Author + information
- Published online April 6, 2010.
- American College of Cardiology Foundation,
- American Heart Association Task Force on Practice Guidelines,
- American Association for Thoracic Surgery,
- American College of Radiology,
- American Stroke Association,
- Society of Cardiovascular Anesthesiologists,
- Society for Cardiovascular Angiography and Interventions,
- Society of Interventional Radiology,
- Society of Thoracic Surgeons,
- Society for Vascular Medicine,
- North American Society for Cardiovascular Imaging,
- Loren F. Hiratzka, MD, Chair*,
- George L. Bakris, MD, WRITING GROUP MEMBER†,
- Joshua A. Beckman, MD, MS, WRITING GROUP MEMBER‡,
- Robert M. Bersin, MD, WRITING GROUP MEMBER§,
- Vincent F. Carr, DO, WRITING GROUP MEMBER‖,
- Donald E. Casey Jr, MD, MPH, MBA, WRITING GROUP MEMBER¶,
- Kim A. Eagle, MD, WRITING GROUP MEMBER*,#,
- Luke K. Hermann, MD, WRITING GROUP MEMBER**,
- Eric M. Isselbacher, MD, WRITING GROUP MEMBER*,
- Ella A. Kazerooni, MD, MS, WRITING GROUP MEMBER††,
- Nicholas T. Kouchoukos, MD, WRITING GROUP MEMBER‡‡,
- Bruce W. Lytle, MD, WRITING GROUP MEMBER§§,
- Dianna M. Milewicz, MD, PhD, WRITING GROUP MEMBER,
- David L. Reich, MD, WRITING GROUP MEMBER‖‖,
- Souvik Sen, MD, MS, WRITING GROUP MEMBER¶¶,
- Julie A. Shinn, RN, MA, CCRN, WRITING GROUP MEMBER†,
- Lars G. Svensson, MD, PhD, WRITING GROUP MEMBER## and
- David M. Williams, MD, WRITING GROUP MEMBER#,***
- ACC/AHA Clinical Practice Guideline
- thoracic aortic disease
- thoracic aortic dissection
- thoracic aortic aneurysm
- intramural hematoma
- genetic syndromes associated with thoracic aortic aneurysm
- emergency department
- acute thoracic aortic disease presentation and evaluation
ACCF/AHA Task Force Members
Alice K. Jacobs, MD, FACC, FAHA, Chair 2009–2011; Sidney C. Smith, Jr, MD, FACC, FAHA, Immediate Past Chair 2006–2008†††; Jeffery L. Anderson, MD, FACC, FAHA, Chair-Elect; Cynthia D. Adams, MSN, PhD, FAHA†††; Christopher E. Buller, MD, FACC; Mark A. Creager, MD, FACC, FAHA; Steven M. Ettinger, MD, FACC; Robert A. Guyton, MD, FACC, FAHA; Jonathan L. Halperin, MD, FACC, FAHA; Sharon A. Hunt, MD, FACC, FAHA†††; Harlan M. Krumholz, MD, FACC, FAHA†††; Frederick G. Kushner, MD, FACC, FAHA; Bruce W. Lytle, MD, FACC, FAHA†††; Rick Nishimura, MD, FACC, FAHA†††; Richard L. Page, MD, FACC, FAHA†††; Barbara Riegel, DNSc, RN, FAHA***; William G. Stevenson, MD, FACC, FAHA; Lynn G. Tarkington, RN; Clyde W. Yancy, MD, FACC, FAHA
Table of Contents
1.1 Methodology and Evidence Review.......e31
1.2 Organization of the Writing Committee.......e32
1.3 Document Review and Approval.......e32
1.4 Scope of the Guideline.......e33
1.4.1 Critical Issues.......e35
1.5 Glossary of Terms and Abbreviations Used Throughout Guideline.......e35
2. The Thoracic Aorta.......e36
2.1 The Normal Aorta.......e36
2.2 Normal Thoracic Aortic Diameter.......e36
3. Thoracic Aortic Histopathology.......e37
3.2 Aneurysms and Dissections.......e37
3.3 Vasculitis and Inflammatory Diseases.......e38
4. Imaging Modalities.......e39
4.1 Recommendations for Aortic Imaging Techniques to Determine the Presence and Progression of Thoracic Aortic Disease.......e39
4.2 Chest X-Ray.......e40
4.3 Computed Tomographic Imaging.......e40
4.3.1 Computed Tomographic Imaging Technique.......e42
4.4 Magnetic Resonance Imaging.......e42
4.4.1 Magnetic Resonance Imaging Technique .......e43
4.4.2 Black Blood Imaging.......e43
4.4.3 Noncontrast White Blood Imaging.......e43
4.4.4 Contrast-Enhanced Magnetic Resonance Angiography.......e43
4.4.5 Phase Contrast Imaging.......e43
4.5 Standards for Reporting of the Thoracic Aorta on Computed Tomography and Magnetic Resonance Imaging.......e43
4.7.1 Echocardiographic Criteria for Thoracic Aortic Aneurysms.......e45
4.7.2 Echocardiographic Criteria for Aortic Dissection.......e45
22.214.171.124 Diagnostic Accuracy of Echocardiography for Aortic Dissection.......e46
126.96.36.199 Diagnostic Accuracy of Echocardiography for Acute Intramural Hematoma.......e46
188.8.131.52 Role of Echocardiography in Following Patients With Chronic Aortic Disease.......e47
5. Genetic Syndromes Associated With Thoracic Aortic Aneurysms and Dissections.......e47
5.1 Recommendations for Genetic Syndromes.......e47
5.1.1 Marfan Syndrome.......e48
5.1.2 Loeys-Dietz Syndrome.......e49
5.1.3 Ehlers-Danlos Syndrome, Vascular Form or Type IV.......e49
5.1.4 Turner Syndrome.......e49
5.1.5 Other Genetic Syndromes With Increased Risk for Thoracic Aortic Aneurysms and Dissections.......e50
5.1.6 Recommendations for Familial Thoracic Aortic Aneurysms and Dissections.......e50
6. Other Cardiovascular Conditions Associated With Thoracic Aortic Aneurysm and Dissection.......e52
6.1 Recommendations for Bicuspid Aortic Valve and Associated Congenital Variants in Adults.......e52
6.2 Aberrant Right Subclavian Artery.......e53
6.3 Coarctation of the Aorta.......e53
6.4 Right Aortic Arch.......e53
7. Inflammatory Diseases Associated With Thoracic Aortic Disease.......e53
7.1 Recommendations for Takayasu Arteritis and Giant Cell Arteritis.......e53
7.2 Takayasu Arteritis.......e54
7.3 Giant Cell Arteritis.......e56
7.4 Behçet Disease.......e57
7.5 Ankylosing Spondylitis (Spondyloarthropathies) .......e57
7.6 Infective Thoracic Aortic Aneurysms.......e57
8. Acute Aortic Syndromes.......e58
8.1 Aortic Dissection.......e58
8.1.1 Aortic Dissection Definition.......e58
8.1.2 Anatomic Classification of Aortic Dissection.......e58
8.1.3 Risk Factors for Aortic Dissection.......e61
8.1.4 Clinical Presentation of Acute Thoracic Aortic Dissection.......e62
184.108.40.206 Symptoms of Acute Thoracic Aortic Dissection.......e62
220.127.116.11 Perfusion Deficits and End-Organ Ischemia.......e62
8.1.5 Cardiac Complications.......e64
18.104.22.168 Acute Aortic Regurgitation.......e64
22.214.171.124 Myocardial Ischemia or Infarction.......e64
126.96.36.199 Heart Failure and Shock.......e64
188.8.131.52 Pericardial Effusion and Tamponade.......e64
8.1.7 Neurologic Complications.......e65
8.1.8 Pulmonary Complications.......e65
8.1.9 Gastrointestinal Complications.......e65
8.1.10 Blood Pressure and Heart Rate Considerations.......e65
8.1.11 Age and Sex Considerations.......e65
8.2 Intramural Hematoma.......e66
8.3 Penetrating Atherosclerotic Ulcer.......e67
8.4 Pseudoaneurysms of the Thoracic Aorta.......e67
8.5 Traumatic Rupture of the Thoracic Aorta.......e67
8.6 Evaluation and Management of Acute Thoracic Aortic Disease.......e68
8.6.1 Initial Evaluation and Management.......e68
184.108.40.206 Recommendations for Estimation of Pretest Risk of Thoracic Aortic Dissection.......e68
220.127.116.11 Laboratory Testing.......e68
18.104.22.168 Recommendations for Screening Tests.......e69
22.214.171.124 Recommendations for Diagnostic Imaging Studies.......e69
126.96.36.199 Recommendations for Initial Management.......e69
188.8.131.52 Recommendations for Definitive Management.......e69
8.6.2 Evaluation and Management Algorithms.......e70
8.6.3 Initial Management.......e71
184.108.40.206 Blood Pressure and Rate Control Therapy.......e71
220.127.116.11 Additional Antihypertensive Therapy.......e73
18.104.22.168 Pain Control.......e73
22.214.171.124 Determining Definitive Management.......e73
8.6.4 Recommendation for Surgical Intervention for Acute Thoracic Aortic Dissection.......e73
8.6.5 Endovascular Interventions.......e73
8.6.6 Principles of Treatment for Intramural Hematoma and Penetrating Atherosclerotic Ulcer.......e74
126.96.36.199 Intimal Defect Without Intramural Hematoma.......e74
188.8.131.52 Intimal Defect With Intramural Hematoma.......e74
184.108.40.206 Recommendation for Intramural Hematoma Without Intimal Defect.......e74
8.7 Treatment for the Management of Traumatic Aortic Rupture.......e74
9. Thoracic Aortic Aneurysms.......e75
9.1 General Approach to the Patient.......e76
9.1.1 Recommendation for History and Physical Examination for Thoracic Aortic Disease.......e76
220.127.116.11 Coronary Artery Disease.......e77
18.104.22.168 Associated Renal Ischemia.......e78
22.214.171.124 Associated Mesenteric Ischemia.......e78
126.96.36.199 Associated Peripheral Ischemia.......e78
9.1.2 Differential Diagnosis.......e78
188.8.131.52 Physical Findings.......e78
9.1.3 Considerations for Imaging.......e79
9.2 General Medical Treatment and Risk Factor Management for Patients With Thoracic Aortic Disease.......e79
9.2.1 Recommendation for Medical Treatment of Patients With Thoracic Aortic Diseases.......e79
184.108.40.206 Recommendations for Blood Pressure Control.......e80
220.127.116.11 Recommendation for Dyslipidemia.......e80
18.104.22.168 Recommendation for Smoking Cessation.......e80
9.2.2 Surgical and Endovascular Treatment by Location of Disease.......e81
22.214.171.124 Ascending Aorta and Aortic Sinuses.......e81
126.96.36.199.1 Recommendations for Asymptomatic Patients With Ascending Aortic Aneurysm.......e81
188.8.131.52.2 Recommendation for Symptomatic Patients With Thoracic Aortic Aneurysm.......e81
184.108.40.206.3 Endovascular Grafting for Ascending Aortic Aneurysm.......e82
220.127.116.11.4 Recommendations for Open Surgery for Ascending Aortic Aneurysm.......e82
18.104.22.168 Recommendations for Aortic Arch Aneurysms.......e84
22.214.171.124.1 Open Surgery.......e84
126.96.36.199 Descending Thoracic Aorta and Thoracoabdominal Aorta.......e84
188.8.131.52.1 Recommendations for Descending Thoracic Aorta and Thoracoabdominal Aortic Aneurysms.......e84
184.108.40.206.2 Endovascular Versus Open Surgical Approach.......e85
220.127.116.11.3 End-Organ Preservation During Thoracic Endograft Implantation.......e86
18.104.22.168.4 Periprocedural Complications of Endograft Procedures.......e87
22.214.171.124.5 Open Surgical.......e88
126.96.36.199.6 End-Organ Preservation During Open Thoracoabdominal Repairs.......e89
188.8.131.52.7 Aortic Dissection With Malperfusion.......e89
10. Special Considerations in Pregnant Patients With Aortic Disease.......e89
10.1 Effects of Pregnancy on the Aorta.......e89
10.2 Epidemiology of Chronic and Acute Aortic Conditions in Pregnancy.......e89
10.3 Counseling and Management of Chronic Aortic Diseases in Pregnancy.......e90
10.3.1 Recommendations for Counseling and Management of Chronic Aortic Diseases in Pregnancy.......e89
10.4 Evaluation and Management of Acute Aortic Syndromes During Pregnancy.......e90
11. Aortic Arch and Thoracic Aortic Atheroma and Atheroembolic Disease.......e90
11.1 Recommendations for Aortic Arch and Thoracic Aortic Atheroma and Atheroembolic Disease.......e90
11.2 Clinical Description.......e90
11.3 Risk Factors.......e91
11.5.1 Anticoagulation Versus Antiplatelet Therapy.......e91
11.5.2 Lipid-Lowering Agent.......e92
11.5.3 Surgical and Interventional Approaches.......e92
12. Porcelain Aorta.......e92
13. Tumors of the Thoracic Aorta.......e93
14. Perioperative Care for Open Surgical and Endovascular Thoracic Aortic Repairs.......e93
14.1 Recommendations for Preoperative Evaluation.......e93
14.1.1 Preoperative Risk Assessment.......e94
14.2 Recommendations for Choice of Anesthetic and Monitoring Techniques.......e95
14.2.1 Temperature Monitoring.......e95
14.2.2 Hemodynamic Monitoring.......e95
14.2.3 Transesophageal Echocardiography.......e96
14.2.4 Transesophageal Echocardiography for Endovascular Repairs of the Descending Thoracic Aorta.......e96
14.3 Airway Management for Descending Thoracic Aortic Repairs.......e96
14.4 Recommendation for Transfusion Management and Anticoagulation in Thoracic Aortic Surgery.......e96
14.5 Organ Protection.......e97
