortic bruit; and narrowing of the aorta or a major branch. The presence of three of these six criteria carries high diagnostic accuracy.

Arteriography typically shows long areas of smooth narrowing interspersed with areas that appear normal. Aneurysm and occlusions are also common. Duplex ultrasound MR angiography or CT scans may show wall thickening resulting from inflammation and edema of the media and adventitia. MR or CT angiography of the entire aorta and iliac vessels is recommended for all patients with suspected Takayasu disease to define the extent of disease, identify aneurysms, and estimate the activity of disease. A “macaroni sign” on echocardiographic evaluation has recently been described, signifying the thick wall of the vessel with the small, linear residual lumen.²²⁸ Positron emission tomography (PET) scanning is showing promise for diagnosis by identifying “hot spots” at sites of active Takayasu lesions.²²⁹

Management

Corticosteroid therapy appears effective in suppressing inflammation during the active phase, and favorable results have been reported with immunosuppressive and cytotoxic agents.²³⁰ The mainstay of medical treatment continues to be administration of corticosteroids.²³⁰ However, the use of biologic targeted treatments (infiximab, rituximab, toclizumab) has been quite effective and safe in patients refractory to standard therapy or in those who are glucocorticoid dependent.²³¹ Operative treatment may be used to relieve symptoms caused by arterial obstruction, and percutaneous angioplasty and stenting are associated with mixed results. When performed for stenosis, the restenosis rate is substantial in both open and endovasculuar surgery.^232,233,234 These procedures are best reserved for patients in whom the acute inflammatory stage of the disease has been controlled.²¹⁷

Giant-Cell Arteritis

Giant-cell arteritis (temporal arteritis, polymyalgia rheumatica) involves extracranial arteries, including the aorta, in 10% to 13% of cases. A peak incidence late in life sets giant-cell arteritis apart from other nonspecific arteritides. Recent evidence suggests that giant-cell arteritis is occurring even later in life than in prior decades.²³⁵ Similar to Takayasu disease, giant-cell arteritis may produce narrowing of the brachiocephalic arteries, aneurysms of the ascending aorta, aortic dissection, and aortic regurgitation.²³⁶ Despite clinical, angiographic, and pathologic similarities to Takayasu arteritis, giant-cell arteritis almost always occurs in individuals older than 50 years of age. Although the most common presentation involves polymyalgia rheumatica with temporal arteritis, any large artery may be involved.

Treatment of giant-cell arteritis usually involves oral prednisone in an initial dose of 40 to 60 mg/d. Induction therapy with intravenous steroids has been found beneficial in a randomized trial. During clinical follow-up, CT or MR angiography or [18F]-fludeoxyglucose PET/MRI²³⁷ can be of value in monitoring improvement in arterial abnormalities with therapy. In unresponsive cases (< 10%) and in those who relapse as the dose is tapered, cytotoxic agents such as cyclosporine, azathioprine, and methotrexate may be helpful. One randomized, double-blind trial found a significant reduction in the rate of relapse and the cumulative mean dose of corticosteroid medication with methotrexate compared with placebo in corticosteroid-treated patients,²³⁸ but another study did not.^239,240

For patients with symptomatic ascending aortic involvement in giant-cell arteritis (pain, inflammation), we have, anecdotally, had excellent results with surgical aortic resection—often permitting dramatic reduction or elimination of corticosteroid treatment.

IgG4-Related Diseases

IgG-related diseases are a relatively recently recognized immune-mediated condition that may involve nearly every organ system.^241,242 Hallmarks of this disease are tumor-like swelling of the involved organs, dense lymphoplasmacitic infiltrate that contains IgG4-positive plasma cells, and storiform fibrosis. Usually there are elevated IgG levels in the serum. A number of conditions are considered part of IgG-related diseases, such as autoimmune pancreatitis, sclerosing mesenteritis, Mikulicz syndrome (salivary and lacrimal glands), Riedel thyroiditis, eosinophilic angiocentric fibrosis, Küttner tumor (submandibular glands), inflammatory pseudotumor, mediastinal fibrosis, and hypocomplementemic tubulointerstitial nephritis.²⁴² The vascular involvement includes IgG4-related retroperitoneal fibrosis, IgG4-related abdominal aortitis, and IgG4-related perianeurysmal fibrosis or inflammatory aortic aneurysms.²⁴³ It is now recognized that IgG4-related disease may be the cause of idiopathic retroperitoneal fibrosis (previously called Ormonds disease) in up to two-thirds of cases.²⁴⁴

Presentations of periaortitis may be nonspecific, leading to a delay in the diagnosis. The most common symptoms are pain in the lower abdomen, flanks, and back. The aorta may be extremely tender to palpation. The thoracic aorta may also be involved leading to aneurysm or dissection.²⁴⁵ Although Takayasu arteritis and giant-cell arteritis affect the primary aortic branches, IgG4-related disease generally spares the branches off the aortic arch.²⁴²

