Aortic Dissection Nature.pdf

PRIMER Aortic dissection Christoph A. Nienaber1,2, Rachel E. Clough3,4, Natzi Sakalihasan5, Toru Suzuki6, Richard Gibbs7, Firas Mussa8, Michael P. Jenkins7, Matt M. Thompson9, Arturo Evangelista10, James S. M. Yeh1,2, Nicholas Cheshire1, Ulrich Rosendahl1 and John Pepper1,2 Abstract | Aortic dissection is a life-threatening condition caused by a tear in the intimal layer of the aorta or bleeding within the aortic wall, resulting in the separation (dissection) of the layers of the aortic wall. Aortic dissection is most common in those 65–75 years of age, with an incidence of 35 cases per 100,000 people per year in this population. Other risk factors include hypertension, dyslipidaemia and genetic disorders that involve the connective tissue, such as Marfan syndrome. Swift diagnostic confirmation and adequate treatment are crucial in managing affected patients. Contemporary management is multidisciplinary and includes serial non-invasive imaging, biomarker testing and genetic risk profiling for aortopathy. The choice of approach for repairing or replacing the damaged region of the aorta depends on the severity and the location of the dissection and the risks of complication from surgery. Open surgical repair is most commonly used for dissections involving the ascending aorta and the aortic arch, whereas minimally invasive endovascular intervention is appropriate for descending aorta dissections that are complicated by rupture, malperfusion, ongoing pain, hypotension or imaging features of high risk. Recent advances in the understanding of the underlying pathophysiology of aortic dissection have led to more patients being considered at substantial risk of complications and, therefore, in need of endovascular intervention rather than only medical or surgical intervention.

Correspondence to C.A.N. Cardiology and Aortic Centre, Royal Brompton Hospital, Royal Brompton and Harefield NHS Foundation Trust, Sydney Street, London SW3 6NP, UK. [email protected] Article number: 16053 doi:10.1038/nrdp.2016.53 Published online 21 July 2016; corrected online 25 Aug 2016

The wall of the aorta comprises three layers: the intima, which faces the bloodstream, the media and the outer adventitia. Acute aortic dissection is characterized by rapid development of an intimal flap, which is caused by blood flowing into the media and forcing the intima and the adventitia apart. This intimal flap separates the true lumen (the normal pathway of blood flow in the aorta) from a false lumen (a new route of blood flow in the media) (FIG. 1). In the majority of patients, intimal tears are present as sites of communication between these two lumens, but bleeding within the wall of the aorta might also be the cause of dissection in some patients1,2. As blood continues to flow into the aortic wall, the ­intimal flap can extend in both an antegrade and a retrograde direction from the site of the initial tear or bleeding and can progress to involve side-branch arteries. The first 2 weeks after the onset of an aortic dissection is considered the acute stage, in which patients are highly vulnerable to life-threatening complications and death, particularly when the ascending aorta is involved. This is followed by a subacute stage (3 months since onset) and then the chronic stage. This classification system is becoming less helpful in guiding therapeutic decision making. Rather, treatment decisions are increasingly being made on the basis of anatomical

classification systems that take into account the location of the dissection and/or origin of the intimal tear and the extent of the dissection. Stanford type A (or DeBakey type I and type II) aortic dissections involve the ascending aorta and usually require swift open surgical repair, whereas Stanford type B (or DeBakey type IIIa and type IIIb) dissections involve the descending but not the ascending aorta, and are conventionally treated by endovascular repair and/or medical therapy (FIG. 2). Imaging early in the disease course is important to understand the evolution of the disease and to guide management. This Primer refers to the Stanford classification as it is the most commonly used system among surgeons and ­physicians, despite its limitations. Aortic dissection can be associated with a very poor outcome, and rapid diagnosis and decision making are crucial. The diversion of normal blood flow into the false lumen can cause signs or symptoms of ischaemia, complications such as malperfusion (reduced blood supply to tissues and organs), aortic valve insufficiency (leakage of blood across the aortic valve into the left ventricle) and tamponade (compression of the heart). These can lead to sudden aortic rupture, circulatory failure and death in the majority of patients without timely treatment. To ensure that physicians are equipped with the

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 1 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER Author addresses Cardiology and Aortic Centre, Royal Brompton Hospital, Royal Brompton and Harefield NHS Foundation Trust, Sydney Street, London SW3 6NP, UK. 2 Faculty of Medicine, Imperial College London, London, UK. 3 Aortic Centre, Hôpital Cardiologique, Centre Hospitalier Régional Universitaire de Lille (CHRU de Lille), Lille, France. 4 Division of Imaging Sciences and Biomedical Engineering, King’s College London, London, UK. 5 Centre Hospitalier Universitaire de Liege, Department de Chirurgie Cardiovasculaire, Liege, Belgium. 6 Department of Cardiovascular Sciences, Cardiovascular Research Centre at Glenfield General Hospital, University of Leicester, Leicester, UK. 7 Department of Vascular Surgery, St Mary’s Hospital, Imperial College Healthcare NHS Trust, London, UK. 8 Department of Vascular Surgery, Columbia University, New York, New York, USA. 9 Department of Vascular Surgery, St George’s University of London, London, UK. 10 Department of Cardiology, University Hospital Vall d’Hebron, Barcelona, Spain. 1

right tools to effectively manage patients with both acute and chronic aortic conditions, the European Society of Cardiology has recently updated its guideline on the diagnosis and treatment of aortic diseases3. This update was made following the publication of several important studies and registries that focused on the management and outcome of aortic dissection1,4,5. The pathology of the aorta is attracting increasing attention in the light of ageing western and Asian popu­ lations and with the development of new diagnostic modalities and emerging therapeutic options. The complexity of aortic dissection demands a team approach of cardiac surgeons, cardiologists and vascular interventionists. Aortic dissection is considered a complex vascular scenario, the understanding of which has been improved by access to inflammatory biomarkers that help to understand the asymptomatic signs of high mortality risk

besides the classic symptoms of malperfusion and imminent rupture. In this Primer, we describe insights into the incidence and pathophysiology of acute aortic syndrome — the modern clinical term that encompasses all stages of aortic dissection from intramural haematoma to ruptured dissecting aneurysm — and discuss classic and emerging strategies to improve outcomes and long-term prognosis of this complex and c­ hallenging condition.

Epidemiology Dissection is the most common catastrophic event to affect the aorta, with an incidence exceeding that of ruptured abdominal aortic aneurysm in western populations2,6,7. Studies of acute aortic syndromes recorded at hospitals suggest that the incidence of acute aortic dissec­ tion is 3–5 cases per 100,000 people per year, which is half that of symptomatic aortic aneurysm8,9. However, this figure may underestimate the incidence of aortic dissection as hospital-based reports do not account for pre-­admission deaths10–15. Indeed, a prospective analysis of 30,412 middle-aged men and women with 20 years of ­follow‑­up reported 15 cases per 100,000 patient-years at risk for aortic dissection, with a 67.5% male preponderance16. In those 65–75 years of age, the incidence may even be as high as 35 cases per 100,000 people per year 17 (FIG. 3). Women with aortic dissection are more likely to present at an older age than men and to have atypical symptoms, which often delays diagnosis, leading to higher mortality. This finding remains true even after adjustment for age and arterial hypertension (which are important risk factors for developing the condition), and despite the higher incidence of aortic dissection in men than in women (16 cases per 100,000 men per year versus 8 cases per 100,000 women per year in those >50 years of age)18.

a Aorta

b Smooth muscle cell

Vasa vasorum

Endothelial cell

False lumen

True lumen

True lumen Intramural haematoma

Intimal flap

Intima Internal elastic membrane

Media

Adventitia External elastic membrane

Figure 1 | Aortic dissection is caused by bleeding within the aortic wall. The aortic wall comprises three layers: the intima, the media and the adventitia. Bleeding within the medial layer forces the layers apart to form an intimal flap. a | Aortic dissection probably results from an intimal tear in

Nature Reviews | Disease Primers most cases. b | In some cases, aortic dissection may also be caused by rupture of the vasa vasorum (the capillaries that supply the aortic wall). This bleeding results in an intramural haematoma that may progress to form an aortic dissection.

2 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER The International Registry for Aortic Dissection has shown that patients with aortic dissections who are of African descent living in Europe or the United States present at a younger age and with a higher incidence of cocaine abuse, hypertension and diabetes than white patients, although in‑hospital and 3‑year mortality were similar between the two groups19. Although reliable

Ascending aorta

a

Vertebral artery Tubular ascending aorta 22–36 mm (15 ± 1 mm/m2) Sinotubular junction 22–36 mm (15 ± 1 mm/m2) Sinus of Valsalva 24–40 mm (18 ± 2 mm/m2)

Common carotid artery Descending thoracic aorta 20–30 mm

Aortic annulus 20–31 mm (13 ± 1 mm/m2)

Diaphragm

Coeliac trunk Superior mesenteric artery Abdominal aorta

Left renal artery R Inferior mesenteric artery Left internal iliac artery

Right common iliac artery Right external iliac artery

b

Dissection involving the ascending aorta

Innominate artery

c

Dissection involving both the ascending and the descending aorta

L Aortic arch 22–36 mm Subclavian artery Mammary artery

d Dissection limited to

the descending aorta

DeBakey IIIa

True lumen False lumen

DeBakey IIIb

Stanford A DeBakey II

Stanford A DeBakey I

Stanford B DeBakey III

Figure 2 | Classification systems for aortic dissection. a | Schematic of aortic Nature Reviews | Disease Primers anatomy, including the aortic arch (inset, with R denoting right-sided arteries and L denoting left-sided arteries). The aortic arch spans from the origin of the innominate artery to just beyond the origin of the left subclavian artery and the descending thoracic aorta spans from distal to the origin of the left subclavian artery to the diaphragm. Not shown are the posterior intercostal arteries (n = 11 pairs; the first two pairs arising from the supreme intercostal artery, a branch of the costocervical trunk of the subclavian artery, and the rest arising directly from the thoracic aorta) and the lumbar arteries (n = 4 pairs, which arise directly from the abdominal aorta). b–d | Anatomic classification of aortic dissection. Adapted from REF. 132, Nature Publishing Group.

age-specific population-based incidence figures of aortic dissection are available for western countries, such data are not presently available in Asia16,17. However, aortic dissection is thought to be on the rise in China and Japan, as a reflection of the high prevalence of arterial hypertension20.