14.5.1 Recommendations for Brain Protection During Ascending Aortic and Transverse Aortic Arch Surgery.......e97
14.5.2 Recommendations for Spinal Cord Protection During Descending Aortic Open Surgical and Endovascular Repairs.......e98
184.108.40.206 Monitoring of Spinal Cord Function in Descending Thoracic Aortic Repairs.......e99
220.127.116.11 Maintenance of Spinal Cord Arterial Pressure.......e99
18.104.22.168 Cerebrospinal Fluid Pressure and Drainage.......e99
22.214.171.124 Glucocorticoids and Mannitol.......e100
14.5.3 Recommendations for Renal Protection During Descending Aortic Open Surgical and Endovascular Repairs.......e100
14.6 Complications of Open Surgical Approaches.......e100
14.7 Mortality Risk for Thoracic Aortic Surgery.......e101
14.8 Postprocedural Care.......e102
14.8.1 Postoperative Risk Factor Management.......e102
14.8.2 Recommendations for Surveillance of Thoracic Aortic Disease or Previously Repaired Patients.......e102
15. Nursing Care and Patient/Family Education.......e103
15.1 Nursing Care of Medically Managed Patients.......e103
15.2 Preprocedural Nursing Care.......e103
15.3 Postprocedural Nursing Care.......e103
15.4 Nursing Care of Surgically Managed Patients.......e104
16. Long-Term Issues.......e105
16.1 Recommendation for Employment and Lifestyle in Patients With Thoracic Aortic Disease.......e105
17. Institutional/Hospital Quality Concerns.......e106
17.1 Recommendations for Quality Assessment and Improvement for Thoracic Aortic Disease.......e106
17.2 Interinstitutional Issues.......e107
18. Future Research Directions and Issues.......e107
18.1 Risks and Benefits of Current Imaging Technologies.......e107
18.2 Mechanisms of Aortic Dissection.......e108
18.3 Treatment of Malperfusion and Reperfusion Injury.......e108
18.4 Gene-Based Mechanisms and Models.......e108
18.4.1 Aortic Disease Management Based on the Underlying Genetic Defects.......e108
18.4.2 Biomarkers for Acute Aortic Dissection.......e108
18.4.3 Genetic Defects and Molecular Pathway Analyses.......e108
18.4.4 Clinical Trials for Medical Therapy for Aortic Aneurysms.......e108
18.5 Aortic Atheroma and Atherosclerosis Identification and Treatment.......e108
18.6 Prediction Models of Aortic Rupture and the Need for Preemptive Interventions.......e108
18.7 National Heart, Lung, and Blood Institute Working Group Recommendations.......e108
Appendix 1. Author Relationships With Industry and Other Entities.......e126
Appendix 2. Reviewer Relationships With Industry and Other Entities.......e128
Appendix 3. Abbreviation List.......e130
It is essential that the medical profession play a central role in critically evaluating the evidence related to drugs, devices, and procedures for the detection, management, or prevention of disease. Properly applied, rigorous, expert analysis of the available data documenting absolute and relative benefits and risks of these therapies and procedures can improve outcomes and reduce costs of care by focusing resources on the most effective strategies. One important use of such data is the production of clinical practice guidelines that, in turn, can provide a foundation for a variety of other applications such as performance measures, appropriate use criteria, clinical decision support tools, and quality improvement tools.
The American College of Cardiology Foundation (ACCF) and the American Heart Association (AHA) have jointly engaged in the production of guidelines in the area of cardiovascular disease since 1980. The ACCF/AHA Task Force on Practice Guidelines is charged with developing, updating, and revising practice guidelines for cardiovascular diseases and procedures, and the Task Force directs and oversees this effort. Writing committees are charged with assessing the evidence as an independent group of authors to develop, update, or revise recommendations for clinical practice.
Experts in the subject under consideration have been selected from both organizations to examine subject-specific data and write guidelines in partnership with representatives from other medical practitioner and specialty groups. Writing committees are specifically charged to perform a formal literature review, weigh the strength of evidence for or against particular treatments or procedures, and include estimates of expected health outcomes where data exist. Patient-specific modifiers, comorbidities, and issues of patient preference that may influence the choice of tests or therapies are considered. When available, information from studies on cost is considered, but data on efficacy and clinical outcomes constitute the primary basis for recommendations in these guidelines.
The ACCF/AHA Task Force on Practice Guidelines makes every effort to avoid actual, potential, or perceived conflicts of interest that may arise as a result of industry relationships or personal interests among the writing committee. Specifically, all members of the writing committee, as well as peer reviewers of the document, are asked to disclose all current relationships and those 24 months prior to initiation of the writing effort that may be perceived as relevant. All guideline recommendations require a confidential vote by the writing committee and must be approved by a consensus of the members voting. Members who were recused from voting are noted on the title page of this document. Members must recuse themselves from voting on any recommendation where their relationships with industry (RWI) apply. If a writing committee member develops a new relationship with industry during his/her tenure, he/she is required to notify guideline staff in writing. These statements are reviewed by the Task Force on Practice Guidelines and all members during each conference call and/or meeting of the writing committee, updated as changes occur, and ultimately published as an appendix to the document. For detailed information regarding guideline policies and procedures, please refer to the methodology manual for ACCF/AHA Guideline Writing Committees (1). RWI and other entities pertinent to this guideline for authors and peer reviewers are disclosed in Appendixes 1 and 2, respectively. Disclosure information for the ACCF/AHA Task Force on Practice Guidelines is also available online at (http://www.acc.org/about/overview/ClinicalDocumentsTaskForces.cfm).
These practice guidelines are intended to assist healthcare providers in clinical decision making by describing a range of generally acceptable approaches for diagnosis, management, and prevention of specific diseases or conditions. Clinicians should consider the quality and availability of expertise in the area where care is provided. These guidelines attempt to define practices that meet the needs of most patients in most circumstances. The recommendations reflect a consensus after a thorough review of the available current scientific evidence and are intended to improve patient care. The Task Force recognizes that situations arise where additional data are needed to better inform patient care; these areas will be identified within each respective guideline when appropriate.
Patient adherence to prescribed and agreed upon medical regimens and lifestyles is an important aspect of treatment. Prescribed courses of treatment in accordance with these recommendations are effective only if they are followed. Because lack of patient understanding and adherence may adversely affect outcomes, physicians and other healthcare providers should make every effort to engage the patient's active participation in prescribed medical regimens and lifestyles.
If these guidelines are used as the basis for regulatory or payer decisions, the goal should be improvement in quality of care and aligned with the patient's best interest. The ultimate judgment regarding care of a particular patient must be made by the healthcare provider and the patient in light of all of the circumstances presented by that patient. Consequently, there are circumstances in which deviations from these guidelines are appropriate.
The guidelines will be reviewed annually by the ACCF/AHA Task Force on Practice Guidelines and considered current unless they are updated, revised, or withdrawn from distribution.
Alice K. Jacobs, MD, FACC, FAHA
Chair, ACCF/AHA Task Force on Practice Guidelines
Sidney C. Smith, Jr, MD, FACC, FAHA
Immediate Past Chair, ACCF/AHA Task Force on Practice Guidelines
1.1 Methodology and Evidence Review
The writing committee conducted a comprehensive search of the medical and scientific literature through the use ofPubMed/MEDLINE. Searches were limited to publications written in the English language. Compiled reports were reviewed and additional articles were provided by committee members. Specifically targeted searches were conducted on the following subtopics: acute aortic dissection, ankylosing spondylitis, aortic dissection and litigation, aortic neoplasm, aortic tumors, Behçet disease, bicuspid aortic valve, calcified aorta, chronic dissection, coarctation of the aorta, D-dimer, dissecting aneurysm, Ehlers-Danlos syndrome, endovascular and aortic aneurysms, medial degeneration, porcelain aorta, giant cell arteritis, imaging and thoracic aortic disease, inflammatory disease, intramural hematoma, Loeys-Dietz syndrome, Marfan syndrome, Noonan syndrome, penetrating aortic ulcer, polycystic kidney disease, thoracic and aortic aneurysms, thoracic aortic disease and patient care, thoracic aortic disease and surgery, thoracic aorta and Kawasaki disease, Takayasu arteritis, thoracoabdominal and aorta or aortic disease, and Turner syndrome. More than 850 references were reviewed, with 829 used as the primary evidence base for the final guideline. The ACCF/AHA Task Force on Practice Guidelines methodology processes were followed to write the text and recommendations. In general, published manuscripts appearing in journals listed in Index Medicus were used as the evidence base. Published abstracts were used only for emerging information but were not used in the formulation of recommendations.
The committee reviewed and ranked evidence supporting current recommendations with the weight of evidence ranked as Level A if the data were derived from multiple randomized clinical trials or meta-analyses. The committee ranked available evidence as Level B when data were derived from a single randomized trial or nonrandomized studies. Evidence was ranked as Level C when the primary source of the recommendation was consensus opinion, case studies, or standard of care. In the narrative portions of these guidelines, evidence is generally presented in chronological order of development. Studies are identified as observational, retrospective, prospective, or randomized. For certain conditions for which inadequate data are available, recommendations are based on expert consensus and clinical experience and are ranked as Level C. An analogous example is the use of penicillin for pneumococcal pneumonia, where there are no randomized trials and treatment is based on clinical experience. When recommendations at Level C are supported by historical clinical data, appropriate references (including clinical reviews) are cited if available. For issues where sparse data are available, a survey of current practice among the clinicians on the writing committee formed the basis for Level C recommendations and no references are cited. The schema for classification of recommendations and level of evidence is summarized in Table 1, which also illustrates how the grading system provides an estimate of the size of the treatment effect and an estimate of the certainty of the treatment effect.