It is important to diagnose IgG4-related diseases early, as effective therapy is available. Glucocorticoids are first-line agents. There are no studies that have evaluated the usual steroid-sparing agents (methotrexate, azathioprine). There are no randomized trials using B-cell depletion therapy (targeting a subset of plasma cells that produce the IgG4) with agents such as rituximab, but this agent seems to be quite effective in treating some patients with IgG4-related diseases who do not respond to glucocorticoids.²⁴²

HLA-B27–Associated Spondyloarthropathies

Aortitis is present in a substantial portion of patients with ankylosing spondylitis and Reiter syndrome; more than 90% have the histocompatibility antigen HLA-B27. Aortic involvement is most common in those with spondylitis of long duration, peripheral joint complaints in addition to spondylitis, and iritis.²⁴⁶ Inflammation of the aortic root and surrounding tissues, manifest by aortic valve regurgitation or cardiac conduction abnormalities in patients with the HLA-B27 histocompatibility antigen, may also occur without spondyloarthropathies. Although the majority of cases occur in the aortic root and ascending aorta, occasionally isolated abdominal aortitis may occur.²⁴⁷ Histologically, the aortic lesion in this setting resembles the inflammation seen in syphilis, with focal destruction of medial elastic tissue and thickening of the intima and adventitia. Aortic dissection has been reported.²⁴⁸

Infectious Aortitis

Primary infection of the aortic wall is a rare cause of aortic aneurysms, which are more often saccular than fusiform. Infectious or “mycotic” aneurysms may arise secondarily from an infection occurring in a preexisting aneurysm of another cause. Staphylococcus, Salmonella, and Pseudomonas species are the most frequent pathogens causing primary aortic infections.²⁴⁸ Many cases arise as complications of infective endocarditis or arterial catheterization. An intrinsically abnormal aorta, however, may become infected as a consequence of bacteremia. Such infection produces suppurative aortitis, leading to weakness of a portion of the aortic wall. In these cases, aneurysms are typically saccular, yet there is a comparatively high propensity to rupture. Infection of the aorta related to endovascular stents is a new and increasing clinical entity that is difficult to treat.^49,250

Syphilitic Aortitis

Treponemal infection produces chronic aortitis in approximately 10% of patients with untreated tertiary syphilis and is the primary cause of death in about the same proportion of cases, but there is evidence of the process at autopsy in about half of patients who have had untreated syphilis for more than 10 years.²⁵¹ During the spirochetemic phase of primary syphilis, Treponema pallidum organisms lodge in the adventitia of the vasa vasorum and initiate an inflammatory response characterized by perivascular lymphocytic and plasma cell infiltrate. This is followed by obliterative endarteritis, resulting in patchy medial necrosis, elastic fiber fragmentation, weakening of the aortic wall, and aneurysm formation. The intima of the aorta has a characteristic wrinkled appearance, frequently with superimposed atherosclerotic plaques. Because the infection is seeded through the vasa vasorum, the process is most severe in the ascending aorta and the arch, where the density of these vessels is greatest. Luetic aneurysms are typically saccular and involve the ascending aorta whether or not the transverse and descending portions are also affected. Aortic aneurysms resulting from cardiovascular syphilis follow interruption of the elastic fibers as a result of periaortitis and mesoaortitis, which thicken, but weaken, the aortic wall. Rupture is the major complication, but the enlarging aneurysm may also compress or erode adjacent structures of the mediastinum. Because the inflammatory process tends to interrupt the medial layer by transverse scars, dissection is distinctly uncommon.

Aortic involvement may be asymptomatic or associated with aortic regurgitation, coronary ostial stenosis, or aortic aneurysm. Asymptomatic aortitis may sometimes be identified by linear calcification of the ascending aorta, evident on chest radiographs. Valvular regurgitation, present in 20% to 30% of patients with syphilitic aortitis, is mainly a consequence of aortic root dilatation. Syphilitic coronary ostial stenosis, only a century ago more common than coronary atherosclerosis as a cause of angina pectoris, occurs in 25% to 30% of such patients, most of whom also have aortic regurgitation. MI is rare. The least frequent manifestation of syphilitic aortitis is aneurysm formation, which occurs in 5% to 10% of affected patients. Although the prognosis for patients with uncomplicated syphilitic aortitis is comparable to that of the general population, the outlook is poor when syphilitic aneurysms of the aorta are large enough to produce symptoms. The diagnosis of cardiovascular syphilis may be difficult in patients older than 50 years of age, when hypertensive and atherosclerotic disease often coexist.