Mechanisms/pathophysiology Structure and function of the aorta Understanding aortic dissection first requires an understanding of the anatomy and physiology of the aorta. The aorta is the largest blood vessel in the body and can be categorized into four anatomical regions: the ascending aorta (which arises from the heart and from which the coronary arteries branch off ), the aortic arch (which curves over the heart and rotates towards the posterior thoracic wall and gives rise to branches that supply the head, neck and arms), the descending thoracic aorta (which extends through the posterior thoracic cavity in proximity to the spine) and the abdominal aorta (which is the region of the aorta distal to the diaphragm from which most of the major abdominal arteries branch off and further split into the paired iliac arteries in the lower abdomen) (FIG. 2). Dissection involving the ascending aorta is generally more ­common than dissections in other regions and is seen in two-thirds of cases21; the reason for this could be exposure to the systolic jet from the left ventricle or missed distal dissection, which occurs frequently. The normal ascending aorta diameter is 22–36 mm depending on age, size and sex, whereas the descending thor­ acic aorta diameter is 20–30 mm (REF. 22). The diameter of the abdominal aorta is generally 20 mm; importantly, however, the aorta is an elastic and dynamic structure that expands and shrinks with each heart beat (also known as the Windkessel effect)23. Finally, the aortic root is the portion of the aorta that is attached to the heart. The aortic root includes the aortic valve and has a normal diameter of 30–35 mm (REF. 22). The aorta is a composite tube consisting of three layers: an intima comprising endothelial cells on a basement membrane that provides a smooth surface for blood to flow across; an elastin-rich and smooth muscle-rich media that enables the aorta to expand and contract; and a collagen-rich and fibroblast-rich adventitia that provides additional support and structure to the aorta (FIG. 1). Key features of aortic dissection. Whether due to an inherent instability of the aortic wall (such as in the case of an inherited connective tissue disease) or an acquired condition (such as atherosclerotic degeneration due to ageing), compromised aortic integrity is a fundamental component of the underlying pathology of aortic dissec­tion. Once the structural or functional properties of the aorta are compromised, existing dissections are aggravated by mechanical stress owing to blood flow. Although the two main mechanisms that contribute to this degeneration are extracellular matrix degradation and inflammation, the precise trigger of aortic dissection is unknown.

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 3 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER 50

Stanford type A

45

Total

Men

40

Cases per 100,000 individuals

Stanford type B

Women Total

35 30 25 20 15 10 5

>8 5

<3 5 35 –4 4 45 –5 4 55 –6 4 65 –7 4 75 –8 4

>8 5

35 –4 4 45 –5 4 55 –6 4 65 –7 4 75 –8 4

<3 5

>8 5

35 –4 4 45 –5 4 55 –6 4 65 –7 4 75 –8 4

<3 5

0

Age (years)

Figure 3 | Incidence of aortic dissection. Age-specific and sex-specific rates per 100,000 individuals for incidence of Nature Reviews | Disease Primers acute aortic dissection by subtype (2002–2012). Data from REF. 17.

Two plausible events have been suggested as the initiating event in aortic dissection: a tear in the intima that causes blood to flow into the aortic wall or rupture of the vasa vasorum (the small vessels that supply the aortic wall), leading to weakening of the inner aortic wall (FIG. 1). The recognition that intramural haematoma, or bleeding into the aortic wall, is related to aortic dissection has demonstrated that both conditions can coexist and may constitute a spectrum24.

Cardiovascular risk factors Various factors can increase the risk of aortic dissection, many of which have been shown to predispose the aorta to become fragile (BOX 1). Older age, dyslipidaemia, increased levels of apolipoprotein AI and arterial hypertension can promote atherosclerotic degeneration of the aorta, leading to aortic wall fragility, which includes intimal thickening, fibrosis, calcification, extracellular fatty acid deposition and extracellular matrix degradation that compromise the elastic properties of the wall16,25. In addition, hypertension can place increased pressure on the aortic wall, which might trigger the development of an intimal tear. Hypertension might also contribute to the production of pro-inflammatory cytokines and matrix metalloproteinases (MMPs) that lead to excessive extracellular matrix degradation (discussed in more detail below)25. Approximately 80% of patients who develop an aortic dissection have hypertension16,21. In addition, 31% of patients have atherosclerosis, 15% have undergone previous cardiac surgery and 4% have an iatrogenic cause (such as trauma from previous catheter-based procedures or surgery)26. Patients who are <40 years of age are less likely to have a history of hypertension (34% of these patients) or atherosclerosis (1% of these patients), and are more likely to have a monogenic syndrome or bicuspid aortic valve (59% of these patients) than older patients27. The role of associated diabetic conditions in aortic dissection is controversial28.

Genetic associations Genetic studies have established that mutations in speci­ fic genes can distinguish patients who are at risk of aortic dissection (TABLE 1). This information has the potential to improve the management of these patients and to optimize the timing of aortic surgery. Genetic disorders that are associated with aortic dissections include Marfan syndrome, Loeys–Dietz syndrome, Ehler–Danlos syndrome and Turner syndrome29,30, with Marfan syndrome and Loeys–Dietz syndrome the best studied among them. Approximately 5% of patients with aortic dissection have Marfan syndrome26, which is an autosomal domin­ ant disorder caused by a mutation in FBN1 (REF. 31) or FBN2 (‘Marfan-like’ syndrome)32. These genes encode fibrillin 1 and fibrillin 2, respectively, which are components of elastin-associated microfibrils that are located mainly in the medial layer of the aorta. Mutations in FBN1‑causing Marfan syndrome result in a predisposition to develop predominantly aortic aneurysms and dissections, as well as skeletal and ocular features31. Loeys–Dietz syndrome has some similarity to Marfan syndrome in that patients with the disorder are prone to developing aneurysms and dissections of other arteries as well as the aorta32. Loeys–Dietz syndrome results generally from autosomal dominant mutations in either TGFBR1 (which encodes transforming growth factor‑β (TGFβ) receptor 1) or TGFBR2 (REFS 31,33). Non-syndromic mutations that appear like a normal phenotype, such as in ACTA2 (which encodes aortic smooth muscle actin) and SMAD2 (which encodes an intracellular signal transducer involved in TGFβ signalling), are also being explored regarding their associ­ ations with aortic dissection34. In addition, studies using exon sequencing in families in which aortic dissection is over-represented have shown that mutations in LOX (which encodes lysyl oxidase, a protein involved in the modification of collagen and elastin precursors)35, FOXE3 (which encodes a transcription factor involved

4 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER in the cellular differentiation of, for example, smooth muscle cells in the aorta)36 and MFAP5 (which encodes the extracellular matrix component microfibrillar-­ associated protein 5)37 are also associated with the condition, with ascending aortic dissections most common in patients with LOX mutations35. Genetic testing using a 21‑gene panel for thoracic aortic aneurysms and ­dissections is currently under investigation30.

Aortic aneurysm The presence of a thoracic aortic aneurysm is associated with increased risk of dissection at the level of the diseased segment 38. Patients with aneurysms of >60 mm in diameter have a yearly rate of rupture or dissection of ≥6.9% and a death rate of 11.8%; the mechanisms for the association between aneurysm size and dissection remain unknown, but increasing aneurysm size has been shown to be more strongly associated with an increased risk of rupture than an increased risk of dissection; risk factors such as longstanding hypertension and older age are suggested to have an influence on this association between diameter and rupture39. Dilatation of the aortic root Dilatation of the aortic root is an important risk factor of aortic dissection. However, 60% of non-syndromic patients with a Stanford type A aortic dissection have aortic diameters of <50 mm, whereas patients with Marfan syndrome or bicuspid aortic valve tend to dissect at a larger diameter 40 in the absence of hypertension. The diameter of the enlarged aortic root is a crucial trigger for aortic dissection in the setting of monogenetic dis­ orders (increased wall tension in the presence of weakened connective tissue causes it to enlarge slowly before dissection occurs), whereas, in patients with hypertension, factors other than aortic root dimension are important, such as systolic jet direction, micro-trauma from arterial hypertension and aortic wall degeneration. These factors may lead to rupture of the vasa vasorum and the evolution of an intramural haematoma. The triggers for compromised structural integrity of the aortic wall may be dysfunctional mechanosensing and mechano­regulation of the extracellular matrix by cells in the medial layer 41. Animal models of aortic dissection Although the pathogenetic mechanisms of aortic dissec­ tion have remained elusive, recent investigations using animal models have improved our understanding. Animal models of genetic aortopathies have focused on Marfan syndrome using genetically modified mice (that is, fibrillin-deficient mice)33 and SMAD3‑deficient mice42, among other models, to try and understand the molecules and pathways that contribute to aortic aneurysm and aortic root dilatation or dissection of the ascending aorta. Patients with connective tissue diseases (such as Marfan syndrome) have aortic dissections that most commonly involve the ascending aorta3. The main mechanistic pathway that leads to aortic pathology in these mouse models is the loss of structural integrity of the aortic wall. In these models, the main feature is cystic

medial degeneration, which is characterized by the loss and fragmentation of elastic fibres, the loss of smooth muscle cells and the accumulation of proteoglycans in the medial layer of the aorta43. Other studies have focused on using animal m ­ odels to understand dissection of the descending aorta, a condition more similar to that observed in elderly people. In this setting, aortic wall degeneration is associated with ageing, atherosclerosis, hypertension and accompanying inflammation. For instance, one study used a combination of angiotensin II infusion (to induce ­arterial hypertension), calcium chloride application and granulocyte–macrophage colony-stimulating factor (GM‑CSF) infusion (to induce inflammation) in wildtype mice. This study showed that macrophages invade the aorta as a consequence of cytokine stimulation (namely, GM‑CSF and downstream IL‑6), and that this invasion causes aortic fragility and subsequent dissection44. Other studies have shown that angiotensin II infusion alone in wild-type mice also results in dissection of the descending aorta (although to a lesser extent), and Box 1 | Risk factors for aortic dissection* Lifestyle factors • Long-term arterial hypertension • Smoking • Dyslipidaemia • Cocaine, crack cocaine or amphetamine use Connective tissue disorders • Marfan syndrome • Loeys–Dietz syndrome • Ehlers–Danlos syndrome • Turner syndrome Hereditary vascular disease • Bicuspid aortic valve • Coarctation of the aorta Vascular inflammation • Autoimmune disorders -- Giant-cell arteritis -- Takayasu arteritis -- Beçhet disease -- Ormond disease • Infection -- Syphilis -- Tuberculosis Deceleration trauma • Car accident • Fall from height Iatrogenic factors • Catheter or instrument intervention • Valvular or aortic surgery -- Side-clamping, cross-clamping or aortotomy -- Graft anastomosis -- Patch aortoplasty *See REF. 131. Adapted from Nienaber, C. A. & Powell, J. T. Management of acute aortic syndromes. Eur. Heart J., 2012, 33, 1, 26–35b, by permission of Oxford University Press.