To provide clinicians with a comprehensive set of data, whenever possible, the exact event rates in various treatment arms of clinical trials are presented to permit calculation of the absolute risk difference (ARD), number needed to harm (NNH); the relative treatment effects are described either as odds ratio (OR), relative risk (RR), or hazard ratio (HR) depending on the format in the original publication. Along with all other point statistics, confidence intervals (CIs) for those statistics are added when available.
The writing committee recognized that the evidence base for this guideline is less robust in terms of randomized clinical trials than prior ACCF/AHA guidelines, particularly those focused on coronary artery disease (CAD) and heart failure. As the reader will discern, much of the evidence base for this topic consists of cohort studies and retrospective reviews, which largely emanate from centers with a specialized interest in specific types of thoracic aortic disease. The writing committee attempted to focus on providing the practitioner with recommendations for evaluation and treatment wherever possible and where controversy exists, identified as such in the text.
The writing committee acknowledges the expertise of the highly experienced and effective practice guidelines staff of the ACCF and AHA. The writing committee chair also acknowledges the commitment and dedication of the diverse writing committee members who were able to put aside issues of specialty “turf” and focus on providing the medical community with a guideline aimed at optimal patient care.
1.2 Organization of the Writing Committee
The guideline was written by a committee comprised of experts in cardiovascular medicine, surgery, radiology, and nursing. For many of the previous ACCF/AHA practice guidelines, writing expertise has been available within these 2 organizations. Because of the broad scope and diversity of thoracic aortic diseases, as well as the specialists who treat such patients, the ACCF and AHA sought greater involvement from many specialty organizations. Most, but not all, specialty organizations that represent the major stakeholders caring for patients with thoracic aortic diseases provided writing committee members and financial support of the project, and they are recognized as marquee level partners with the ACCF and AHA. These organizations included the American Association for Thoracic Surgery (AATS), American College of Radiology (ACR), American Stroke Association (ASA), Society of Cardiovascular Anesthesiologists (SCA), Society for Cardiovascular Angiography and Interventions (SCAI), Society of Interventional Radiology (SIR), Society of Thoracic Surgeons (STS), and Society for Vascular Medicine (SVM). The American College of Emergency Physicians (ACEP) and the American College of Physicians (ACP) were also represented on the writing committee. Where additional expertise was needed, the scientific councils of the AHA were contacted for writing committee representatives. Representation was provided or facilitated by the Councils on Cardiovascular Nursing, Cardiovascular Surgery and Anesthesia, Cardiovascular Radiology and Intervention, and Clinical Cardiology, Council for High Blood Pressure Research, and Stroke Council.
1.3 Document Review and Approval
This document was reviewed by 3 outside reviewers nominated by the ACCF and 2 outside reviewers nominated by the AHA, as well as 1 or 2 reviewers from each of the following organizations: the AATS, ACP, ACEP, ACR, ASA, SCA, SCAI, SIR, STS, and the SVM. It was also reviewed by 6 individual content reviewers—2 content reviewers from the ACCF Catherization Committee and 1 content reviewer from the ACCF Interventional Council. All reviewer RWI information was collected and distributed to the writing committee and is published in this document (see Appendix 2).
This document was approved for publication by the governing bodies of the ACCF and the AHA and the AATS, ACR, ASA, SCA, SCAI, SIR, STS, and SVM and was endorsed by the North American Society for Cardiovascular Imaging.
1.4 Scope of the Guideline
The term “thoracic aortic disease” encompasses a broad range of degenerative, structural, acquired, genetic-based, and traumatic disease states and presentations. According to the Centers for Disease Control and Prevention death certificate data, diseases of the aorta and its branches account for 43 000 to 47 000 deaths annually in the United States (2). The precise number of deaths attributable to thoracic aortic diseases is unclear. However, autopsy studies suggest that the presentation of thoracic aortic disease is often death due to aortic dissection (AoD) and rupture, and these deaths account for twice as many deaths as attributed to ruptured abdominal aortic aneurysms (AAAs) (3). The diagnosis of acute thoracic AoD or rupture is often difficult and delayed, and errors in diagnosis may account for deaths otherwise attributed to cardiac arrhythmia, myocardial infarction (MI), pulmonary embolism, or mesenteric ischemia.
The University HealthSystem Consortium (UHC) is an alliance of more than 100 academic medical centers and affiliate hospitals. UHC's Clinical DataBase/Resource Manager allows comparison of patient-level risk-adjusted outcomes for performance improvement. The UHC provided the writing committee with a summary of recent information based on ICD-9 codes for thoracic aortic disease–related hospitalizations from the Clinical DataBase/Resource Manager (Tables 2A and 2B).⇓⇓ This data table demonstrates a high number of hospital discharges (more than 135 000) for thoracic, abdominal, thoracoabdominal, and “unspecified” aortic aneurysms in the 5-year period between 2002 and 2007. Subcategories include those with dissection, those with rupture, and those with neither. In the most recent 1-year period assessed (fourth quarter 2006 through third quarter 2007), there were nearly 9000 cases representing all patients with thoracic aortic disease discharged from UHC hospitals. Although these data are unavailable for the entire United States, they provide important estimates of the magnitude of the prevalence of thoracic aortic disease. Additional information regarding the patient admission source, particularly those with acute presentations, is pertinent to the discussion regarding interinstitutional transfer (see Section 17.2).
Most patients with significant thoracic aortic disease will be directed to specialized practitioners and institutions. However, the importance of early recognition and prompt treatment and/or referral for various thoracic aortic diseases by all healthcare professionals provides the rationale for this document. This guideline will attempt to provide the practitioner with a sufficient description of background information, diagnostic modalities, and treatment strategies so that appropriate care of these patients can be facilitated and better understood. The goal of this guideline is to improve the health outcomes and quality of life for all patients with thoracic aortic disease.
Other practice guidelines developed by ACCF and AHA address the management of patients with cardiac and vascular diseases. The ACCF/AHA guidelines on peripheral arterial disease (4) include recommendations for lower extremity, renal, mesenteric, and abdominal aortic diseases. Data standards on this topic are currently in development, as is a practice guideline on extracranial carotid and vertebral artery diseases. The ACCF/AHA guidelines are published in the Journal of the American College of Cardiology and Circulation and are available on both the ACC (www.acc.org) and AHA (my.americanheart.org) Web sites.
This guideline includes diseases involving any or all parts of the thoracic aorta with the exception of aortic valve diseases (5) and includes the abdominal aorta when contiguous thoracic aortic diseases are present. Specific disease states are described in the following sections, and the reader is referred to the glossary of terminology in Section 1.5 and Appendix 3 for abbreviations used throughout the guideline.
The reader will note that several topics or referenced areas may appear more than once throughout the guideline. This will appear to be redundant to those who choose to read the entire document, but the writing committee believed that because of the multidisciplinary nature of and organizational involvement in this project, individuals representing specific disciplines may choose to read and extract portions of the document for their own use. Inclusion of the narrative text and references was thought to be appropriate to facilitate a more complete understanding for these disciplines and individuals. Accordingly, the organization of the guideline is meant to be less of a textbook presentation of the various topics but rather a more clinically oriented document applicable to a variety of disciplines.
1.4.1 Critical Issues
As the writing committee developed this guideline, several critical issues emerged:
• Thoracic aortic diseases are usually asymptomatic and not easily detectable until an acute and often catastrophic complication occurs. Imaging of the thoracic aorta with computed tomographic imaging (CT), magnetic resonance imaging (MR), or in some cases, echocardiographic examination is the only method to detect thoracic aortic diseases and determine risk for future complications (see Section 4).
• Radiologic imaging technologies have improved in terms of accuracy of detection of thoracic aortic disease. However, as the use of these technologies has increased, so also has the potential risk associated with repeated radiation exposure, as well as contrast medium–related toxicity. Whether these technologies should be used repeatedly as a widespread screening tool is discussed in Section 4. In addition, the writing committee formulated recommendations on a standard reporting format for thoracic aortic findings as discussed in Section 4.5.
• Imaging for asymptomatic patients at high risk based on history or associated diseases is expensive and not always covered by payers.
• For many thoracic aortic diseases, results of treatment for stable, often asymptomatic, but high-risk conditions are far better than the results of treatment required for acute and often catastrophic disease presentations. Thus, the identification and treatment of patients at risk for acute and catastrophic disease presentations (e.g., thoracic AoD, thoracic aneurysm rupture) prior to such an occurrence are paramount to eliminating the high morbidity and mortality associated with acute presentations (see Section 8.1).
• A subset of patients with acute AoD are subject to missed or delayed detection of this catastrophic disease state. Many present with atypical symptoms and findings, making diagnosis even more difficult (see Sections 8.1.4 and 8.6). This issue has come under greater medical-legal scrutiny, and specific cases have been widely discussed in the public domain. Widespread awareness of the varied and complex nature of thoracic aortic disease presentations has been lacking, especially for acute AoD. Risk factors and clinical presentation clues are noted in Section 8.1.4. The collaboration and cosponsorship of multiple medical specialties in the writing of this guideline will provide unique opportunities for widespread dissemination of knowledge to raise the level of awareness among all medical specialties.
• There is rapidly accumulating evidence that genetic alterations or mutations predispose some individuals to aortic diseases (see Section 5). Therefore, identification of the genetic alterations leading to these aortic diseases has the potential for early identification of individuals at risk. In addition, biochemical abnormalities involved in the progression of aortic disease are being identified through studies of patients' aortic samples and animal models of the disease (6,7). The biochemical alterations identified in the aortic tissue have the potential to serve as biomarkers for aortic disease. Understanding the molecular pathogenesis may lead to targeted therapy to prevent aortic disease. Medical and gene-based treatments are beginning to show promise for reducing or delaying catastrophic complications of thoracic aortic diseases (see Section 9.2).
• As noted in Section 18, there are several areas where greater resources for research and both short- and long-term outcomes registries are needed.
1.5 Glossary of Terms and Abbreviations Used Throughout Guideline
Aneurysm (or true aneurysm): a permanent localized dilatation of an artery, having at least a 50% increase in diameter compared with the expected normal diameter of the artery in question. Although all 3 layers (intima, media, and adventitia) may be present, the intima and media in large aneurysms may be so attenuated that in some sections of the wall they are undetectable.
Pseudoaneurysm (or false aneurysm): contains blood resulting from disruption of the arterial wall with extravasation of blood contained by periarterial connective tissue and not by the arterial wall layers (see Section 8.4). Such an extravascular hematoma that freely communicates with the intravascular space is also known as a pulsating hematoma (8–10).
Ectasia: arterial dilatation less than 150% of normal arterial diameter.
Arteriomegaly: diffuse arterial dilatation involving several arterial segments with an increase in diameter greater than 50% by comparison to the expected normal arterial diameter.
Thoracoabdominal aneurysm (TAA): aneurysm involving the thoracic and abdominal aorta (see Section 126.96.36.199).
Abdominal aortic aneurysm (AAA): aneurysm involving the infradiaphragmatic abdominal aorta.
Aortic dissection (AoD): disruption of the media layer of the aorta with bleeding within and along the wall of the aorta. Dissection may, and often does, occur without an aneurysm being present. An aneurysm may, and often does, occur without dissection. The term “dissecting aortic aneurysm” is often used incorrectly and should be reserved only for those cases where a dissection occurs in an aneurysmal aorta (see Section 8.1).