The frequency of cardiovascular syphilis has fallen dramatically over recent decades as a consequence of early identification and treatment of the disease. However, a recent report indicates that syphilitic ascending aortic aneurysm is still surprisingly common and must remain in the consciousness of the caring medical and surgical teams.²⁵²

Adequate antimicrobial therapy of early syphilis is the most important preventive measure, although whether such treatment retards the progression of disease once aortitis has developed has not been clearly established. Without surgical intervention, symptomatic syphilitic aortic aneurysms are associated with a high mortality.

Recently, Roberts and colleagues²⁵³ described the features of syphilitic aortitis at surgery in the hope of permitting intraoperative identification and prompt institution of treatment. The distinguishing features include sparing of the aortic root, involvement of the tubular portion of the ascending aorta, uniform involvement of all of the surface area of the affected aortic portions, and inflammation in all three layers of the aortic wall. The aorta is said to have a “tree bark” appearance characteristic of syphilitic aortitis.

Tuberculous Aortitis

Tuberculous aneurysms usually result from direct extension of infection from hilar lymph nodes and subsequent granulomatous destruction of the medial layer, leading to loss of aortic wall elasticity. The posterior or posterolateral aortic wall is usually the site of saccular aneurysm formation in these cases. Caseating granulomatous lesions affecting the medial layer of the aortic wall characterize the histology. Pseudoaneurysm formation,²⁵⁴ perforation, or aortoenteric fistula may result. Infection may occasionally invade the aortic valve ring and adjacent structures, producing a caseating paravalvular abscess. Rupture of tuberculous aortic lesions may occur. It is important to recognize that miliary tuberculous may manifest in a manner very similar to Takayasu arteritis. Therefore, if there is any suspicion that tuberculosis is present, testing should be performed before starting immunosuppressant medications.

This is an exciting time in the care of the thoracic aorta. Although clinical trials of ARBs have not revealed the hoped-for medical panacea, other exciting advances are being made.

As the familial nature of aneurysmal diseases becomes familiar to a broader array of physicians, radiologic testing of relatives of affected individuals is likely to become standard practice and reimbursable by insurers. Recognizing an aneurysm in advance of symptoms is the optimal method to enhance survival, and family members represent the most cost-effective candidates for diagnostic testing.

Surgical therapy for the thoracic aorta has become safer and safer, approaching the low mortality of conventional cardiac surgical procedures, like coronary artery bypass and valve replacement. This trend is likely to continue, becoming more widespread and proliferating beyond leading large-volume centers.

Endovascular devices and technologies are advancing at a rapid rate, providing alternative options, especially for elderly and infirm patients. It is hoped that these endovascular therapies will advance further, perhaps even mimicking the success and durability of open techniques.

The aorta is being understood more and more thoroughly from an engineering standpoint. Engineering calculations are currently on the way towards becoming an integral part of surgical decision making.²⁵⁵

Once a thoracic aortic aneurysm has been identified, we can keep the patient safe by following algorithms based on patient symptoms and aortic size. Perhaps the biggest problem in thoracic aortic disease is the silent nature of thoracic aortic aneurysms. Via a “guilt by associaton” paradigm, we can identify more and more asymptomatic patients with thoracic aortic aneuryms.

The most significant advances in aneurysm disease of recent years have been on the genetic front. We are starting to understand thoracic aortic aneurysm at the genetic level. Now that whole-exome sequencing has become a relatively inexpensive clinical reality, our genetic understanding of these diseases will advance by leaps and bounds. New genes and new variants in known genes are being discovered on a regular basis. This genetic revolution holds promise in numerous areas. Genetic screening tests for the general population (and for family members) are in the process of becoming a clinical reality. Furthermore, knowledge of the specific genetic mutation borne by an individual thoracic aortic aneurysm patient permits personalized care at the genetic level. That is to say, not all aneurysms behave similarly. Certain mutations (eg, ACTA2 and MYLK) produce aortic dissection at small aortic sizes. Affected patients violate the standard size algorithms for surgical intervention. The personalized mutation is now being invoked into the surgical decision-making process. The future will see this genetic incorporation increase steadily, with the specific mutation becoming an integral factor in the decision-making algorithm.

As our genetic understanding is enhanced, new pathogenic pathways will be identified, with direct potential correlates in terms of promising avenues for novel drug development. Directed drug development will be feasible when enhanced genetic and proteomic profiling of affected individuals leads to better appreciation of exactly how aortic wall proteins are deficient.

The 1960s saw “disruptive” advances in aneurysm care via the development of bold, novel surgical techniques. Today, we find ourselves on the threshold of similar “disruptive” insights and advances at the genetic level.

More than 100 years ago, Sir William Osler stated: “There is no disease more conducive to clinical humility than aneurysm of the aorta.” This is still true today because aneurysms and dissections remain virulent processes that challenge the skill and experience of physicians and surgeons. However, advances in progress at the present time promise to tame this virulent disease.