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 5 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER that this dissection is associated with the accumulation of T helper 17 lymphocytes. Neutralization of IL‑17A with antibodies effectively reduced the frequency of dissection in this model, indicating that hypertension leads to immune modulation that may subsequently ­contribute to aortic dissection45. The treatment of mice with the lysyl oxidase inhib­ itor β‑aminopropionitrile monofumarate, a compound that disrupts the collagen and elastin crosslinking that is crucial for maintaining vessel wall integrity, causes dissec­tion of the descending aorta. Specifically, the absence of crosslinking leads to a mechanically fragile aorta and dissection within 24 hours of infusion of angiotensin II in approximately 70% of these mice, demonstrating that factors disrupting the extracellular matrix can increase the risk of aortic dissection. In this model, the diseased aorta showed leukocyte accumulation with increased intracellular levels of

granulocyte colony-stimulating factor (G‑CSF) along with upregulated expression of chemokines, such as CXC-chemokine ligand 1 (CXCL1)46. The use of an anti-CXC-chemokine receptor 2 (CXCR2) antibody reduced neutrophil accumulation and aortic rupture after dissection46, underlining the causative role of inflammation in aortic dissection in this context. Collectively, these studies show that inflammation through a cascade of cytokines (such as GM‑CSF, IL‑6, G‑CSF and CXCL1) and inflammatory cells (such as macrophages, lymphocytes and neutrophils) contributes to the onset of aortic dissection as well as subsequent aortic rupture (FIG. 4). At the cellular and molecular levels, activated macrophages can induce vascular smooth muscle cell apoptosis by activating the tumour necrosis factor (TNF) and nitric oxide signalling pathways47. Macrophages can also secrete pro-­inflammatory cytokines (such as IL‑6, TNF and

Table 1 | Genes associated with aortic dissection Gene

Key functions

Consequences of mutations

Refs

FBN1

• Encodes fibrillin 1 • Involved in the formation of microfibrils and elastogenesis • Promotes TGFβ2 bioavailability • Confers a smooth muscle cell phenotype on cells that express it

• Increases the risk of dissection in the ascending and thoracic aorta • Marfan syndrome (OMIM #154700)

Pyeritz (2016)31

EFEMP2

• Encodes fibulin 4 • Involved in the formation of elastic fibres

• Increases the risk of dissection in the ascending aorta • Cutis laxia autosomal recessive IIA (OMIM #219200)

Baldwin et al. (2013)134

TGFB1

• Encodes a cytokine with multiple functions

• Increases the risk of aortic dissection

Loeys et al. (2006)34

TGFBR1 and TGFBR2

• Encode receptors involved in TGFβ signalling

• Increases the risk of dissection in the thoracic aorta • Loeys–Dietz syndrome (OMIM #609192)

Loeys et al. (2006)34

MYH11

• Encodes smooth muscle myosin heavy chain • Involved in smooth muscle cell contraction

• Increases the risk of dissection in the thoracic aorta • Familial thoracic aortic aneurysm with patent ductus arteriosus (OMIM #132900)

Saratzis and Bown (2014)124

ACTA2

• Encodes aortic smooth muscle actin • Involved in smooth muscle cell contraction

• Increases the risk of dissection in the thoracic aorta • Familial thoracic aortic aneurysm (OMIM #611788)

Saratzis and Bown (2014)124

COL3A1

• Encodes type III collagen • A component of connective tissue

• Increases the risk of dissection in the thoracic aorta • Alters the composition of the extracellular matrix • Ehlers–Danlos vascular type IV (OMIM #130050)

Pyeritz (2014)129

SLC2A10

• Encodes GLUT10 (also known as SLC2A10) • Involved in glucose homeostasis

• Increases the risk of dissection in the aorta and other arteries • Decreases the levels of GLUT10 in the TGFβ pathway • Arterial tortuosity syndrome (OMIM #208050)

Callewaert, De Paepe and Coucke (1993)125

SMAD3

• Modulates transcription factors involved with the extracellular matrix

• Increases the risk of aortic dissection • Impairs TGFβ signal transduction • Syndromic form of aortic aneurysm and dissection (OMIM #613795)

Ye et al. (2013)44

LOX

• Encodes the majority of lysyl oxidase in the aorta

• Associated with aortic aneurysm and dissections (limited data)

Guo et al. (2016)37

FOXE3

• Encodes a transcription factor involved in cellular differentiation (for example, of smooth muscle cells in the aorta or epithelial cells in the lens)

• Associated with increased risk of aortic dissection

Kuang et al. (2016)38

MFAP5

• Encodes structural proteins of elastin fibres and microfibrils

• Associated with increased risk of aortic dissection

Schubert et al. (2016)126

MAT2A

• Encodes the enzyme MAT IIa, which catalyses the transfer of the adenosyl moiety from ATP to l‑methionine to synthesize S‑adenosylmethionine

• Associated with thoracic aortic aneurysm (limited data)

Guo et al. (2015)127

PRKG1

• Encodes type I cyclic GMP-dependent protein kinase, which controls smooth muscle cell relaxation

• Associated with thoracic aortic aneurysm and acute aortic dissection (limited data)

Guo et al. (2013)128

GLUT10, glucose transporter type 10; OMIM, Online Mendelian Inheritance in Man; TGFβ, transforming growth factor-β.

6 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER Angiotensin II

Hypertension

Dyslipidaemia

Direct pressure on the wall Intima

Intimal tear

Internal elastic membrane

Atherosclerosis

Recruitment of inflammatory cells

Media

IL-6, TNF and MCP1

Activation of TNF and NO signalling

Extracellular matrix degradation

Smooth muscle cell apoptosis

Aortic wall fragility

External elastic membrane Adventitia Endothelial cell

Fibroblast

Lymphocyte

Macrophage

Neutrophil

Smooth muscle cell

Nature Reviews | Disease Primers Figure 4 | Pathophysiology of aortic dissection involving inflammation. A schematic of the main players involved in inflammation of the aortic wall, which leads to aortic wall fragility (a precursor to aortic dissection and rupture). MCP1, monocyte chemoattractant protein 1; NO, nitric oxide; TNF, tumour necrosis factor.

monocyte chemoattractant protein  1 (MCP1; also known as CCL2)) and MMPs, which lead to extracellular matrix degradation. Vascular smooth muscle cell apoptosis and extracellular matrix degradation induce a­ ortic wall fragility. Pro-inflammatory cytokines secreted by macrophages and angiotensin II that stim­u lates smooth muscle cells and fibroblasts to produce pro-­ inflammatory cytokines (such as IL‑6 and MCP1) can both, in turn, promote the recruitment of further inflammatory cells to the aortic wall48. The potential mech­ anism underlying angiotensin II‑induced dissec­tion involves the promotion of atherosclerosis and increased pressure on the aortic wall. This direct pressure might provide a direct trigger for the initial tear and promote the production of inflammatory cytokines and MMPs by macrophages that result in the destruction of the ­extracellular matrix 48.

Diagnosis, screening and prevention Diagnostic approaches Aortic dissection poses a special diagnostic challenge for physicians because of its relative rarity and because symptoms of aortic dissection are likely to mimic other more common conditions (BOX 2); for these reasons, the correct diagnosis is missed on initial presentation and delayed in >30% of cases49. Early diagnosis is important as morbidity and mortality are related to late implementation of treatment. Imaging, biomarkers and genetic predisposition are key

to understanding the risk of developing an aortic dissec­ tion, to confirm a suspected diagnosis and to determine the appropriate intervention for any given patient. Specific features that relate to management decisions are important, such as the presence of rupture, the extent of the dissection, the involvement of branch vessels and end-organ ischaemia. Imaging techniques. Determining the appropriate imaging modality for aortic dissection depends on the haemodynamic stability of the patient, availability and local expertise. The advantages and limitations of common imaging approaches for evaluating aortic dissec­ tion are summarized in TABLE 2. Routine investig­ations such as electrocardiogram (ECG) and chest X‑ray are often non-diagnostic. Invasive angiography has been largely replaced by non-invasive imaging techniques, such as CT (FIG. 5) and MRI (FIG. 6). MRI takes longer to acquire and its restriction on metal objects (such as needles) in the vicinity makes it unsuitable for scanning unwell and/or unstable patients. A low threshold for CT imaging is recommended in current guidelines3. In the subacute and chronic stages of aortic dissection, MRI is currently advised instead of CT to avoid ionizing irradiation, which is particularly important in young patients. MRI can provide quantitative information about anatomy, haemodynamics and bio­ mechanics, and so may be able to provide information for risk stratification of patients with aortic dissection50.

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 7 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER Box 2 | Common signs and symptoms of aortic dissection Pain in thoracic aortic dissection is characterized by anterior chest or interscapular location (heart breaking or crucifying), migrant with maximum severity immediately mimicking a myocardial infarction and sometimes associated with vasovagal events, such as sweating, vomiting and fainting. The major signs of aortic dissection are: • Diastolic murmur of aortic insufficiency (40–50% of Stanford type A) • Pulsus paradoxus (tamponade) • Cardiogenic shock (tamponade, aortic insufficiency, major coronary occlusion by compression or dissection flap) • Hypovolaemic shock (aortic rupture) • Abdominal bruit • Loss of peripheral or femoral pulses • Hemiplegia, hemiparesis or paraplegia

Triple rule-out CT angiography is an emerging technology for the evalu­ation of coronary and pulmonary arteries, as well as the aorta and adjacent structures in a single ECG-gated multidetector or dual-source CT scan51. Motion artefacts and false-positive findings present frequent diagnostic challenges for the assessment of aortic dissec­tion, especially in the setting of a Stanford type A dissection, but can be avoided by ECG-gated acquisition. In an emergency setting, bedside trans­ thoracic echocardiography (FIG. 7) or trans­oesophageal echo­cardiography are useful for identifying dissection involving the ascending aorta52. Transthoracic echocardiography is dependent on the skill of the operator and has limitations imposed by the narrow acoustic

window and overlying lungs, which obstruct the view of the descending thoracic aorta. However, it is indispen­ sable for rapid bedside identification of very-highrisk patients who need immediate life-saving surgery (for example, those with features such as pericardial ­effusion, tamponade and/or aortic insufficiency). Accurate measurements of aortic dimension can be difficult to obtain because the aorta is a complex geometrical structure25,53. Standardized measurements made perpendicular to the axis of blood flow are important to assess changes in aortic size over time and to avoid erroneous findings of arterial growth. Interobserver and intra-observer variability in measurements of aneurysm diameter on CT imaging have been shown to be 5 mm and 3 mm, respectively 54–57. Thus, any change of 5 mm on a serial CT can be considered a significant change, but smaller changes are difficult to interpret, as are diastolic and systolic changes. Compared with CT, ultrasound imaging techniques systematically underestimate aortic dimensions by an average of 1–3 mm; thus, the same imaging technique should be used for serial measurements in any given patient. Biomarkers. Similar to troponins in the setting of acute coronary syndrome, biomarkers can be useful for both diagnosing and determining the risk of developing an aortic dissection. Biomarkers that are currently being evaluated as new techniques for the detection of aortic dissection1 are compared in TABLE 3. These include

Table 2 | Comparison of imaging techniques for the diagnosis of aortic dissection Technique

Advantages

Limitations

CT

• Widely available • Quick acquisition times • Enables evaluation of the entire aorta, its branches and the surrounding organs • Enables evaluation of the femoral and iliac artery access route for catheter-based interventions

• Exposes the patient to ionizing radiation • Requires the use of iodinated contrast media • Does not provide functional or dynamic assessment of the aorta

MRI

• Produces high-resolution images of the aorta and the aortic wall • Does not require ionizing radiation or iodinated contrast media, making it ideal for surveillance • Can provide functional and biomechanical information

• Relatively long acquisition times • Not recommended for use in patients who are haemodynamically unstable • Limited availability, especially in emergency settings • Provides limited assessment of access routes for catheter-based interventions because calcifications are not visualized

TTE

• Portable and widely available • Enables quick bedside assessment of cardiac function, aortic valve function, the ascending aorta and pericardium, making it highly valuable in the differential diagnosis and assessment of acute aortic syndromes

• Provides poor or insufficient assessment of the aorta beyond the distal arch • Needs to be combined with another imaging modality for a thorough assessment of aortic dissection • Does not have a role in the surveillance or assessment of chronic Stanford type B aortic dissection

TOE

• Provides high diagnostic accuracy in the thoracic aorta • Enables dynamic and functional assessment of the heart and the aorta • Extremely valuable during endovascular procedures, including for patient monitoring, assessment of the true and false lumen, and positioning and success of the stent graft (or grafts)

• Semi-invasive procedure that is dependent on the operator and requires sedation • Risk of decompensation during the procedure (best performed in a surgical theatre) • Cannot reliably assess most of the aortic arch nor the abdominal aorta (below the diaphragm) • Needs to be combined with another imaging modality for thorough assessment of aortic dissection

TOE, transoesophageal echocardiography; TTE, transthoracic echocardiography. Adapted from REF. 133.