2 The Thoracic Aorta
2.1 The Normal Aorta
The thoracic aorta is divided into 4 parts: the aortic root (which includes the aortic valve annulus, the aortic valve cusps, and the sinuses of Valsalva); the ascending aorta (which includes the tubular portion of the ascending aorta beginning at the sinotubular junction and extending to the brachiocephalic artery origin); the aortic arch (which begins at the origin of the brachiocephalic artery and is the origin of the head and neck arteries, coursing in front of the trachea and to the left of the esophagus and the trachea); and the descending aorta (which begins at the isthmus between the origin of the left subclavian artery and the ligamentum arteriosum and courses anterior to the vertebral column, and then through the diaphragm into the abdomen).
The normal human adult aortic wall is composed of 3 layers, listed from the blood flow surface outward (Figure 1):
Intima: endothelial layer on a basement membrane with minimal ground substance and connective tissue.
Media: bounded by an internal elastic lamina, a fenestrated sheet of elastic fibers; layers of elastic fibers arranged concentrically with interposed smooth muscle cells; bounded by an external elastic lamina, another fenestrated sheet of elastic fibers.
Adventitia: resilient layer of collagen containing the vasa vasorum and nerves. Some of the vasa vasorum can penetrate into the outer third of the media.
2.2 Normal Thoracic Aortic Diameter
Since then, it has been recognized that the “normal aortic diameter” is influenced by a number of factors, including patient age, sex, and body size; location of aortic measurement; method of measurement; and the robustness and type of imaging methods used. Hannuksela et al (13) noted that diameter increased by 0.12 to 0.29 mm/y at each level measured by CT for 41 men and 36 women aged 18 to 82 years (Figure 2). Aortic diameter for men was larger than that for women, but the difference decreased with age. Body mass index also affected aortic diameter by 0.27 mm (0.14 to 0.44 mm) per unit of body mass index (13).
Aortic diameter gradually tapers downstream from the sinuses of Valsalva. Hager et al (14) examined 46 men and 24 women without cardiovascular disease (age range 1 to 89 years; mean age 50.2 years) using helical CT (Figure 3). For these patients, there was no correlation with weight, height, or body surface area, but aortic diameter increased with age and was larger for men than for women (14).
Two-dimensional echocardiography has been used to define the “normal” range for aortic diameter at the sinuses of Valsalva in different age categories (and stratified by body surface area) (15). Adjusting for 2 of the key determinants of aortic diameter allows a more precise characterization of aortic size in otherwise healthy individuals (15) (Figure 4). Again, age and sex affected aortic root diameter, but the influence of sex was neutralized when diameter was indexed to body surface area (Table 4).
These tables and definitions help define the presence or absence of a thoracic aortic aneurysm and help define the threshold for considering further treatment for such patients. However, patients with certain genetic syndromes and abnormal tissue morphology may in fact have a normal aortic diameter at the time of acute AoD rupture (see Section 5.1.2). Another challenge relates to abnormal morphology of one aortic segment compared with another. For example, if the diameter of the ascending aorta exceeds the diameter of the aorta at the level of the sinuses Valsalva, even if both are within normal range, then the ascending aorta is considered to be enlarged. To adjust for body habitus variation, the use of aortic diameter indexed to height has been reported to better indicate surgical timing than might be recommended from aortic diameter alone for an otherwise asymptomatic patient with Marfan syndrome or bicuspid aortic valve (16). Whenever possible, the writing committee has inserted aortic diameter thresholds for further action, whether the action is for continued surveillance or for endovascular or surgical intervention.
3 Thoracic Aortic Histopathology
Atherosclerosis is characterized by intimal lesions called atheromata, or atheromatous or fibrofatty plaques, which protrude into the arterial lumen and weaken the underlying media often associated with calcification. With aging, presence of risk factors, and genetic predisposition, this progresses to complicated lesions with surface defects, hemorrhage, and/or thrombosis. A 1995 consensus document from the AHA defines the types and histologic classes of atherosclerosis (17) (Figure 5).
Thoracic aortic atherosclerosis is less common than abdominal aortic atherosclerosis, but the clinical importance is great. Clinical presentations and problems associated with aortic atherosclerosis and atheroma are discussed extensively in Section 11.
3.2 Aneurysms and Dissections
The pathology associated with thoracic aortic aneurysms and dissections was initially termed cystic medial necrosis but this term is a misnomer; the disease is not associated with necrosis of the aorta or with cyst formation. Aortic aneurysm histopathology, more accurately termed medial degeneration, is characterized by disruption and loss of elastic fibers and increased deposition of proteoglycans (Figure 1). Typically, there are areas of loss of smooth muscle cells in the aortic media, but whether there is a total loss of smooth muscle cells in the aortic wall is not clear. There can be atherosclerosis lesions present, but again, these changes are typically superimposed on medial degenerative disease. Although medial degeneration was initially described as a noninflammatory disease, recent literature supports the presence of inflammatory cell infiltration in this disease (18,19).
The biochemical pathways and proteins involved with medial degeneration have not been clearly delineated. However, multiple studies have found increased immunostaining for a subset of matrix metalloproteinases (MMPs) in the media of thoracic aortic aneurysms, particularly MMP-2 and MMP-9 (20–23). Immunostaining of aortic media from patients with Marfan syndrome has demonstrated increases in MMP-2 and MMP-9, which were associated with smooth muscle cells at the borders of areas of medial degeneration and on the surface of disrupted elastic fibers. Elevated MMP-2 and MMP-9 immunostaining has been demonstrated in ascending aneurysms from patients with either tricuspid or bicuspid aortic valves (21,23) and inconsistently in ascending aortic tissue from patients with tricuspid aortic valves (22). These 2 MMPs are known to have elastolytic activity. Variable expression of MMPs and tissue inhibitors of MMPs has also been demonstrated in aortic tissue of patients with Marfan syndrome versus patients without Marfan syndrome (24). Although accumulation of proteoglycans in the aortic media is another consistent finding in thoracic aortic aneurysms, no studies have determined why this accumulation occurs or whether these are causative in nature.
Medial degeneration is also associated with focal loss of vascular smooth muscle cells. Although there are regions of smooth muscle cell loss, morphometric analysis of aortic tissue has suggested that hyperplastic cellular remodeling of the media in ascending thoracic aortic aneurysms may be an initial adaptive response to minimize increased wall stress resulting from vascular dilatation (25). More recent studies of the aortic pathology associated with myosin heavy chain 11, smooth muscle (MYH11), and actin, alpha 2, smooth muscle aorta (ACTA2) mutations leading to ascending aortic aneurysms demonstrate a hyperplastic response by smooth muscle cells in the aortic media. The aortic media in aneurysm tissue taken from patients harboring mutations in these genes demonstrated focal hyperplasia associated with smooth muscle cells that were remarkable for a lack of structured orientation parallel to the lumen of the aorta, but instead, the smooth muscle cells were oriented randomly with respect to one another (26,27).
3.3 Vasculitis and Inflammatory Diseases
A variety of inflammatory vasculitides may also result in thoracic aortic disease. These include giant cell arteritis (GCA), Takayasu arteritis, and Behçet disease (see Section 7). The pathophysiology of GCA shares important features with Takayasu arteritis (28). T-cell clonal expansion suggests a specific antigenic response, which currently remains unelucidated. The inflammatory response, which begins in the adventitial layer, is marked by augmented cytokine and MMP production causing granuloma formation. Granuloma formation both shields the vessel from the inciting antigen and causes vessel destruction (29). Behçet disease is a vasculitis affecting both arteries and veins, of all sizes.
4 Imaging Modalities
4.1 Recommendations for Aortic Imaging Techniques to Determine the Presence and Progression of Thoracic Aortic Disease
1. Measurements of aortic diameter should be taken at reproducible anatomic landmarks, perpendicular to the axis of blood flow, and reported in a clear and consistent format (seeTable 5).(Level of Evidence: C)
2. For measurements taken by computed tomographic imaging or magnetic resonance imaging, the external diameter should be measured perpendicular to the axis of blood flow. For aortic root measurements, the widest diameter, typically at the mid-sinus level, should be used. (Level of Evidence: C)
3. For measurements taken by echocardiography, the internal diameter should be measured perpendicular to the axis of blood flow. For aortic root measurements, the widest diameter, typically at the mid-sinus level, should be used. (Level of Evidence: C)
4. Abnormalities of aortic morphology should be recognized and reported separately even when aortic diameters are within normal limits. (Level of Evidence: C)
5. The finding of aortic dissection, aneurysm, traumatic injury and/or aortic rupture should be immediately communicated to the referring physician. (Level of Evidence: C)
6. Techniques to minimize episodic and cumulative radiation exposure should be utilized whenever possible. (30,31) (Level of Evidence: B)
1. If clinical information is available, it can be useful to relate aortic diameter to the patient's age and body size. (Level of Evidence: C)
Definitive identification or exclusion of thoracic aortic disease or one of its anatomic variants requires dedicated aortic imaging. Selection of the most appropriate imaging study may depend on patient-related factors (i.e., hemodynamic stability, renal function, contrast allergy) and institutional capabilities (i.e., rapid availability of individual imaging modalities, state of the technology, and imaging specialist expertise). Consideration should be given to patients with borderline abnormal renal function (serum creatinine greater than 1.8 to 2.0 mg/dL)—specifically, the tradeoffs between the use of iodinated intravenous contrast for CT and the possibility of contrast-induced nephropathy, and gadolinium agents used with MR and the risk of nephrogenic systemic fibrosis (32).
Radiation exposure should be minimized (31,33–36). The risk of radiation-induced malignancy is the greatest in neonates, children, and young adults (31). Generally, above the age of 30 to 35 years, the probability of radiation-induced malignancy decreases substantially (30,31). For patients who require repeated imaging to follow an aortic abnormality, MR may be preferred to CT. MR may require sedation due to longer examination times and tendency for claustrophobia.
CT as opposed to echocardiography can best identify thoracic aortic disease, as well as other disease processes that can mimic aortic disease, including pulmonary embolism, pericardial disease, and hiatal hernia. After intervention or open surgery, CT is preferred to detect asymptomatic postprocedural leaks or pseudoaneurysms because of the presence of metallic closure devices and clips.
We recommend that external aortic diameter be reported for CT or MR derived size measurements. This is important because lumen size may not accurately reflect the external aortic diameter in the setting of intraluminal clot, aortic wall inflammation, or AoD. A recent refinement in the CT measurement of aortic size examines the vessel size using a centerline of flow, which reduces the error of tangential measurement and allows true short-axis measurement of aortic diameter. In contrast to tomographic methods, echocardiography-derived sizes are reported as internal diameter size. In the ascending aorta, where mural thrombus in aneurysms is unusual, the discrepancy between the internal and external aortic diameters is less than it is in the descending or abdominal aorta, where mural thrombus is common. Standardization of aortic diameter measurements is important to the planning of endovascular treatment of an aneurysm, where the diameter of the aorta in the seal zone must be matched to the diameter of an endograft. Here, the choice of internal or external diameter is specified by the manufacturer of the endograft.
4.2 Chest X-Ray
Routine chest x-ray may occasionally detect abnormalities of aortic contour or size that require definitive aortic imaging. Chest x-ray often serves as a part of the evaluation of patients with potential acute AoD, primarily to identify other causes of patient's symptoms, but also as a screening test to identify findings due to a dilated aorta or bleeding. However, chest x-ray is inadequately sensitive to definitively exclude the presence of AoD in all except the lowest-risk patients and therefore rarely excludes the disease. Pooled data from 10 studies places the predictive sensitivity of a widened mediastinum or an abnormal aortic contour associated with significant thoracic aortic disease at 64% and 71%, respectively (37). In the same analysis, however, including all abnormal radiographic findings increased the sensitivity to 90%, suggesting that a completely normal chest x-ray does lower the likelihood of AoD and may provide meaningful clinical information in very low-risk patients (37). The presence of a widened mediastinum or other radiographic findings suggestive of thoracic aortic disease increases the likelihood of AoD, particularly among patients who lack a clear alternative source for their symptoms. In a blinded prospective study of 216 patients who underwent evaluation for acute thoracic aortic disease, the specificity of chest x-ray for aortic pathology was 86% (38).
For patients with chest trauma, chest x-ray is a poor screening test for the diagnosis of aortic injury (39–41). A sharply demarcated normal mediastinal contour is sometimes used to exclude mediastinal hematoma (suggesting a wide mediastinum, abnormal mediastinal contour, left apical cap, loss of the aortic knob, depression of the left main bronchus, and deviation of an indwelling esophageal tube). However, these signs of hemomediastinum are more often false positive than true positive for aortic injury (40). Figure 6 illustrates the normal appearance of the thoracic aorta, and its components, on a posteroanterior chest x-ray.
4.3 Computed Tomographic Imaging
CT scanning has been used for more than 2 decades to identify acute AoD and to diagnose and measure other thoracic aortic diseases (42) (see Section 5.1). Advantages include near universal availability—the ability to image the entire aorta, including lumen, wall, and periaortic regions; to identify anatomic variants and branch vessel involvement; and to distinguish among types of acute aortic syndromes (i.e., intramural hematoma [IMH], penetrating atherosclerotic ulcer [PAU], and acute AoD)—and the short time required to complete the imaging process and the 3-dimensional data. Electrocardiogram (ECG)-gated techniques have made it possible to generate motion-free images of the aortic root and coronary arteries, similar to coronary CT angiographic imaging.
Reports of newer-generation multidetector helical CT scanners show sensitivities of up to 100% and specificities of 98% to 99% (43–46). Data from the International Registry of Acute Aortic Dissection (IRAD) show that for patients with aortic dissection, CT was the initially used diagnostic modality in 61% of patients and transthoracic echocardiography (TTE) and/or transesophageal echocardiography (TEE) was used first in 33% of patients. Whether this was because the first test was insufficient to diagnose dissection or because additional information about cardiac and aortic valve function was required is unclear (47).
The sequence for CT performed in the potential setting of acute AoD generally would include a noncontrast study to detect subtle changes of IMH (Figure 7), followed by a contrast study to delineate the presence and extent of the dissection flap, identify regions of potential malperfusion, and demonstrate contrast leak indicating rupture. Imaging of the vascular tree from the thoracic inlet to the pelvis, including the iliac and femoral arteries, provides sufficient information to plan surgical or endovascular treatment, if needed. Prompt interpretation and communication of findings to the appropriate treating physicians are essential in the acute setting.
For trauma patients, the sensitivity, specificity, and accuracy of contrast-enhanced multidetector CT for traumatic aortic injury are 96%, 99%, and 99%, respectively (48–50). The negative predictive value of contrast-enhanced CT approaches 100% in many studies, most with in-hospital follow-up of negative CT examinations (36,48,51,52) (Figure 8).
An outcome study of 278 patients undergoing contrast-enhanced CT for blunt chest trauma revealed 6 patients with aorta/great vessel injury and confirmed the 100% negative predictive value in the remaining patients using an extensive review of medical databases with a median follow-up of 20.5 months (53).
Regarding other thoracic aortic diseases, CT has been shown to have a 92% accuracy for diagnosing abnormalities of the thoracic aorta, in a series of examinations that included 33 thoracic aneurysms, 3 ruptured thoracic aortic aneurysms, 6 PAUs, 5 AoDs, and 2 pseudoaneurysms. In addition, CT correctly predicted the need for hypothermic circulatory arrest during surgical repair in 94% of patients (54). For congenital and inflammatory conditions of the thoracic aorta, the literature is primarily descriptive, and accuracy data are not available.
4.3.1 Computed Tomographic Imaging Technique
Although nonhelical CT scanners are capable of diagnosing thoracic aortic disease, the technique used is axial step-and-shoot technology. Inherent limitations include slow scan speed, relatively poor spatial resolution given thicker collimation used to extend the needed craniocaudal coverage, and inherently a noncontiguous dataset. Any patient motion, even minimal, between the acquisition of each image or cluster of images creates a stair-step artifact on multiplanar and 3-dimensional renderings. Helical CT scanners of 4 and greater detector rows consistently provide volumetric acquisitions of the thoracic aorta. Scanners with 16 and greater detector rows provide isotropic resolution in the x, y, and z axes, which allows the datasets to be reconstructed in the optimum imaging plane best suited to any individual vessel.
Technical parameters recommended for image reconstruction are slices of 3-mm or less thickness with a reconstruction interval of 50% or less than the slice thickness at 50% or greater overlap, tube rotation of 1 second or less, and 120 to 140 kVp (55). ECG gating is particularly useful for ascending aortic disease, eliminating motion artifact at the aortic root that can simulate an AoD (Figure 9), and allowing evaluation of aortic valve morphology and function, as well as evaluation of the proximal coronary arteries. If appropriately acquired, an aorta CT and a complete CT coronary angiogram can be obtained in 1 CT acquisition. For AoD, the scan coverage should start above the aortic arch and extend at least to the aortoiliac bifurcation, and probably the groin. This is important to determine both branch vessel involvement, such as lesion extension into the abdominal aorta and iliac arteries and characterization and diagnosis of malperfusion syndrome, and to evaluate access for percutaneous repair when transluminal therapy is being considered. (For further information on technique parameters and anatomic coverage, see the online-only Data Supplement.)
The CT angiographic acquisition uses intravenous contrast delivered at rate of 3 to 5 mL/s by a power injector and usually followed by a saline bolus. The total volume of contrast used should be kept as low as possible, to no greater than 150 mL.
Although axial sections remain the mainstay of interpretation, 2- and 3-dimensional reconstructions, such as maximum intensity projection, multiplanar and curved multiplanar reformations, and volume rendering, may augment interpretation and improve communication of the findings (54). To our knowledge, it has not been scientifically shown that the use of these tools increases diagnostic accuracy or diagnostic confidence among specialists. For example, in 1 study multiplanar reconstructions when added to axial images alone changed the interpretation in only 1 case (54). However, 3-dimensional reconstruction is likely to play an important role in the planning of surgical or endovascular approaches.
4.4 Magnetic Resonance Imaging
MR has been shown to be very accurate in the diagnosis of thoracic aortic disease, with sensitivities and specificities that are equivalent to or may exceed those for CT and TEE (44,58–62). Like CT, MR provides a multiplanar evaluation of the thoracic aorta, and the examination can be extended in coverage to include the length of the aorta and branch vessels, from skull base to the toes. Advantages of MR include the ability to identify anatomic variants of AoD (IMH and PAU), assess branch artery involvement, and diagnose aortic valve pathology and left ventricular dysfunction without exposing the patient to either radiation or iodinated contrast. Disadvantages include prolonged duration of imaging acquisition during which the patient is inaccessible to care providers; inability to use gadolinium contrast in patients with renal insufficiency; contraindication in patients with claustrophobia, metallic implants, or pacemakers; and lack of widespread availability on an emergency basis. Although time-resolved MR techniques are improving, MR often cannot clearly characterize the relationship of an intimal flap and aortic root structures, specifically the coronary arteries. Likely as a result of these considerations, data from IRAD found that MR was the least-used imaging study, used in only 1% of patients as the initial diagnostic study (63). For traumatic injury and congenital and inflammatory conditions of the thoracic aorta, the literature is primarily descriptive, and accuracy data are not available.
4.4.1 Magnetic Resonance Imaging Technique
A comprehensive MR examination of the thoracic aorta may include many components, including black blood imaging, may include basic spin-echo sequences, noncontrast white blood imaging, contrast-enhanced MR angiography using gadolinium based agents, and phase-contrast imaging. MR of the thoracic aorta in the acute setting can be shortened to meet urgent patient assessment needs. The use of each is described more specifically later. MR examinations have become faster, with advances in gradient hardware that have significantly reduced repetition times, resulting in ultrafast MR angiography techniques. However, MR examinations remain 2 to 4 times longer than CT examinations. MR terminology across imaging vendors is less consistent than that of CT; therefore, imaging sequences are described later as general types of sequences without vendor-specific details (58,59).
4.4.2 Black Blood Imaging
Black blood imaging, using spin-echo sequences, is used to evaluate aortic anatomy and morphology (such as aortic size and shape), and the aortic wall for hematoma or other causes of thickening such as vasculitis. They may be repeated after the administration of gadolinium-based contrast agents to identify wall enhancement. These sequences should use ECG gating at end-diastole and may be performed with or without double inversion recovery techniques that null the signal from blood. These sequences generate 2-dimensional images that can be obtained in the axial, sagittal, and coronal planes, as well as the oblique sagittal or “candy cane” view. T1-weighted gradient echo sequences may be used in place of black blood images and are usually performed both before and after contrast administration. They can be obtained in the axial, sagittal, and coronal planes.
4.4.3 Noncontrast White Blood Imaging
This is performed using either basic gradient echo sequences or the more advanced balanced steady-state free precession techniques (SSFP) that are T2*weighted and generate images with subsecond temporal resolution. Signal is generated from blood, making it appear white in the absence of contrast. SSFP techniques use ultrashort repetition times that require MR scanners with high-performance gradients. For the latter, an in-plane resolution of approximately 2.0×1.5 mm2 can be obtained using a repetition time of 3.2 ms, TE of 1.6 ms, flip angle of 60 to 70 degrees, and a 256×256 matrix. SSFP techniques can be performed to generate 2-dimensional, 3-dimensional, and cine images; the 2-dimensional images are usually performed in the axial, sagittal, and coronal planes in an interleaved manner and use ECG gating with triggering at end-diastole to generate images in less than 500 ms that do not require breath holding. Cine SSFP sequences require 7- to 9-second breath holding and are usually only performed at specific anatomic locations of interested, as determined by the findings on the previously generated images. For patients who cannot hold their breath, cine SSFP sequences can be performed without using ECG gating or breath holding but with substantially lower spatial and temporal resolution.
4.4.4 Contrast-Enhanced Magnetic Resonance Angiography
For the thoracic aorta, contrast-enhanced MR angiography is usually performed with ECG gating. Although this increases acquisition time, it provides motion-free images of the aortic root and ascending aorta. For patients without a contraindication to receiving a gadolinium-based contrast agent, contrast-enhanced MR angiography is often the sequence of choice from which most of the diagnostic information is obtained. Contrast-enhanced MR angiography images are obtained as a 3-dimensional volumetric dataset, which can be manipulated and viewed in much the same manner as CT scan datasets. Advances in MR, particularly gradient strength, have markedly reduced the acquisition times of contrast-enhanced MR angiography possible to subsecond acquisition times. This temporally resolved subsecond contrast-enhanced MR angiography is particularly useful in critically ill patients or patients who cannot hold their breath, where motion artifact can degrade other sequences that take longer to acquire, with the tradeoff being a reduction of in-plane resolution.
4.4.5 Phase Contrast Imaging
These sequences are usually performed to evaluate gradients across an area of stenosis, across an intimal or cardiac valve. Image contrast is produced by differences in velocity. Images are obtained using ECG gating or triggering. Two-dimensional images are generated that center on the area of concern. Peak flow and velocity measurements may be calculated, and time–flow and time–velocity curves are generated.
4.5 Standards for Reporting of the Thoracic Aorta on Computed Tomography and Magnetic Resonance Imaging
Viewing and measuring are best accomplished at a picture archiving and communications system workstation or an independent computer workstation, in which the aorta can be rotated into the best orientation to review each segment of the aorta and the aortic branches. This minimizes the chance of inadvertently confusing normal structures for vascular abnormality (Figure 10).
The writing committee believes that specific qualitative and quantitative elements are important to include in reports (Table 5).