8 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER a

b

Nature Reviews Figure 5 | CT imaging of aortic dissection. Standard CT (part a) and | Disease Primers electrocardiogram (ECG)-gated CT (part b). The aortic dissection is clearly seen only with ECG-gating (part b; arrow). The non-gated image (part a) may not enable diagnostic assessment, with the arrows indicating either artefacts or a new aortic dissection. Adapted from REF. 133.

smooth muscle markers, such as smooth muscle myosin heavy chain and calponin, which reflect the destruction and release of medial smooth muscle cellular components into the circulation at the time of dissection. Extracellular matrix proteins, such as MMPs and soluble elastin fragments, are also released into the circulation from aortic insult. The thrombosis (fibrinolysis) marker D‑dimer probably reflects the dynamic coagulo­pathic state resulting from blood flow into the aortic wall. D‑dimer has a pooled sensitivity of 94% for aortic wall dissection with a varying specificity of 40–100%58. Inflammatory biomarkers, such as C‑reactive protein, that are increased post-dissection reflect the extent of damage to the aorta and ongoing inflammation. Within this category, IL‑6 is an interesting biomarker that origin­ates from the liver after stimulation with cytokines and relates to the severity of dissection and

Figure 6 | MRI detection of aortic dissection. Time-resolved MRI (left|to right), Primers Nature Reviews Disease demonstrating contrast flow in the true (red arrow) and false (blue arrow) lumens. From REF. 130. Adapted with permission from Clough, R. E., Zymvragoudakis, V. E., Biasi, L. & Taylor, P. R. Usefulness of new imaging methods for assessment of type B aortic dissection, Ann. Cardiothorac Surg. 3, 3, 314–318.

the time after presentation 59. An activated renin–­ angiotensin system might reflect uncontrolled hypertension and thereby might be a risk marker for vascular complications, such as dissection, in the future60. Other biomarkers being developed for diagnosis include endothelin 1 and pro-brain natriuretic peptide1.

Screening In patients with genetic disorders, regular clinical and imaging follow‑up is recommended and screening by imaging and/or genetic testing should be considered for family members3. Patients with Marfan syndrome require ultrasound interrogation at the time of diagnosis and during follow‑up. If the aortic root, which is usually amenable to ultrasonography, is stable, then patients should undergo annual surveillance. Patients with Loeys–Dietz syndrome should have complete aortic imaging at the time of diagnosis and at 1‑year intervals thereafter, preferentially using MRI screening from the cerebrovascular circulation down to the femoral arteries3. Prevention As discussed above, arterial hypertension is an important risk factor for aortic dissection. The incidence of dissection is approximately 21.4 cases per 100,000 individ­uals per year in those with hypertension compared with 5.4 cases per 100,000 individuals per year in those with normal blood pressure. This increased risk indicates that approximately 50% of aortic dissections could potentially be prevented by timely and effective blood pressure control16. Elimination of dyslipid­aemia and cessation of smoking are presumably equally important in the prevention of aortic dissection. Patients with segmental aortic dilation may require preventive interventions; this concept involves external aortic support in the phase of early enlargement (40–45 mm) and prophylactic surgical root replacement when the aorta has reached ≥45 mm in diameter 3,61. Although aortic root dilation is particularly important in those with connective tissue disorders, it is less important in individuals with hypertension but no connective tissue disorder (that is, non-syndromic patients). As such, pre-emptive surgery for aortas of ≥50 mm in diameter would fail to prevent dissection in 40% of non-syndromic patients. Similarly, in 21% of patients with Stanford type B aortic dissection, the aortic diameter is <35 mm; new biomarkers are needed to identify patients at risk within this population62. Patients with any aortic dilatation should also receive blood-­pressurelowering treatment; β‑blocking agents and sartans are the preferred medications. These patients should also avoid high-impact exercise3. Management Aortic dissection can be managed by surgical or endovascular intervention, or medically. Non-elective surgical intervention for Stanford type A aortic dissection is associated with a high incidence of death, in the region of 30%, particularly in cases with late presentation63. These data reinforce the need for swift diagnosis to

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 9 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER a

c True lumen

b

d

e

False lumen

Nature | Disease Primers Figure 7 | Transoesophageal echocardiography detection ofReviews aortic dissection. Transoesophageal echocardiography imaging demonstrating the in‑folding and unwinding (arrows) of the intimal flap at the level of the ascending aorta (part a and part b), and fluctuation of the intimal flap at the level of the descending aorta during a cardiac cycle (parts c–e). Adapted from REF. 133.

prevent associated complications, such as aortic rupture or pericardial effusion, which may lead to cardiac tamponade (in which blood accumulates in the pericardial sac, compressing the heart) and death. Patients with aortic dissection, or suspected dissection, should be managed in the acute setting with close monitoring of ECG, blood pressure and urine output. Pain and systolic blood pressure should be carefully monitored in particular, with the aim of lowering the systolic blood pressure to <120 mmHg using intravenous β‑blockers (such as labetalol or esmolol) and vasodilators. In the setting of malperfusion, the outcome is likely to be compromised in cases of older age, partial or complete aortic rupture, comorbid disease (particularly coronary artery disease and emphysema), shock and hypotension (which is defined as a systolic blood pressure of <100 mmHg)1,5. Acute aortic dissection and other types of acute aortic syndrome are best managed in aortic centres by experienced teams. Most patients can be transported safely provided they are not in cardiogenic shock and both pain and blood pressure are well controlled. A useful therapeutic algorithm for aortic dissection is shown in FIG. 8.

Stanford type A aortic dissection General principles. Surgery for Stanford type A aortic dissection involves resection of the diseased part of the aorta containing the intimal tear. This segment is then replaced with a synthetic (Dacron) vascular prosthesis. In cases of severe aortic insufficiency, aortic valve replacement or aortic valve-sparing resuspension will be done at the same time as surgery for the dissection. If required, re‑implantation of the coronary arteries can also be performed at this time. In cases of a critically dilated aortic root (>50 mm in diameter), replacement including coronary re-implantation are often inevitable.

Modern surgical approaches. Modern surgery for the ascending aorta emerged with the introduction of extracorporeal circulation (cardiopulmonary bypass or heart–lung machine). The first total replacement of the ascending aorta was performed and established in 1956 (REFS 64–67). Successful replacement of the ascending aorta including the aortic valve was performed by Wheat68 and Bentall69; Yacoub replaced the ascending aorta with preservation of the aortic valve, leading the way to more-recent aortic valve-sparing techniques70,71. Valve-sparing techniques should be encouraged, as they avoid the need for lifelong anticoagulation medication and can be combined with hemi-arch or complete-arch replacement strategies. Surgery to replace the damaged portion of the aorta requires cardiopulmonary bypass, through cannulation of the vena cava and femoral or right axillary artery. Although surgery is feasible under normothermia, it is usually performed with the patient cooled to 24–30 °C to protect the brain; antegrade brain perfusion is preferred as an adjunct. In essence, after clamping the aorta below the innominate artery, if the dissection left both the sinuses and the coronary ostia intact, a ­single vascular channel is re‑established by a proximal suture to a graft at the level of the sinotubular junction. Conversely, if the coronary arteries or sinuses are compromised by the dissection, a valve conduit is required, which is sutured to the aortic annulus below the coronary ­orifices (the openings of the coronary arteries in the aortic root). This procedure necessarily involves relocat­ing the coronary orifices into the prosthesis itself. Distally, the prosthesis, which is essentially a valve-bearing conduit, is sutured end‑to‑end to the origin of the transverse aorta (a technique called the Bentall procedure). In the conventional Bentall procedure, the coronary orifices are sutured directly to the aortic conduit 69. In the original variant of this technique performed by Cabrol72, a prosthetic graft is attached to the coronary orifices and then secondarily on to the tubular aortic graft. The Cabrol procedure is rarely adopted today due to complications. Complementary measures. In cases of complicated aortic arch dissection, complementary measures might need to be taken. Control of the descending aorta is a great challenge for the surgeon and, in the most difficult cases, it is recommended to perform the Bentall procedure under complete circulatory arrest and deep hypothermia at 18 °C69. The ‘elephant trunk’ technique is a further addition to the management of complex dissec­tions and aneurysm involving both the aortic arch and the descending aorta71. The classic elephant trunk is rarely used in acute aortic dissection, but may be encouraged in the acute setting in an open e­ ndovascular form of a frozen (snug-fit) elephant trunk73.

Stanford type B dissection Endovascular repair. Endovascular repair of the thor­ acic aorta (thoracic endovascular aneurysm repair (TEVAR)) has changed the management algorithm for patients with acute Stanford type B aortic dissection.

10 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER The core principle underpinning TEVAR is to place a covered stent graft over the entry tear in the descending thoracic aorta. The addition of this stent drives changes in the haemodynamics of the dissected aorta, resulting in depressurization of the false lumen and rapid expansion of the true lumen74 (FIG. 9). Patients with malperfusion issues, tears that are large but with a single entry point into the aortic wall, repetitive symptoms, or refractory or difficult-to-control hypertension are particularly likely to develop complications of acute dissection and, therefore, benefit from endovascular treatment 5. Expansion of the true lumen will amelior­ ate dynamic malperfusion of the viscera and lower limbs, whereas false lumen collapse and thrombosis will prevent bleeding from false lumen rupture; the endovascular approach may induce re‑approximation and remodelling of the dissected segment and remove the need for open surgery.

TEVAR is usually performed via femoral artery access and the technique has evolved since its inception in 1999 (REFS  75,76). To prevent an iatrogenic (treatment-related) retrograde Stanford type A dissec­ tion — which may occur in up to 2% of elective cases with a mortality rate of >30%, or occur in up to 23% of patients with extensive debranching procedure performed in the acute setting of an aortic arch dissec­tion77 — endo­grafts should be sized to the diameter of the non-­dissected proximal landing zone in the healthy region of the aorta78 (usually just proximal to the left subclavian artery) to avoid oversizing; b ­ alloon dilata­ tion of the stent graft should be avoided. To ensure optimal outcomes, optional additional features of a TEVAR procedure can include transoesophageal echocardio­graphy or intravascular ultrasonography to monitor the procedure and to ensure the endografts are placed in the true lumen in the exact location desired.

Table 3 | Biomarkers and tests for acute aortic conditions Serum biomarker or test

Advantages

Limitations

Availability

Increased levels of C‑reactive protein

• Provides prognostic information of dissection

• Not specific for aortic dissection • Not useful for identifying patients who are at risk of aortic dissection • Not useful for rapid diagnosis of acute aortic conditions

Widely available

Increased levels of IL‑6

• Common marker of inflammation

• Not specific for aortic dissection

Widely available

Increased levels of smooth • Can be used to facilitate rapid diagnosis muscle myosin heavy chain of acute aortic conditions

• Does not provide information about the risk or prognosis of aortic dissection

Research only

• Can be used to facilitate rapid diagnosis of acute aortic conditions

• Does not provide information about the risk or prognosis of aortic dissection

Research only

• May discern between aortic dissection and myocardial infarction

• Not specific for aortic dissection • Does not provide information about the risk or prognosis of aortic dissection

Widely available

Increased levels of creatine • May discern between aortic dissection kinases and myocardial infarction

• Not specific for aortic dissection • Does not provide information about the risk or prognosis of aortic dissection

Widely available

• Not specific for aortic dissection

Widely available

Inflammation

Smooth muscle damage

Increased levels of calponin Cardiac stress or damage Increased levels of cardiac troponins

Increased levels of  pro-brain natriuretic peptide

• May indicate heart failure

Extracellular matrix damage Increased levels of matrix metalloproteinases

• May be useful for identifying patients who are at risk of aortic dissection • May be useful for rapid diagnosis of acute aortic conditions

• Very limited early data

Research only

Increased levels of soluble elastin fragments

• May be used to facilitate rapid diagnosis of acute aortic conditions

• Very limited early data

Research only

• Can be used to facilitate rapid diagnosis of acute aortic conditions • Provides prognostic information on subacute aortic dissection

• Not specific for aortic dissection • Does not provide prognostic information in chronic aortic dissection

Widely available

• Provides good information about the risk of aortic dissection and prognosis

• Does not enable rapid diagnosis of acute aortic conditions

Limited availability

Thrombosis or fibrinolysis Increased levels of D‑dimer

Inherited predisposition Genetic testing

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 11 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER The length of the aorta that should be covered by endograft remains controversial and is dictated by clinical context. In the setting of a contained rupture in a­ ortic dissection, the descending thoracic aorta should be scaffolded from the left subclavian artery to just above the coeliac trunk to prevent retrograde perfusion of the false lumen. In patients with dynamic (intermittent) malperfusion, coverage of the proximal entry tear using an endo­graft will usually expand the true lumen sufficiently to reper­fuse ischaemic viscera and legs. In cases in which it is not clear whether reperfusion has taken place, the true lumen may be further expanded by implantation of a bare stent distally 79,80. Adjunctive bare metal stents may be required to treat mal­perfusion ­arising from static malperfusion and to induce false lumen thrombosis. Pre-emptive revascularization of the left sub­clavian artery is needed when the left subclavian artery origin is to be covered (occluded) to prevent clinical sequelae, such as arm ischaemia, myocardial ischaemia (from the left internal mammary arterial coronary bypass graft), posterior circulation stroke or spinal cord ischaemia (SCI)81. The favourable clinical results of TEVAR in the setting of complicated Stanford type B aortic dissection have changed the management of this condition. Most large registries report 30‑day mortality rates of approximately 10%, with the majority of patients dying from consequences of the dissection21,82. In addition, a meta-analysis showed that, following TEVAR, the

rate of in‑hospital mortality, stroke and paraplegia was 9%, 3.1% and 1.9%, respectively 83. In the longer term, TEVAR for acute dissection seems to provide reasonable protection from mid-term aortic-related death. In the MOTHER registry 84, the aortic-related mortality per 100 patient-years was 1.2, although there was a requirement for re‑intervention in 6.7 per 100 patient-years. The prevention of aortic-related death may be linked to the morphological changes in the realigned and remodelled aorta and with the occurrence of true lumen expansion, false lumen collapse and false lumen thrombosis in the stented part of the aorta in 84% of patients85. The recent trend towards favouring TEVAR for early intervention in acute and subacute Stanford type B aortic dissections has partly been driven by wide uptake, relative ease of the procedure and poor results from open surgery in this group of patients86,87. Surgical replacement. Open surgery for chronic Stanford type B aortic dissection is elective, rarely performed and involves replacement of the dissected aortic segment with excision of the septum and re‑­implantation of the visceral, renal and, where appropriate, intercostal arteries88. Typically, left heart bypass is used, but reasonable results have also been reported using deep hypothermia to 18 °C at circulatory arrest 89. Although not systematically studied, the durability of this approach seems to be superior to TEVAR in a similar patient group and these differences are increased in patients with connective

Patient with chest pain History and physical examination Routine blood test, troponin, D-dimer, ECG and chest X-ray Diagnosis uncertain or suspect aortic dissection

Negative

Urgent CT scan — triple rule out Pulmonary embolism

Stanford type A

Open surgery unless futile

Aortic dissection

Acute coronary syndrome

Stanford type B

Uncomplicated Medical treatment with or without pre-emptive endovascular (stent) treatment in the subacute phase

Complicated Endovascular (stent) treatment

Consider other diagnosis and/or tests (such as echocardiography or endoscopy) or early repeat CT to exclude progression to detectable acute aortic syndrome

Figure 8 | An algorithm for the identification and treatment of aortic dissection inNature patients presenting with Reviews | Disease Primers chest pain. Acute aortic syndrome often presents with chest pain. The differential diagnosis of chest pain includes acute coronary syndrome (unstable angina and myocardial infarction), pulmonary embolism (blood clot in the pulmonary artery of the lung) and others (for example, musculoskeletal pain, oesophagitis, gastro-oesophageal reflux, chest infection and pericarditis). Open surgery for Stanford type B dissection is not recommended and is rarely performed. The term ‘complicated’ type B dissection means dissection with aortic rupture or impending rupture, end-organ or limb ischaemia, ongoing pain or hypertension despite medical therapy, large entry tear (>10 mm), early false lumen expansion (>22 mm false lumen diameter or >40 mm total lumen diameter), partial false lumen thrombosis or ongoing aortic inflammation on PET imaging.

12 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER ischaemic neurons ceasing to function. These defects are reversible with improved blood flow. As perfusion pressure is directly related to arterial pressure and inversely related to cerebrospinal fluid pressure99, cerebro­spinal drainage, permissive hypertension and re-­ implantation of intercostal arteries can reduce or abolish SCI. Simultaneously, neuronal metabolic demand can be reduced with local or systemic hypo­thermia97, as described above. Development of SCI remains a risk after intervention, and continuation of spinal cord protection is important for 48 hours while the collateral network is recruited98. New strategies for SCI prevention, such as staged repair and preconditioning, rely on this time-dependant collateral recruitment 100–102.

Nature Reviews Disease is Primers Figure 9 | Endovascular repair of Stanford type B aortic dissection. A |catheter used to insert an expandable stent graft into the aorta to cover the site of the intimal tear. In cases in which the stent graft occludes a branching artery, pre-emptive surgery is required (hybrid intervention) to vascularize the occluded artery. The covered stent graft excludes the false lumen, which collapses. Bleeding from the false lumen rupture is prevented by thrombosis within the false lumen and the covered stent graft. The endovascular approach may induce re‑approximation and remodelling of the dissected segment. In the figure on the left, the arrow indicates blood flow out of the true lumen and into the false lumen.

tissue disease90,91. Contemporary experience suggests that open surgery can be performed safely in selected patients who are deemed fit for surgery with low stroke and paraplegia rates92–94. However, the future role of open surgery remains unclear, as there is increasing evidence supporting the fact that early TEVAR facilitates aortic remodelling, but whether this is sufficient to prevent late dilatation of the abdominal aorta remains to be proven95. The hybrid repair of thoraco-abdominal aneurys­mal degeneration combining open visceral re‑routing and TEVAR of the remaining aorta may be a possible strategy in the chronic post-dissection state; however, the rate of advancement of endovascular techniques and devices has surpassed the moderate improvement in outcomes from classic surgery, which means that the technological evolution of TEVAR may overcome its current deficiencies.

Neuro-monitoring In endovascular and open procedures of the dissected aorta, the principle neurological complication is SCI. The reported risk for thoraco-abdominal repair is 5–20% for open, 0–30% for hybrid open–­endovascular and 9–16% for endovascular procedures96,97. The mechanism of neuronal injury is reduction of spinal cord perfusion pressure and hypoxia from impaired c­ ollateral supply to spinal arteries fed from the vertebral, internal mammary, intercostal, lumbar and hypogastric (internal iliac) arteries97,98. Replacement with a graft or coverage with an endovascular stent graft can compromise collateral perfusion. Monitoring of the spinal cord function can be achieved using somatosensory-evoked and motorevoked potentials that assess neuronal function in the ventral grey matter and sensory dorsal horn. Attenuation of signals implies critically reduced perfusion, with

Medical management and follow‑up Some patients with Stanford type B aortic dissection who are managed medically initially may require intervention (TEVAR) at a later subacute or chronic stage103. Medical management involves β-blockers, sartans and statins to control blood pressure and inflammation3. All patients with or without TEVAR or who have undergone surgical intervention should maintain a normal blood pressure and should be kept under surveillance with re‑imaging for asymptomatic changes that might require intervention (or re-intervention). The greatest risk for patients with aortic dissection seems to be in the first 24 months after presentation, although late changes may occur at a slower pace; follow-up protocols can be adjusted ­accordingly with longer follow‑up intervals4,103,104. Quality of life Quality of life is difficult to measure after surgery, but there seems to be agreement that replacement of the dissected ascending aorta reduces the chance of death and improves quality of life105,106. With an excellent surgical result, the life of the patient after surgery or after an endovascular procedure can be comparable to the pre-­ dissection state, but the lifestyle perceptions, feelings and expectations of the individual are subject to change106,107. For instance, heavy weight lifting and some contact sports need to be abandoned. In the absence of any peri-operative or peri-interventional neurological complications, such as paraparesis and paraplegia, a good quality of life may return within a matter of weeks108 (FIG. 10). Conversely, neurological complications increase the length of hospital stay and may incur a lengthy recovery in a neurorehabilitation centre. In addition, the type of neuroprotection used during the intervention might affect mid-term (3 months) quality of life; antegrade cerebral perfusion is considered best in this regard in the field. Even after successful surgery or endovascular repair, many patients remain with a residual dissected segment of aorta and require a strict clinical and imaging follow‑up and, sometimes, re-intervention108–111. Outlook Outstanding pathophysiological questions There are no reports on cellular and molecular mechanistic differences between dissections affecting the ascending and descending aorta, with inflammation

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 13 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER 70

Before aortic dissection

60

After aortic dissection

Patients (%)

50 40 30 20 15 10

Go lf th er of fo ca rm rd s io Bo w lin M g an ua l la bo Hu ur nt in g a Ya fi nd rd shi /or w o r ng k cl and ea / ni o r n Sw g im m in g No ex er ci se O

g

sp or ts Te nn ra is cq an Se ue d/o xu tba r al l re l la tio ns

m

Te a

g in

W al k in

g

tli ft gh

cy cl in

W ei

Bi

Ru n

ni

ng

0

Activity

Figure 10 | Engagement of physical activities before and after aortic dissection. Percentage of patients engaging in Nature Reviews | Disease Primers various physical activities before and after aortic dissection. Adapted with permission from REF. 107, © 2015 Wiley Periodicals, Inc.

and extracellular matrix degradation being the common processes that lead to all scenarios of dissection. Most experimental studies using genetic models of aorto­pathy (such as Marfan syndrome) have focused on the effects on the ascending aorta. Translation of the findings in mice to patients remains unclear as experimental models do not necessarily match the phenotype in patients. Improved understanding of the underlying mechanisms that cause aortic dissection may enable the identification of targets for the development of new therapies, and thus help to move beyond the current medical treatment strategy, which is limited mainly to antihypertensive agents111,112. Indeed, in the near future, the underlying processes of extracellular matrix degrad­ ation and cellular ­inflammation may become targets for specific therapy.

acquired in 4D might enable haemodynamic assessment of flow pattern and derived aortic wall stress parameters, providing information that reaches beyond anatomical assessment for the extent of dissection and aortic diameter 52,113. Another form of functional imaging that is potentially useful in aortic dissection is PET, which is able to depict the distribution of injected 18 F-fluorodeoxyglucose along the entire aorta and identify areas of enhanced metabolic activity, usually from accumulated inflammatory cells or invading mono­ cyctes in areas of acute injury. PET in conjunction with morphological imaging using CT or MRI is already being used to identify areas of tissue inflammation. These approaches may be instrumental in judging the risk of rupture and the need for preventive treatment in chronic dissection113,114.