Diameter measurements taken from axial images are inherently incorrect unless the artery being measured is perfectly aligned in cross section on the image (Figure 11). It is preferable to make diameter measurements perpendicular to the longitudinal or flow axis of the aorta to correct for the variable geometry of the aorta. Suggested standard anatomic locations are noted in Figure 12. The use of standardized measurements helps minimize errant reports of significant aneurysm growth due to technique or interreader variability in measuring technique.
Angiography provides accurate information about the site of dissection, branch artery involvement, and communication of the true and false lumens (60,61). Additionally, angiographic and catheter-based techniques allow for evaluation and treatment of coronary artery (62) and aortic branch (visceral and limb artery) disease, as well as assessment of aortic valve and left ventricular function (60,63).
Disadvantages of angiography compared with other less invasive modalities include 1) not being universally available because it requires the presence of an experienced physician to perform the study; 2) being an invasive procedure that is time consuming and requires exposure to iodinated contrast; 3) having poor ability to diagnose IMH given a lack of luminal disruption; 4) potentially producing false negative results when a thrombosed false lumen prevents adequate opacification to identify the dissection; and 5) reported sensitivities and specificities of angiography for the evaluation of acute AoD that are slightly lower than those for the other less invasive modalities (42,64,65). Thus, CT, MR, and TEE have replaced catheter-based angiography as the first-line diagnostic tests to establish the presence of the acute aortic syndrome (60,63,66,67). Multidetector CT has also largely replaced angiography for the anatomic studies required for treatment planning and monitoring of aortic disease (68).
The aorta and its major branches can be visualized with echocardiography using a variety of imaging fields as well as methods of ultrasound. The suprasternal view is best for visualizing the aortic arch, whereas the aortic root and ascending aorta are best seen in the left (and sometimes right) parasternal projection. When involvement of the innominate artery, left subclavian artery, or left common carotid artery is suspected, orthogonal and longitudinal scanning planes are helpful. Imaging of the descending thoracic aorta is less easily accomplished with echocardiography compared with other imaging modalities. Abdominal scan planes can be used to visualize the caudal descending aorta.
In general, TEE is superior to TTE for assessment of the thoracic aorta. Through the use of multiplane image acquisi-tion, 3-dimensional Doppler TEE is safe and can be performed at the bedside, with a low risk of complications (less than 1% overall, less than 0.03% for esophageal perforation), most of which are related to conscious sedation (69,70). Reconstruction of the aorta can be performed. Intraoperative use of TEE is noted in Sections 14.2.3 and 14.2.4.
4.7.1 Echocardiographic Criteria for Thoracic Aortic Aneurysms
The echocardiographic diagnosis of thoracic aortic aneurysms is determined on demonstration of aortic enlargement relative to the expected aortic diameter, based on age- and body size–adjusted nomograms (Figures 2, 3, 4, and 13).⇓ For other aortic segments, such nomograms have not been developed, so aortic dilatation is diagnosed when the aorta exceeds a generally agreed to standard diameter (Table 3) or when a given aortic segment is larger than contiguous aortic segment(s) of apparently normal size. Beyond establishing the presence of aortic enlargement, echocardiography may reveal associated cardiac pathology that suggests the underlying etiology of the aortic disease (e.g., bicuspid aortic valve).
4.7.2 Echocardiographic Criteria for Aortic Dissection
The echocardiographic diagnosis of an AoD requires the identification of a dissection flap separating true and false lumens (Figure 14). However, one of the major limitations of both TTE and TEE is the frequent appearance of artifacts that mimic a dissection flap (Figure 15). These usually arise from a mirror image or reverberation artifact that appears as a mobile linear echodensity overlying the aortic lumen. It is therefore essential that the echocardiographer make certain to distinguish true dissection flaps from such artifacts. The first step is to confirm the presence of the dissection flap from several angles and from several transducer locations. The second step is to confirm that the dissection flap has a motion independent of surrounding structures and that the apparent flap is contained within the aortic lumen (i.e., does not pass through the aortic wall in any view). The third step is to usecolor-flow Doppler to demonstrate differential flow on the 2 sides of the dissection flap. Often one can visualize 1 or more sites of intimal tears, with color-flow Doppler demonstrating sites of flow between the 2 lumens. The true lumen typically exhibits expansion during systole and collapse during diastole. Additional signs of the true lumen include little or nospontaneous echocardiographic contrast, systolic jets directed away from the lumen, and forward flow during systole. The false lumen often exhibits diastolic expansion, evidence of spontaneous contrast, complete or partial thrombosis, and reversed, delayed, or absent blood flow. Imaging criteria for distinguishing the true and false lumens have also been described for intravascular ultrasound (71) and CT (72). Features that characterize the false lumen of an acute dissection include the wedgelike angle (the “beak sign”) where the dissection flap meets the aortic wall, the presence of strandlike structures (“cobwebs”) in the lumen, and the lack of a laminar structure in the outer wall of the lumen.
Associated echocardiographic findings of clinical importance include identification of a pericardial effusion. Indeed, it is often possible to see a layer of echogenic material within the pericardial fluid that is indicative of hemopericardium. Echocardiography can provide important information about right and left ventricular function and myocardial ischemia based on assessment of left ventricular segmental wall motion. On TEE, one can identify the ostia of the 2 coronary arteries and detect involvement by the dissection flap. Echocardiography also identifies the presence of aortic regurgitation and permits grading of its severity. On TEE, the mechanisms of acute aortic regurgitation can be defined, and these data can be used to guide the surgeon's efforts to spare the valve at the time of aortic repair.
188.8.131.52 Diagnostic Accuracy of Echocardiography for Aortic Dissection
TTE has a sensitivity of 77% to 80% and a specificity of 93% to 96% for identification of proximal AoD. For distal AoD, the sensitivity of TTE is lower. TEE improves the diagnostic accuracy substantially, particularly when a patient's body shape, chest wall, or concomitant pulmonary disease restricts the transthoracic windows for aortic imaging. With TEE, sensitivity for proximal AoD is 88% to 98% with a specificity of 90% to 95% (46). A 2006 meta-analysis that evaluated the diagnostic utility of TEE in suspected thoracic AoD included 630 patients from 10 different studies. TEE was shown to both accurately identify and rule out acute AoD with sensitivities and specificities of 98% (95% CI 95% to 99%) and 95% (95% CI 92% to 97%), respectively (46).
Major advantages of TEE include its portability (allowing for bedside patient evaluation), rapid imaging time, and lack of intravenous contrast or ionizing radiation. Additionally, dissection-related cardiac complications can be evaluated including aortic regurgitation, proximal coronary artery involvement, and the presence of tamponade physiology. Disadvantages of TEE include a potential lack of availability (particularly at small centers and during off hours) and sedation requirements that may include endotracheal intubation (73). The accuracy of TEE can be quite operator dependent. For the very distal ascending aorta and the proximal aortic arch, TEE may be limited by a blind spot caused by interposition of the trachea and left main bronchus between the esophagus and aorta. Small, circumscribed AoDs or IMHs in this region are less well visualized by TEE. Another limitation of TEE is its inability to visualize the abdominal aorta; if there is concern of a malperfusion syndrome, an abdominal CT may also be required.
184.108.40.206 Diagnostic Accuracy of Echocardiography for Acute Intramural Hematoma
IMH of the aorta has a distinctly different appearance on echocardiography. In contrast to classic AoD, in IMH there is no mobile intimal flap within the aortic lumen, and in most cases the aortic lumen has a relatively normal round appearance. Instead, in IMH there is thickening of the aortic wall that is typically crescentic in shape and extends along a length of the aorta. In some cases it can be difficult to distinguish the wall thickening of IMH from diffuse aortic atherosclerosis or mural thrombus lining an aortic aneurysm. However, atheroma and mural thrombus protrude from the aortic intima into the lumen, thus giving both the aortic wall and the lumen an irregular shape, whereas in IMH the inner lining of the aortic lumen remains smooth. Moreover, in the presence of intimal calcification, intramural thrombus will present as thickening external to calcification, whereas mural thrombus will be internal to calcification.
220.127.116.11 Role of Echocardiography in Following Patients With Chronic Aortic Disease
Given that TEE is a semi-invasive procedure; it is usually not preferred for surveillance of patients with thoracic aortic diseases. Also, although TTE is noninvasive, its failure to visualize consistently and measure accurately the tubular portion of the ascending thoracic aorta is problematic. It is not typically used to follow aneurysms in that aortic segment. However, because TTE does accurately visualize the aortic root, its primary role as an imaging method for serial follow-up is in patients with aortic disease limited to the root, particularly those with Marfan syndrome. It is also used, often in conjunction with CT or MR, to observe patients with concomitant structural heart disease, such as bicuspid aortic valve, mitral valve prolapse, cardiomegaly, or cardiomyopathy.
5 Genetic Syndromes Associated With Thoracic Aortic Aneurysms and Dissection
5.1 Recommendations for Genetic Syndromes
1. An echocardiogram is recommended at the time of diagnosis of Marfan syndrome to determine the aortic root and ascending aortic diameters and 6 months thereafter to determine the rate of enlargement of the aorta. (Level of Evidence: C)
2. Annual imaging is recommended for patients with Marfan syndrome if stability of the aortic diameter is documented. If the maximal aortic diameter is 4.5 cm or greater, or if the aortic diameter shows significant growth from baseline, more frequent imaging should be considered. (Level of Evidence: C)
3. Patients with Loeys-Dietz syndrome or a confirmed genetic mutation known to predispose to aortic aneurysms and aortic dissections (TGFBR1, TGFBR2, FBN1, ACTA2, or MYH11) should undergo complete aortic imaging at initial diagnosis and 6 months thereafter to establish if enlargement is occurring. (74–77) (Level of Evidence: C)
4. Patients with Loeys-Dietz syndrome should have yearly magnetic resonance imaging from the cerebrovascular circulation to the pelvis. (23,78,79) (Level of Evidence: B)
5. Patients with Turner syndrome should undergo imaging of the heart and aorta for evidence of bicuspid aortic valve, coarctation of the aorta, or dilatation of the ascending thoracic aorta. (80) If initial imaging is normal and there are no risk factors for aortic dissection, repeat imaging should be performed every 5 to 10 years or if otherwise clinically indicated. If abnormalities exist, annual imaging or follow-up imaging should be done. (Level of Evidence: C)
1. It is reasonable to consider surgical repair of the aorta in all adult patients with Loeys-Dietz syndrome or a confirmed TGFBR1 or TGFBR2 mutation and an aortic diameter of 4.2 cm or greater by transesophageal echocardiogram (internal diameter) or 4.4 to 4.6 cm or greater by computed tomographic imaging and/or magnetic resonance imaging (external diameter). (78) (Level of Evidence: C)
2. For women with Marfan syndrome contemplating pregnancy, it is reasonable to prophylactically replace the aortic root and ascending aorta if the diameter exceeds 4.0 cm. (74) (Level of Evidence: C)
3. If the maximal cross-sectional area in square centimeters of the ascending aorta or root divided by the patient's height in meters exceeds a ratio of 10, surgical repair is reasonable because shorter patients have dissection at a smaller size and 15% of patients with Marfan syndrome have dissection at a size less than 5.0 cm. (16,76,81) (Level of Evidence: C)
1. In patients with Turner syndrome with additional risk factors, including bicuspid aortic valve, coarctation of the aorta, and/or hypertension, and in patients who attempt to become pregnant or who become pregnant, it may be reasonable to perform imaging of the heart and aorta to help determine the risk of aortic dissection. (Level of Evidence: C)
5.1.1 Marfan Syndrome
Marfan syndrome is a heritable disorder of the connective tissue with a high penetrance but variable expression. Approximately 25% of patients do not have a family history and represent new cases due to sporadic mutations for the condition. Marfan syndrome results from mutations in the FBN1 gene, with over 600 mutations currently entered into the FBN1 mutation database causing Marfan syndrome or related conditions (http://www.umd.be/). The FBN1 gene encodes fibrillin-1, a large glycoprotein that is secreted from cells and deposited in the extracellular matrix in structures called microfibrils. Microfibrils are found at the periphery of elastic fibers, including the elastic fibers in the medial layer of the ascending aorta, and in tissues not associated with elastic fibers (82). Only 12% of FBN1 mutations causing Marfan syndrome have been observed more than once in unrelated individuals, a fact that complicates using mutational detection for diagnosis. A second locus for Marfan syndrome, termed MFS2, was recently identified to be caused by mutations in the transforming growth factor-beta type II receptor (TGFBR2) (83). The phenotype of this locus may overlap for Loeys-Dietz syndrome. The criteria for Marfan syndrome is based primarily on clinical findings in the various organ systems affected in the Marfan syndrome, along with family history and FBN1 mutations status (84).