Emerging imaging techniques The increased availability of diagnostic imaging will enable increasingly rapid and accurate detection of acute aortic conditions53–55 (BOX 3). Current precision imaging may enable us to understand the underlying pathophysiological processes beyond just the diameter in aortic dissection, including early intramural haema­ toma, more-subtle forms of aortic wall disease and patients who are at risk of developing a dissection. For instance, phase-contrast sequences and velocity maps

Advances in treatment Recent evolution of treatment strategies and intra-procedural CT imaging 1 point towards earlier endovascular treatment for subacute and chronic Stanford type B aortic dissection within the 3‑month window of plasticity (when the aorta still retains the capacity to remodel)4,114,115, as well as endovascular solutions for ascending (Stanford type A) aorta dissections116–120. Current challenges include maintaining an open true lumen in Stanford type B dissection, minimizing the risk of significant cerebral emboli, SCI and damage to the aortic wall. In the setting of malperfusion, recovery of an open true lumen using adjunctive procedures, such as the PETTICOAT stent technique, is established, and embolization techniques for the false lumen have been used with some success80,120,121. Simple tube grafts have evolved to fenestrated and branched devices and are used in scenarios involving the aortic arch and visceral branch vessels. The two largest series to use these approaches report technical success in all patients, with a mortality of 0–8.7%119,121. The improving stent-graft

Box 3 | Future developments in clinical approaches to aortic dissection • Identification of biomarkers and advances in modern imaging technology (anatomical and functional imaging or multi-modality imaging) for early diagnosis, risk stratification and guiding treatment strategies • Advances in stent-graft technology and techniques for broader endovascular treatment applications • Availability of genetic and biomarker profiling for individualized preventive and treatment strategies

14 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER technology and user expertise means that, in future, the majority of patients will probably have stent-graft treatment if anatomically suitable, regardless of the ­initial clinical situation122. Robotically steerable catheters can be used in these procedures for precise navigation through d ­ ifficult anatomy 123.

Risk stratification and early diagnosis Regardless of technical improvement on the treatment side, a major part of aortic dissection management is the early detection and identification of individ­ uals who are at risk of death or progression of disease. Recent advances in genomic technologies and their

Clough, R. E. & Nienaber, C. A. Management of acute aortic syndrome. Nat. Rev. Cardiol. 12, 103–114 (2015). 2. Vilacosta, I. & Roman, J. A. Acute aortic syndrome. Heart 85, 365–368 (2001). 3. Erbel, R. et al. 2014 ESC Guidelines on the diagnosis and treatment of aortic diseases: document covering acute and chronic aortic diseases of the thoracic and abdominal aorta of the adult. The Task Force for the Diagnosis and Treatment of Aortic Diseases of the European Society of Cardiology (ESC). Eur. Heart J. 35, 2873–2926 (2014). This paper describes a multidisciplinary guideline highlighting the changing trends in the management of acute aortic conditions. 4. Nienaber, C. A. et al. Endovascular repair of type B aortic dissection: long-term results of the randomized investigation of stent grafts in aortic dissection trial. Circ. Cardiovasc. Interv. 6, 407–416 (2013). This paper reports the first randomized long-term comparison between medical management and pre-emptive stent grafting in Stanford type B aortic dissection, showing a long-term benefit using the latter approach. 5. Nienaber, C. A. & Clough, R. E. Management of acute aortic dissection. Lancet 385, 800–811 (2015). This article provides the most recent update on changing trends in the management of aortic dissection. 6. Bickerstaff, L. K. et al. Thoracic aortic aneurysms: a population-based study. Surgery 92, 1103–1108 (1982). 7. Clouse, W. D. et al. Acute aortic dissection: population-based incidence compared with degenerative aortic aneurysm rupture. Mayo Clin. Proc. 79, 176–180 (2004). 8. Olsson, C., Thelin, S., Stahle, E., Ekbom, A. & Granath, F. Thoracic aortic aneurysm and dissection: increasing prevalence and improved outcomes reported in a nationwide population-based study of more than 14,000 cases from 1987 to 2002. Circulation 114, 2611–2618 (2006). 9. Acosta, S. et al. Increasing incidence of ruptured abdominal aortic aneurysm: a population-based study. J. Vasc. Surg. 44, 237–243 (2006). 10. Mehta, R. H. et al. Chronobiological patterns of acute aortic dissection. Circulation 106, 1110–1115 (2002). 11. Collins, J. S. et al. Differences in clinical presentation, management, and outcomes of acute type a aortic dissection in patients with and without previous cardiac surgery. Circulation 110, II237–II242 (2004). 12. Suzuki, T. et al. Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the International Registry of Aortic Dissection (IRAD). Circulation 108 (Suppl. 1), II312–II317 (2003). 13. Ramanath, V. S., Oh, J. K., Sundt, T. M. 3rd & Eagle, K. A. Acute aortic syndromes and thoracic aortic aneurysm. Mayo Clin. Proc. 84, 465–481 (2009). 14. Feldman, M., Shah, M. & Elefteriades, J. A. Medical management of acute type A aortic dissection. Ann. Thorac Cardiovasc. Surg. 15, 286–293 (2009). 15. Thrumurthy, S. G., Karthikesalingam, A., Patterson, B. O., Holt, P. J. & Thompson, M. M. The diagnosis and management of aortic dissection. BMJ 344, d8290 (2012). 1.

imminent availability provides an opportunity to integrate these techniques into routine clinical practice. The link between various phenotypes and genetic variants may refine our ability to risk stratify individuals based on their genetic profile124–129. Monitoring of key patho­ physiological modulators, such as the TGFβ pathway (among others)21, in combination with metabolic technologies might enable the identification of clinically useful biomarkers for rapid diagnosis of acute dissection at the bedside130–135. Use of biomarkers might also facilitate treatment strategies that are based primarily on individualized endovascular solutions, forming the first steps towards personalized medicine in aortic dissection1,126.

16. Landenhed, M. et al. Risk profiles for aortic dissection and ruptured or surgically treated aneurysms: a prospective cohort study. J. Am. Heart Assoc. 4, e001513 (2015). This paper provides a prospective analysis of 30,412 middle-aged individuals with 20 years of follow‑up, reporting 15 cases per 100,000 patient-years for aortic dissection with a male preponderance. 17. Howard, D. P. et al. Population-based study of incidence and outcome of acute aortic dissection and premorbid risk factor control: 10‑year results from the Oxford Vascular Study. Circulation 127, 2031–2037 (2013). This article describes a population-based study reporting an aortic dissection incidence of 35 cases per 100,000 patient-years in those 65–75 years of age. 18. Nienaber, C. A. et al. Gender-related differences in acute aortic dissection. Circulation 109, 3014–3021 (2004). 19. Bossone, E. et al. Acute aortic dissection in blacks: insights from the International Registry of Acute Aortic Dissection. Am. J. Med. 126, 909–915 (2013). 20. Yeh, T. Y. et al. Epidemiology and medication utilization pattern of aortic dissection in Taiwan: a population-based study. Medicine (Baltimore) 94, e1522 (2015). This article demonstrates that the incidence of aortic dissection is rising in China and Japan alongside a high prevalence of arterial hypertension. 21. Hagan, P. G. et al. The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA 283, 897–903 (2000). This paper reports contemporary figures on the outcomes of Stanford type A and Stanford type B aortic dissection as a function of surgical or medical management from the International Registry for Aortic Dissection. 22. Evangelista, A. et al. Echocardiography in aortic diseases: EAE recommendations for clinical practice. Eur. J. Echocardiogr. 11, 645–658 (2010). 23. Hirsch, A. T. et al. ACC/AHA guidelines for the management of patients with peripheral arterial disease (lower extremity, renal, mesenteric, and abdominal aortic): a collaborative report from the American Associations for Vascular Surgery/Society for Vascular Surgery, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, Society of Interventional Radiology, and the ACC/AHA Task Force on Practice Guidelines (writing committee to develop guidelines for the management of patients with peripheral arterial disease) — summary of recommendations. J. Vasc. Interv. Radiol. 17, 1383–1397 (2006). 24. von Kodolitsch, Y. et al. Intramural hematoma of the aorta: predictors of progression to dissection and rupture. Circulation 107, 1158–1163 (2003). 25. Yin, H. & Pickering, J. G. Cellular senescence and vascular disease: novel routes to better understanding and therapy. Can. J. Cardiol. 32, 612–623 (2016). 26. Pape, L. A. et al. Presentation, diagnosis, and outcomes of acute aortic dissection: 17‑year trends from the International Registry of Acute Aortic Dissection. J. Am. Coll. Cardiol. 66, 350–358 (2015).

27. Januzzi, J. L. et al. Characterizing the young patient with aortic dissection: results from the International Registry of Aortic Dissection (IRAD). J. Am. Coll. Cardiol. 43, 665–669 (2004). 28. Nienaber, C. A. Diabetes mellitus and thoracic aortic disease: are people with diabetes mellitus protected from acute aortic dissection? J. Am. Heart Assoc. 1, e001404 (2012). 29. Milewicz, D. M. & Regalado, E. S. Use of genetics for personalized management of heritable thoracic aortic disease: how do we get there? J. Thorac Cardiovasc. Surg. 149, S3–S5 (2015). 30. Ziganshin, B. A. et al. Routine genetic testing for thoracic aortic aneurysm and dissection in a clinical setting. Ann. Thorac Surg. 100, 1604–1611 (2015). This is an interesting report on the clinical experience of routine genetic testing in non-syndromic and syndromic patients with aortic aneurysm and dissection. 31. Pyeritz, R. E. Recent progress in understanding the natural and clinical histories of the Marfan syndrome. Trends Cardiovasc. Med. 26, 423–428 (2016). 32. Deng, H. et al. Identification of a novel missense FBN2 mutation in a Chinese family with congenital contractural arachnodactyly using exome sequencing. PLoS ONE 11, e0155908 (2016). 33. Pyeritz, R. E. & McKusick, V. A. The Marfan syndrome: diagnosis and management. N. Engl. J. Med. 300, 772–777 (1979). 34. Loeys, B. L. et al. Aneurysm syndromes caused by mutations in the TGF-β receptor. N. Engl. J. Med. 355, 788–798 (2006). 35. Habashi, J. P. et al. Losartan, an AT1 antagonist, prevents aortic aneurysm in a mouse model of Marfan syndrome. Science 312, 117–121 (2006). 36. Michelena, H. I. et al. Incidence of aortic complications in patients with bicuspid aortic valves. JAMA 306, 1104–1112 (2011). 37. Guo, D. C. et al. LOX mutations predispose to thoracic aortic aneurysms and dissections. Circ. Res. 118, 928–934 (2016). 38. Kuang, S. Q. et al. FOXE3 mutations predispose to thoracic aortic aneurysms and dissections. J. Clin. Invest. 126, 948–961 (2016). 39. Barbier, M. et al. MFAP5 loss‑of‑function mutations underscore the involvement of matrix alteration in the pathogenesis of familial thoracic aortic aneurysms and dissections. Am. J. Hum. Genet. 95, 736–743 (2014). 40. Davies, R. R. et al. Yearly rupture or dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann. Thorac Surg. 73, 17–27; discussion 27–28 (2002). 41. Kim, J. B. et al. Risk of rupture or dissection in descending thoracic aortic aneurysm. Circulation 132, 1620–1629 (2015). 42. Pape, L. A. et al. Aortic diameter >or = 5.5 cm is not a good predictor of type A aortic dissection: observations from the International Registry of Acute Aortic Dissection (IRAD). Circulation 116, 1120–1127 (2007). This is an interesting report from the International Registry for Aortic Dissection supporting the observation that the ascending aorta dissects at a normal diameter or at a diameter of<55 mm in half of patients with Stanford type A aortic dissection.

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 15 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER 43. Humphrey, J. D., Schwartz, M. A., Tellides, G. & Milewicz, D. M. Role of mechanotransduction in vascular biology: focus on thoracic aortic aneurysms and dissections. Circ. Res. 116, 1448–1461 (2015). 44. Ye, P. et al. GM‑CSF contributes to aortic aneurysms resulting from SMAD3 deficiency. J. Clin. Invest. 123, 2317–2331 (2013). 45. Bellini, C., Wang, S., Milewicz, D. M. & Humphrey, J. D. Myh11(R247C/R247C) mutations increase thoracic aorta vulnerability to intramural damage despite a general biomechanical adaptivity. J. Biomech. 48, 113–121 (2015). 46. Son, B. K. et al. Granulocyte macrophage colonystimulating factor is required for aortic dissection/ intramural haematoma. Nat. Commun. 6, 6994 (2015). 47. Ju, X. et al. Interleukin‑6–signal transducer and activator of transcription‑3 signaling mediates aortic dissections induced by angiotensin II via the T‑helper lymphocyte 17–interleukin 17 axis in C57BL/6 mice. Arterioscler. Thromb. Vasc. Biol. 33, 1612–1621 (2013). 48. Anzai, A. et al. Adventitial CXCL1/G‑CSF expression in response to acute aortic dissection triggers local neutrophil recruitment and activation leading to aortic rupture. Circ. Res. 116, 612–623 (2015). 49. Boyle, J. J., Weissberg, P. L. & Bennett, M. R. Tumor necrosis factor-α promotes macrophage-induced vascular smooth muscle cell apoptosis by direct and autocrine mechanisms. Arterioscler. Thromb. Vasc. Biol. 23, 1553–1558 (2003). 50. Ju, X. et al. IL‑6 regulates extracellular matrix remodeling associated with aortic dilation in a fibrillin‑1 hypomorphic mgR/mgR mouse model of severe Marfan syndrome. J. Am. Heart Assoc. 3, e000476 (2014). 51. Nienaber, C. A. & Eagle, K. A. Aortic dissection: new frontiers in diagnosis and management: part II: therapeutic management and follow‑up. Circulation 108, 772–778 (2003). 52. Clough, R. E., Waltham, M., Giese, D., Taylor, P. R. & Schaeffter, T. A new imaging method for assessment of aortic dissection using four-dimensional phase contrast magnetic resonance imaging. J. Vasc. Surg. 55, 914–923 (2012). This article introduces 4D phase-contrast MRI for functional imaging of aortic dissection. 53. Frauenfelder, T. et al. Triple rule-out CT in the emergency department: protocols and spectrum of imaging findings. Eur. Radiol. 19, 789–799 (2009). 54. Moore, A. G. et al. Choice of computed tomography, transesophageal echocardiography, magnetic resonance imaging, and aortography in acute aortic dissection: International Registry of Acute Aortic Dissection (IRAD). Am. J. Cardiol. 89, 1235–1238 (2002). 55. Elefteriades, J. A. et al. Indications and imaging for aortic surgery: size and other matters. J. Thorac Cardiovasc. Surg. 149, S10–S13 (2015). 56. Redheuil, A. et al. Age-related changes in aortic arch geometry: relationship with proximal aortic function and left ventricular mass and remodeling. J. Am. Coll. Cardiol. 58, 1262–1270 (2011). 57. Lederle, F. A. et al. Variability in measurement of abdominal aortic aneurysms. Abdominal Aortic Aneurysm Detection and Management Veterans Administration Cooperative Study Group. J. Vasc. Surg. 21, 945–952 (1995). 58. Singh, K. et al. Intra- and interobserver variability in the measurements of abdominal aortic and common iliac artery diameter with computed tomography. The Tromsø study. Eur. J. Vasc. Endovasc. Surg. 25, 399–407 (2003). 59. Cayne, N. S. et al. Variability of maximal aortic aneurysm diameter measurements on CT scan: significance and methods to minimize. J. Vasc. Surg. 39, 811–815 (2004). 60. Wen, D., Du, X., Dong, J. Z., Zhou, X. L. & Ma, C. S. Value of D‑dimer and C reactive protein in predicting inhospital death in acute aortic dissection. Heart 99, 1192–1197 (2013). This article provides an interesting analysis of D‑dimer and C-reactive protein as common biomarkers to predict in‑hospital death from acute aortic dissection. 61. Okina, N. et al. Utility of measuring C‑reactive protein for prediction of in‑hospital events in patients with acute aortic dissection. Heart Vessels 28, 330–335 (2013). 62. Jing, Q. et al. Angiotensin-converting enzyme I/D polymorphism and the risk of thoracic aortic

dissection in Chinese Han population. Mol. Biol. Rep. 40, 1249–1254 (2013). 63. Treasure, T. & Pepper, J. The potential of personalized external aortic root support (PEARS) to restore and maintain aortic valve competence. Ann. Thorac Surg. 100, 2415 (2015). This paper describes the introduction of a personalized external aortic root support to prevent enlargement of the ascending aorta in aortopathy. 64. Trimarchi, S. et al. Descending aortic diameter of 5.5 cm or greater is not an accurate predictor of acute type B aortic dissection. J. Thorac Cardiovasc. Surg. 142, e101–e107 (2011). 65. Mehta, R. H. et al. Acute type A aortic dissection in the elderly: clinical characteristics, management, and outcomes in the current era. J. Am. Coll. Cardiol. 40, 685–692 (2002). 66. Cooley, D. A. & De Bakey, M. E. Resection of entire ascending aorta in fusiform aneurysm using cardiac bypass. J. Am. Med. Assoc. 162, 1158–1159 (1956). 67. DeBakey, M. E. et al. Surgical management of dissecting aneurysms of the aorta. J. Thorac Cardiovasc. Surg. 49, 130–149 (1965). 68. Wheat, M. W. Jr, Wilson, J. R. & Bartley, T. D. Successful replacement of the entire ascending aorta and aortic valve. JAMA 188, 717–719 (1964). 69. Bentall, H. & De Bono, A. A technique for complete replacement of the ascending aorta. Thorax 23, 338–339 (1968). 70. Yacoub, M. H. et al. Late results of a valve-preserving operation in patients with aneurysms of the ascending aorta and root. J. Thorac Cardiovasc. Surg. 115, 1080–1090 (1998). 71. David, T. E. & Feindel, C. M. An aortic valve-sparing operation for patients with aortic incompetence and aneurysm of the ascending aorta. J. Thorac Cardiovasc. Surg. 103, 617–621; discussion 622 (1992). This article reports the introduction of the surgical aortic valve-sparing concept for patients with aneurysm of the ascending aorta. 72. Cabrol, C., Gandjbakhch, I. & Cham, B. [Aneurysms of the ascending aorta; total replacement with reimplantation of the coronary arteries (author’s transl)]. Nouv. Presse Med. 7, 363–365 (in French) (1978). 73. Karck, M. et al. The frozen elephant trunk technique: a new treatment for thoracic aortic aneurysms. J. Thorac Cardiovasc. Surg. 125, 1550–1553 (2003). 74. Sayer, D. et al. Aortic morphology following endovascular repair of acute and chronic type B aortic dissection: implications for management. Eur. J. Vasc. Endovasc. Surg. 36, 522–529 (2008). 75. Dake, M. D. et al. Endovascular stent-graft placement for the treatment of acute aortic dissection. N. Engl. J. Med. 340, 1546–1552 (1999). 76. Nienaber, C. A. et al. Nonsurgical reconstruction of thoracic aortic dissection by stent-graft placement. N. Engl. J. Med. 340, 1539–1545 (1999). 77. Cochennec, F. et al. Hybrid repair of aortic arch dissections. J. Vasc. Surg. 57, 1560–1567 (2013). This paper outlines endovascular stenting and associated vascular re‑routing surgery for the treatment of dissection that encroaches onto the aortic arch. 78. Canaud, L. et al. Retrograde aortic dissection after thoracic endovascular aortic repair. Ann. Surg. 260, 389–395 (2014). 79. Lombardi, J. V. et al. Prospective multicenter clinical trial (STABLE) on the endovascular treatment of complicated type B aortic dissection using a composite device design. J. Vasc. Surg. 55, 629–640.e2 (2012). 80. Nienaber, C. A. et al. Provisional extension to induce complete attachment after stent-graft placement in type B aortic dissection: the PETTICOAT concept. J. Endovasc. Ther. 13, 738–746 (2006). 81. VIRTUE Registry Investigators. Mid-term outcomes and aortic remodelling after thoracic endovascular repair for acute, subacute, and chronic aortic dissection: the VIRTUE Registry. Eur. J. Vasc. Endovasc. Surg. 48, 363–371 (2014). 82. Fattori, R. et al. Complicated acute type B dissection: is surgery still the best option?: A report from the International Registry of Acute Aortic Dissection. JACC Cardiovasc. Interv. 1, 395–402 (2008). 83. Parker, J. D. & Golledge, J. Outcome of endovascular treatment of acute type B aortic dissection. Ann. Thorac Surg. 86, 1707–1712 (2008).