The cardinal features of Marfan syndrome involve the cardiovascular, ocular, and skeletal systems. Patients with Marfan syndrome are highly predisposed to thoracic aortic aneurysm and/or dissection, with virtually every patient with the syndrome having evidence of aortic disease at some point during their lifetime. Other cardiovascular manifestations include valvular disease, primarily mitral valve prolapse and regurgitation (85). Aortic regurgitation can result from distortion of the aortic valve cusps by an enlarged aortic root. The skeletal manifestations reflect overgrowth of the long bones and include arachnodactyly, dolichostenomelia, kyphoscoliosis, dolichocephaly, and pectus deformities. Abnormalities in the connective tissues are also manifested as joint laxity, recurrent or incisional hernias, striae atrophica, and dural ectasia (84). The ocular manifestation that is both sensitive and fairly specific for Marfan syndrome is ectopia lentis or lens dislocation. The presence of ectopia lentis is a particularly useful clinical finding to differentiate Marfan syndrome from Loeys-Dietz syndrome (78).
Most patients with Marfan syndrome present with dilatation of the aortic root/ascending aorta or Type A dissection. Internal aortic diameter measured at the sinuses of Valsalva provides a baseline for future evaluations because this is the aortic segment that dilates in Marfan syndrome. Of note, echocardiographic studies measure the internal diameter, whereas most patients undergo definitive imaging by CT and/or MR, which measures the external diameter (expected to be 0.2 to 0.4 cm larger than internal diameter), and external diameter is the measurement used in most cases to determine the threshold for surgical repair. There is growing awareness of the importance of relating this measurement to normal values based on age and body surface area (15) (Table 4). The severity of the aortic disease is related to the degree and segment length of aortic dilatation with dilatation limited to the sinuses of Valsalva having a less malignant prognosis than dilatation that extends to the aortic arch (86). After diagnosis, follow-up imaging studies are recommended at 6 months and then annually if stability is documented based on the high likelihood of aortic disease progression. In most cases, TTE can be used to monitor the size of the sinuses of Valsalva.
A subset of patients present with Type B dissection, and a rare patient will present with AAA. The poor outcome of patients with Marfan syndrome with acute Type B dissections has led some to advocate early surgical repair (87).
Studies addressing the efficacy of beta blockade in patients with Marfan syndrome have shown slower aortic root growth, fewer cardiovascular end points (defined as aortic regurgitation, dissection, surgery, heart failure, or death), and improved survival (88). Patients continue to enlarge their aorta and dissect on therapy, so such medication does not preclude the need for routine imaging and prophylactic aortic repair when the diameter of the aorta warrants repair. It is also important to note that significant aortic root dilatation is correlated negatively with therapeutic response (88). Recent studies in a mouse model of Marfan syndrome with aortic disease similar to that seen in humans showed that treatment with losartan normalized aortic root growth (6), and a clinical trial using losartan in Marfan syndrome patients under the age of 25 years is in progress (89,90) (see Section 18.104.22.168).
Surgical repair of the dilated aortic root/ascending aorta for patients with Marfan syndrome is usually performed at a threshold of a external diameter of 5.0 cm (91), smaller than that for other patients because of the greater tendency for AoD at a smaller diameter (see Section 22.214.171.124). Factors that will prompt repair at an external diameter of less than 5.0 cm are rapid growth defined as greater than 0.5 cm/y, family history of AoD at a diameter less than 5.0 cm, or the presence of significant aortic regurgitation.
After prophylactic repair of the ascending aorta, the arch and descending aorta are sites for later-onset aneurysms and dissections in patients with Marfan syndrome, prompting the need for routine imaging of the arch and descending aorta. Survival in patients with Marfan syndrome has been significantly improved with medical and surgical management of the aortic disease (76,92,93). The David valve sparing reimplantation operation for suitable patients undergoing elective aortic root surgery at centers with a high volume of these cases has become standard practice (76,92–99), although some have reported less-optimal long-term results with valve-sparing procedures (100,101).
Pregnant patients with Marfan syndrome are at increased risk for AoD if the aortic diameter exceeds 4 cm (102) (see Section 10). All women with Marfan syndrome warrant frequent cardiovascular monitoring throughout pregnancy and into the puerperium. Limited data on the treatment of women with Marfan syndrome who experience dissections during pregnancy suggest a better outcome with cesarean section with concomitant aortic repair (103).
5.1.2 Loeys-Dietz Syndrome
The Loeys-Dietz syndrome is an autosomal dominant aortic aneurysm syndrome with involvement of many other systems (78,104). Loeys-Dietz syndrome results from mutations in either the transforming growth factor receptor Type I or II (TGFBR1 or TGFBR2) genes and the diagnosis is confirmed through mutational analysis of these genes (105).
The disease is characterized by the triad of arterial tortuousity and aneurysms, hypertelorism and bifid uvula or cleft palate, or a uvula with a wide base or prominent ridge on it. The arterial tortuousity is most commonly observed in the head and neck vessels but can occur in other vessels. These patients also have a spectrum of other features, which include the following: velvety and translucent skin, craniosynostosis, malar hypoplasia, retrognathia, blue sclera, patent ductus arteriosus, skeletal features similar to Marfan syndrome, dural ectasia, atrial septal defects, developmental delay, cervical spine abnormalities, and joint laxity. The vascular disease in these patients is particularly aggressive with a mean age of death of 26 years (78). Most patients have aortic root aneurysms (98%) that lead to AoD. Because there are multiple reports of AoD occurring in patients with Loeys-Dietz when the aortic diameter was less than 5.0 cm, repair is recommended at smaller diameters (78) (see Section 5.1). For young children with severe systemic manifestations of Loeys-Dietz syndrome, specifically prominent craniofacial features that are associated with more severe aortic disease, once the aortic diameter exceeds the 99th percentile for age and the aortic valve annulus reaches 1.8 to 2.0 cm, prophylactic surgery allows for the placement of a graft of sufficient size to accommodate growth. Of note, echocardiographic examinations that measure the internal aortic diameter were used in these studies to determine the threshold for surgical repair.
Patients with Loeys-Dietz also develop aneurysms of other vessels (53%), leading to the recommendation that they have yearly MR imaging from the cerebrovascular circulation to the pelvis. Current studies indicate that aggressive surgical management of the aneurysms in these patients can be achieved with few complications (23,78). Surgical procedures in patients with Loeys-Dietz syndrome are not complicated by tissue fragility (23,79).
5.1.3 Ehlers-Danlos Syndrome, Vascular Form or Type IV
The vascular form of Ehlers-Danlos syndrome is a rare autosomal dominant disorder characterized by easy bruising, thin skin with visible veins, characteristic facial features, and rupture of arteries, uterus, or intestines. Rupture of the gastrointestinal tract is more likely to occur prior to arterial rupture, and the majority of patients survive the gastrointestinal rupture (106). Most of the fatal complications are caused by arterial rupture, with most deaths attributable to arterial dissections or ruptures involving primarily the thoracic or abdominal arteries, including AoDs and ruptures. These arterial ruptures lead to reduced life expectancy, with the median survival of only 48 years, and often no aneurysms are documented. If the rupture of an artery is life threatening, it can be surgically repaired, but tissue fragility, tendency to hemorrhage extensively, and poor wound healing may complicate the surgical repair (107). Whether there is a role for the repair of unruptured aneurysms in patients with this syndrome is not clear, which is in contrast to Loeys-Dietz syndrome, where a role for prophylactic surgical repair of aneurysms is already well established. Nevertheless, when these patients present with AoD or aortic root aneurysms, successful aortic surgery can be achieved with careful handling of tissues and resewing of anastomoses with pledgeted sutures (76). Noninvasive vascular imaging is preferred as fatal complications have been associated with invasive imaging in these patients (106). The outcome of pregnancy in women with Ehlers-Danlos syndrome is poor because of rupture of the gravid uterus and vessel rupture at delivery or in the postpartum period. The diagnosis of vascular Ehlers-Danlos syndrome is based on DNA or protein studies identifying a defect in type III collagen, encoded by the COL3A1 gene.
5.1.4 Turner Syndrome
Turner syndrome is defined as complete or partial absence of 1 sex chromosome in a phenotypic female, most commonly 45, X. Short stature and ovarian failure are the most prevalent finding, but women with Turner syndrome have an increased cardiovascular mortality rate from both structural and ischemic heart disease, especially AoD (108,109). Between 10% and 25% of patients with Turner syndrome have a bicuspid aortic valve. Aortic coarctation is present in approximately 8% of patients. Determining aortic dilatation in patients with Turner syndrome is difficult because aortic dilatation is based on body surface area so the aortas of patients with Turner syndrome are expected to be smaller than those of the general population because of the patient's short stature. If one defines aortic dilatation as an ascending-to-descending aortic diameter ratio of greater than 1.5, then 33% of women with Turner syndrome had aortic dilatation (110).
The average age of AoD in Turner syndrome was 31 years, and less than half of the patients survived the event (111). Data indicate a population-based AoD incidence of 36:100 000 Turner syndrome years (1.4% among individuals with Turner syndrome) compared with 6:100 000 in the general Danish population (112). Therefore, the risk of AoD is much lower in patients with Turner syndrome compared with patients with Marfan syndrome or Loeys-Dietz syndrome. The majority of dissections in women with Turner syndrome occur in patients with known risk factors for dissection, such as cardiovascular malformations (bicuspid aortic valve or coarctation of the aorta), systemic hypertension, or both. Therefore, the evidence base regarding the value of screening for aortic disease in women with Turner syndrome is not available. However, there appears to be an increased risk for dissection in these women, suggesting that imaging of the heart, aorta, and pulmonary veins at the time of diagnosis might be valuable (80,113). For patients with no risk factors for AoD (bicuspid aortic valve, coarctation, dilated aorta), re-evaluation of the aorta has been suggested every 5 to 10 years or if clinically indicated (e.g., attempting pregnancy (114) or transition to an adult clinic). Patients with risk factors for AoD should undergo more frequent imaging. Recently, studies have not shown an effect of growth hormone treatment in women with Turner syndrome on either ascending or descending aortic diameter (113). In addition, studies have not found any evidence of left ventricular hypertrophy in patients with Turner syndrome who were treated with growth hormone (115,116).
5.1.5 Other Genetic Syndromes With Increased Risk for Thoracic Aortic Aneurysms and Dissections
A substantial proportion of patients with Ehlers-Danlos syndrome who do not have the vascular form also have aortic root dilatation, but the progression of this dilatation to AoD is rare (76,117). Similarly, patients with congenital contractural arachnodactyly or Beals syndrome due to mutations in FBN2 have had aortic root enlargement without documented progression to dissection (118,119).
There are other genetic syndromes that have multiple reports or documentation of thoracic aortic aneurysms leading to Type A dissections. There are multiple case reports of AoD in patients with autosomal dominant polycystic kidney disease (120,121). Although AoD is a complication of autosomal dominant polycystic kidney disease, it is less common than cerebral aneurysms leading to subarachnoid hemorrhage in this population. There is insufficient information to gauge the value of routine or screening imaging for these patients.