84. Patterson, B. et al. Aortic pathology determines midterm outcome after endovascular repair of the thoracic aorta: report from the Medtronic Thoracic Endovascular Registry (MOTHER) database. Circulation 127, 24–32 (2013). 85. Conrad, M. F. et al. Late aortic remodeling persists in the stented segment after endovascular repair of acute complicated type B aortic dissection. J. Vasc. Surg. 62, 600–605 (2015). 86. Trimarchi, S. et al. Role and results of surgery in acute type B aortic dissection: insights from the International Registry of Acute Aortic Dissection (IRAD). Circulation 114, I357–I364 (2006). 87. Bozinovski, J. & Coselli, J. S. Outcomes and survival in surgical treatment of descending thoracic aorta with acute dissection. Ann. Thorac Surg. 85, 965–970; discussion 970–971 (2008). 88. Corvera, J. S. & Fehrenbacher, J. W. Open repair of chronic aortic dissections using deep hypothermia and circulatory arrest. Ann. Thorac Surg. 94, 78–81; discussion 82–83 (2012). 89. Kouchoukos, N. T., Kulik, A. & Castner, C. F. Outcomes after thoracoabdominal aortic aneurysm repair using hypothermic circulatory arrest. J. Thorac Cardiovasc. Surg. 145, S139–S141 (2013). 90. Mommertz, G. et al. Thoracoabdominal aortic aneurysm repair in patients with Marfan syndrome. Eur. J. Vasc. Endovasc. Surg. 35, 181–186 (2008). 91. Pujara, A. C. et al. Open repair of chronic distal aortic dissection in the endovascular era: implications for disease management. J. Thorac Cardiovasc. Surg. 144, 866–873 (2012). 92. Estrera, A. L. et al. Early and late outcomes after complete aortic replacement. Ann. Thorac Surg. 100, 528–534 (2015). 93. Tian, D. H., De Silva, R. P., Wang, T. & Yan, T. D. Open surgical repair for chronic type B aortic dissection: a systematic review. Ann. Cardiothorac. Surg. 3, 340–350 (2014). 94. Coselli, J. S. et al. Mortality and paraplegia after thoracoabdominal aortic aneurysm repair: a risk factor analysis. Ann. Thorac Surg. 69, 409–414 (2000). 95. Geisbusch, P. et al. Thoracic aortic endografting in patients with connective tissue diseases. J. Endovasc. Ther. 15, 144–149 (2008). 96. Drinkwater, S. L. et al. The incidence of spinal cord ischaemia following thoracic and thoracoabdominal aortic endovascular intervention. Eur. J. Vasc. Endovasc. Surg. 40, 729–735 (2010). 97. Bisdas, T., Panuccio, G., Sugimoto, M., Torsello, G. & Austermann, M. Risk factors for spinal cord ischemia after endovascular repair of thoracoabdominal aortic aneurysms. J. Vasc. Surg. 61, 1408–1416 (2015). 98. Griepp, E. B. et al. The anatomy of the spinal cord collateral circulation. Ann. Cardiothorac Surg. 1, 350–357 (2012). 99. Verhoeven, E. L. et al. Editor’s choice — ten-year experience with endovascular repair of thoracoabdominal aortic aneurysms: results from 166 consecutive patients. Eur. J. Vasc. Endovasc. Surg. 49, 524–531 (2015). 100. Augoustides, J. G., Stone, M. E. & Drenger, B. Novel approaches to spinal cord protection during thoracoabdominal aortic interventions. Curr. Opin. Anaesthesiol. 27, 98–105 (2014). 101. Griepp, R. B. & Di Luozzo, G. Hypothermia for aortic surgery. J. Thorac Cardiovasc. Surg. 145, S56–S58 (2013). 102. Zipfel, B. et al. Spinal cord ischemia after thoracic stent-grafting: causes apart from intercostal artery coverage. Ann. Thorac Surg. 96, 31–38 (2013). 103. Etz, C. D. et al. Staged repair significantly reduces paraplegia rate after extensive thoracoabdominal aortic aneurysm repair. J. Thorac Cardiovasc. Surg. 139, 1464–1472 (2010). This article reports that the staged repair of thoraco-abdominal aneurysm reduces the rate of paraplegia in this condition. 104. Etz, C. D., Debus, E. S., Mohr, F. W. & Kolbel, T. First‑in‑man endovascular preconditioning of the paraspinal collateral network by segmental artery coil embolization to prevent ischemic spinal cord injury. J. Thorac Cardiovasc. Surg. 149, 1074–1079 (2015). 105. Lohse, F. et al. Quality of life after replacement of the ascending aorta in patients with true aneurysms. Tex. Heart Inst. J. 36, 104–110 (2009). 106. Immer, F. F. et al. Improvement of quality of life after surgery on the thoracic aorta: effect of antegrade cerebral perfusion and short duration of deep hypothermic circulatory arrest. Circulation 110, II250–II255 (2004).

16 | 2016 | VOLUME 2

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

PRIMER 107. Chaddha, A. et al. Survivors of aortic dissection: activity, mental health, and sexual function. Clin. Cardiol. 38, 652–659 (2015). This is the first report on changes in physical activities, mental health and sexual function after aortic dissection. 108. Tang, G. H. et al. Surgery for acute type A aortic dissection in octogenarians is justified. J. Thorac Cardiovasc. Surg. 145, S186–S190 (2013). 109. Fattori, R. et al. Interdisciplinary expert consensus document on management of type B aortic dissection. J. Am. Coll. Cardiol. 61, 1661–1678 (2013). 110. Svensson, L. G. et al. Aortic valve and ascending aorta guidelines for management and quality measures. Ann. Thorac Surg. 95, S1–S66 (2013). This paper presents the most recent guidelines for managing disease of the aortic valve and ascending aorta, including measures on quality of life. 111. Nienaber, C. A., Divchev, D., Palisch, H., Clough, R. E. & Richartz, B. Early and late management of type B aortic dissection. Heart 100, 1491–1497 (2014). 112. Suzuki, T. et al. Type-selective benefits of medications in treatment of acute aortic dissection (from the International Registry of Acute Aortic Dissection [IRAD]). Am. J. Cardiol. 109, 122–127 (2012). 113. Sakalihasan, N. et al. (Tissue PET) Vascular metabolic imaging and peripheral plasma biomarkers in the evolution of chronic aortic dissections. Eur. Heart J. Cardiovasc. Imaging 16, 626–633 (2015). This paper describes a study that found that PET imaging of aortic wall inflammation correlates with biomarker activity and may have prognostic usefulness in the chronic phase of aortic dissection. 114. Hertault, A. et al. Benefits of completion 3D angiography associated with contrast enhanced ultrasound to assess technical success after EVAR. Eur. J. Vasc. Endovasc. Surg. 49, 541–548 (2015). 115. Ulug, P., McCaslin, J. E., Stansby, G. & Powell, J. T. Endovascular versus conventional medical treatment for uncomplicated chronic type B aortic dissection. Cochrane Database Syst. Rev. 11, CD006512 (2012). 116. Metcalfe, M. J. et al. The first endovascular repair of an acute type A dissection using an endograft

designed for the ascending aorta. J. Vasc. Surg. 55, 220–222 (2012). 117. Sobocinski, J. et al. Endovascular approaches to acute aortic type A dissection: a CT‑based feasibility study. Eur. J. Vasc. Endovasc. Surg. 42, 442–447 (2011). This paper describes radiographic evaluation of CT images in proximal aortic dissection, with results suggesting that endovascular treatment is suitable for approximately 50% of patients. 118. Moon, M. C. et al. Computed tomography-based anatomic characterization of proximal aortic dissection with consideration for endovascular candidacy. J. Vasc. Surg. 53, 942–949 (2011). 119. Lu, Q. et al. Endovascular repair of ascending aortic dissection: a novel treatment option for patients judged unfit for direct surgical repair. J. Am. Coll. Cardiol. 61, 1917–1924 (2013). This paper describes the introduction of endovascular repair of the ascending aorta in cases of Stanford type A aortic dissection that are not suitable for surgical repair. 120. Idrees, J., Roselli, E. E., Shafii, S., Reside, J. & Lytle, B. W. Outcomes after false lumen embolization with covered stent devices in chronic dissection. J. Vasc. Surg. 60, 1507–1513 (2014). 121. Oikonomou, K. et al. Outcomes of fenestrated/ branched endografting in post-dissection thoracoabdominal aortic aneurysms. Eur. J. Vasc. Endovasc. Surg. 48, 641–648 (2014). 122. Sobocinski, J. et al. Treatment options for postdissection aortic aneurysms. J. Cardiovasc. Surg. (Torino) 57, 202–211 (2016). 123. Bonatti, J., Vetrovec, G., Riga, C., Wazni, O. & Stadler, P. Robotic technology in cardiovascular medicine. Nat. Rev. Cardiol. 11, 266–275 (2014). 124. Saratzis, A. & Bown, M. J. The genetic basis for aortic aneurysmal disease. Heart 100, 916–922 (2014). 125. Callewaert, B., De Paepe, A. & Coucke, P. Arterial tortuosity syndrome. GeneReviews http:// www.ncbi.nlm.nih.gov/books/NBK253404/ (updated 13 Nov 2014). 126. Schubert, J. A., Landis, B. J., Shikany, A. R., Hinton, R. B. & Ware, S. M. Clinically relevant variants

identified in thoracic aortic aneurysm patients by research exome sequencing. Am. J. Med. Genet. A 170, 1288–1294 (2016). 127. Guo, D. C. et al. MAT2A mutations predispose individuals to thoracic aortic aneurysms. Am. J. Hum. Genet. 96, 170–177 (2015). 128. Guo, D. C. et al. Recurrent gain‑of‑function mutation in PRKG1 causes thoracic aortic aneurysms and acute aortic dissections. Am. J. Hum. Genet. 93, 398–404 (2013). 129. Pyeritz, R. E. Heritable thoracic aortic disorders. Curr. Opin. Cardiol. 29, 97–102 (2014). 130. Clough, R. E., Zymvragoudakis, V. E., Biasi, L. & Taylor, P. R. Usefulness of new imaging methods for assessment of type B aortic dissection. Ann. Cardiothorac Surg. 3, 314–318 (2014). 131. Nienaber, C. A. & Powell, J. T. Management of acute aortic syndromes. Eur. Heart J. 33, 26–35b (2012). 132. LeMaire, S. A. & Russell, L. Epidemiology of thoracic aortic dissection. Nat. Rev. Cardiol. 8, 103–113 (2011). 133. Donati, T., Wilson, J., Kolbel, T. & Clough, R. E. Modern diagnostics for type B aortic dissection. Gefasschirurgie 20, 420–427 (2015). 134. Baldwin, A. K., Simpson, A., Steer, R., Cain, S. A. & Kielty, C. M. Elastic fibres in health and disease. Expert Rev. Mol. Med. 15, e8 (2013). 135. Shalhub, S., Dua, A. & Brooks, J. Biomarkers in descending thoracic aortic dissection. Semin. Vasc. Surg. 27, 196–199 (2014).

Author contributions

Introduction (C.A.N. and J.S.M.Y.); Epidemiology (F.M. and J.S.M.Y.); Mechanisms/pathophysiology (T.S. and J.S.M.Y.); Diagnosis, screening and prevention (C.A.N., R.E.C., A.E. and J.S.M.Y.); Management (C.A.N., N.S., R.G., M.P.J., M.M.T., J.S.M.Y. and J.P.); Quality of life (J.S.M.Y. and U.R.); Outlook (C.A.N., R.E.C., J.S.M.Y. and N.C.); Overview of Primer (C.A.N.).

Competing interests statement

The authors declare no competing interests.

NATURE REVIEWS | DISEASE PRIMERS

VOLUME 2 | 2016 | 17 . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©

CORRECTION CORRECTION

Aortic dissection Christoph A. Nienaber, Rachel E. Clough, Natzi Sakalihasan, Toru Suzuki, Richard Gibbs, Firas Mussa, Michael P. Jenkins, Matt M. Thompson, Arturo Evangelista, James S. M. Yeh, Nicholas Cheshire, Ulrich Rosendahl and John Pepper Nature Reviews Disease Primers 2, 16053 (2016)

In the version of the article originally published, Michael P. Jenkins was incorrectly stated as Michael T. Jenkins. The article has now been corrected.

NATURE REVIEWS | DISEASE PRIMERS

www.nature.com/nrdp . d e v r e s e r s t h g i r l l A . e r u t a N r e g n i r p S f o t r a p , d e t i m i L s r e h s i l b u P n a l l i m c a M 6 1 0 2 ©