Similar to autosomal dominant polycystic kidney disease, there are multiple reports in the literature of patients with Noonan syndrome who are experiencing AoDs (122–124). The value of imaging or routine monitoring of these patients is unknown. A review of 200 patients with Alagille syndrome also identified thoracic aortic disease in a small subset of these patients (125).
5.1.6 Recommendations for Familial Thoracic Aortic Aneurysms and Dissections
1. Aortic imaging is recommended for first-degree relatives of patients with thoracic aortic aneurysm and/or dissection to identify those with asymptomatic disease. (126,127) (Level of Evidence: B)
2. If the mutant gene (FBN1, TGFBR1, TGFBR2, COL3A1, ACTA2, MYH11) associated with aortic aneurysm and/or dissection is identified in a patient, first-degree relatives should undergo counseling and testing. Then, only the relatives with the genetic mutation should undergo aortic imaging. (Level of Evidence: C)
1. If one or more first-degree relatives of a patient with known thoracic aortic aneurysm and/or dissection are found to have thoracic aortic dilatation, aneurysm, or dissection, then imaging of second-degree relatives is reasonable. (126) (Level of Evidence: B)
2. Sequencing of the ACTA2 gene is reasonable in patients with a family history of thoracic aortic aneurysms and/or dissections to determine if ACTA2 mutations are responsible for the inherited predisposition. (26,27,77,78,128,129) (Level of Evidence: B)
1. Sequencing of other genes known to cause familial thoracic aortic aneurysms and/or dissection (TGFBR1, TGFBR2, MYH11) may be considered in patients with a family history and clinical features associated with mutations in these genes. (26,27,77,78,128,129) (Level of Evidence: B)
2. If one or more first-degree relatives of a patient with known thoracic aortic aneurysm and/or dissection are found to have thoracic aortic dilatation, aneurysm, or dissection, then referral to a geneticist may be considered. (Level of Evidence: C)
A genetic basis of nonsyndromic familial thoracic aortic aneurysms and dissection has only recently been defined. Familial aggregation studies of patients referred for repair of thoracic aortic aneurysm and dissection that did not have a genetic defect have indicated that between 11% and 19% of these patients have a first-degree relative with thoracic aortic aneurysms and dissection (127,130). Patients with a family history of thoracic aortic aneurysm and dissection present at a younger mean age than do sporadic patients but at a significantly older age than patients with Marfan or Loeys-Dietz syndrome. The disease is primarily inherited in an autosomal dominant manner with decreased penetrance primarily in women and variable expression (131). These mapping studies have firmly established that there is significant genetic heterogeneity for familial thoracic aortic aneurysm and dissection (i.e., many different genes can be mutated and cause the same clinical condition) (27,77,129,132–134). The causative gene has been identified for the following loci: TAAD2 is due to TGFBR2 mutations, the mutant gene at 16p is the MYH11 gene, and the TAAD4 defective gene is ACTA2.
The defective gene at the TAAD2 locus for familial thoracic aortic aneurysms and dissection was identified as TGFBR2, which is the same gene that is mutated in approximately two thirds of patients with Loeys-Dietz syndrome. Genetic testing of unrelated families with familial thoracic aortic aneurysm and dissections demonstrated that TGFBR2 mutations were only present in 1% to 5% of families (26). All 4 families had mutations that affected TGFBR2 arginine 460 of the receptor, suggesting that missense mutation was associated with familial thoracic aortic aneurysm and dissection (i.e., genotype-phenotype correlation). Although the majority of vascular disease in these families involved the ascending aorta, affected family members also had descending aortic disease and aneurysms of other arteries, including cerebral, carotid, and popliteal aneurysms. It is notable that similar to Loeys-Dietz syndrome, AoDs occur in patients with TGFBR2 mutation at diameters less than 5.0 cm, leading to the recommendation that aortic repair be considered at a internal diameter by echocardiography of 4.2 cm or greater (78) (see Section 5.1).
A large French family with thoracic aortic aneurysm and dissection associated with patent ductus arteriosus was used to map the defective gene causing this phenotype to 16p (135). The defective gene at this locus was identified as the smooth muscle cell–specific myosin heavy chain 11 (MYH11), a major protein in the contractile unit in smooth muscle cells (77). Subsequent analysis of DNA from 93 unrelated families with thoracic aortic aneurysm and dissection failed to identify any MYH11 mutations. Sequencing DNA from 3 unrelated families with thoracic aortic aneurysm and dissection associated with patent ductus arteriosus identified MYH11 mutations in 2 of these families (26); the remaining family had a TGFBR2 mutation as the cause of the thoracic aortic aneurysm and dissection and patent ductus arteriosus. Therefore, MYH11 mutations are responsible for familial thoracic aortic aneurysm and dissection associated with patent ductus arteriosus and a rare cause of familial thoracic aortic aneurysm and dissection.
A defective gene at the locus 10q23–24 was identified in a large family with multiple members with thoracic aortic aneurysm and dissection as ACTA2, which encodes the the smooth muscle-specific alpha-actin, a component of the contractile complex and the most abundant protein in vascular smooth muscle cells (27,136). Approximately 15% of families with thoracic aortic aneurysm and dissection have ACTA2 mutations (27). Features identified in some families with ACTA2 mutation included livedo reticularis and iris flocculi, although the prevalence of these features has not been determined. The majority of affected individuals presented with acute Type A or B dissections or Type B dissections, with 16 of 24 deaths occurring due to Type A dissections. Two of 13 individuals experienced Type A dissections with a documented ascending aortic diameter less than 5.0 cm. AoDs occurred in 3 individuals under 20 years of age, and 2 women died of dissections postpartum. Finally, 3 young men had Type B dissection complicated by rupture or aneurysm formation at the ages of 13, 16, and 21 years.
Identification of the underlying genetic mutation leading to familial thoracic aortic aneurysms and dissections provides critical clinical information for the family. First, only family members who harbor mutations need to be routinely imaged for aortic disease. Second, identification of the underlying mutation may lead to different management of the aortic disease, as is the case for TGFBR2 mutations. In addition to providing information to families, identification of genes leading to familial thoracic aortic aneurysms and dissections has emphasized the roles of smooth muscle contractile function in preventing aortic diseases (11). Individuals who undergo genetic testing for thoracic aortic disease should receive genetic counseling prior to the testing to explain the implications for the testing for their medical follow-up and implications for family members.
The genes leading to nonsyndromic forms of aortic aneurysms and dissections are in the early stages of identification. Tables 6 and 7⇓⇓ summarize the current clinical features associated with mutations in these genes and recommendations for when to sequence these genes in families with multiple members with familial thoracic aortic aneurysm and dissection.
Given the familial risk of thoracic aortic aneurysms, screening the proband's first-degree relatives with appropriate imaging studies is indicated in the absence of identification of the defective gene leading to the disease.
Because thoracic aortic disease is typically asymptomatic until a life-threatening event (e.g., AoD) occurs, evaluating other family members can potentially prevent premature deaths. Most syndromes and familial forms of thoracic aortic disease are inherited in an autosomal dominant manner. Therefore, an individual with an inherited predisposition to thoracic aortic aneurysm and dissections has up to a 50% risk of passing on this predisposition to their children, which is the basis for genetic evaluation of the offspring. In addition, siblings and parents of the patient need to be evaluated for possible predisposition to thoracic aortic aneurysm and dissections. Because of the variable age of onset of aortic disease in familial thoracic aortic aneurysms and dissections, the writing committee believes that imaging of family members at risk of the disease every 2 years is warranted.
6 Other Cardiovascular Conditions Associated With Thoracic Aortic Aneurysm and Dissection
6.1 Recommendations for Bicuspid Aortic Valve and Associated Congenital Variants in Adults
1. First-degree relatives of patients with a bicuspid aortic valve, premature onset of thoracic aortic disease with minimal risk factors, and/or a familial form of thoracic aortic aneurysm and dissection should be evaluated for the presence of a bicuspid aortic valve and asymptomatic thoracic aortic disease. (Level of Evidence: C)
2. All patients with a bicuspid aortic valve should have both the aortic root and ascending thoracic aorta evaluated for evidence of aortic dilatation. (137–140) (Level of Evidence: B)
Bicuspid aortic valves is the most common congenital abnormality affecting the aortic valve and the aorta and is found in 1% to 2% of the population (137). Nine percent of patients have family members who also have bicuspid aortic valves (141). The ACC/AHA Valvular Heart Disease Guidelines specifically address this condition (5). Of importance to this guideline, bicuspid aortic valves can be inherited in families as an autosomal dominant condition and may be associated with thoracic aortic aneurysm formation. It is important to note that in these families, members can have thoracic aortic aneurysms in the absence of bicuspid aortic valves (142). The valves are prone to either aortic valve regurgitation, most commonly seen in younger patients, or aortic valve stenosis, more common in older patients. Bicuspid aortic valve repair for regurgitation has excellent long-term results, an important consideration in the absence of prosthetic aortic valve alternatives in this young population (99,139,140,143). The most common site of fusion of the leaflets is at the left and right leaflet commissure, and less so at the right noncoronary leaflet commissure. The latter is typically more often associated with aortic valve stenosis. In a study of 2000 patients at the Cleveland Clinic who underwent bicuspid aortic valve surgery, 20% had concurrent ascending aortic aneurysms that required repair (139,140). AoD is also common, and as many as 15% of patients with acute AoD have bicuspid aortic valves (144), a frequency more common than Marfan syndrome. AoD was present in 12.5% of patients with bicuspid valves with an aortic diameter less than 5 cm, reminiscent of patients with Marfan syndrome in whom 15% had AoD at a size less than 5 cm (16,143).
The writing committee discussed the potential need to screen relatives of patients who have undergone aortic valve replacement before age 70, as these younger patients were more likely to have had bicuspid aortic valve as their primary pathology. However, 40% of women and around one third of men over age 70 undergoing aortic valve replacement have bicuspid aortic valve disease (145). The yield from screening relatives of younger patients with bicuspid aortic valve disease compared with relatives of older patients is not known.
6.2 Aberrant Right Subclavian Artery
Aberrant right subclavian artery, which arises as the fourth branch from the aorta, courses behind the esophagus in approximately 80% of patients and causes dysphagia in many patients (146–149). The dysphagia usually occurs in adults as the artery enlarges (Kommerell diverticulum) (150,151). In most adult patients, the aorta is also abnormal and is prone to aneurysm formation, dissection, and rupture. Surgical treatment in adults involves resection of the aneurysmal segment of the subclavian artery (the diverticulum) and the adjacent aorta and replacement of the aorta with a graft (152). An alternative treatment is exclusion of the right subclavian artery origin and adjacent aorta using an aortic endograft, although long-term follow-up of this endovascular approach is not available and the compression and aneurysm growth may continue (153). The distal normal segment of the subclavian artery is usually not reimplanted into the descending aorta directly but rather revascularized using an interposition graft or carotid-subclavian bypass. Alternatively, the subclavian artery can be ligated or coiled distal to the aneurysm, implanting the distal segment into the adjacent carotid artery in the neck and thus restoring flow to the upper extremity and the ipsilateral vertebral artery (148,149,152).
6.3 Coarctation of the Aorta
Coarctation of the aorta is a relatively common abnormality that occurs in about 40 to 50 of every 100 000 live births, with a 2:1 ratio in males versus females. Most lesions are treated soon after birth or in childhood, and adults presenting with untreated coarctation of the aorta are rare. More often, patients have had previous procedures and present with later problems such as heart failure, intracranial hemorrhage, hypertension particularly with exercise, aneurysm formation, AoD, rupture of old repairs, undersized grafts of previous repairs, and infections. It is of particular importance in previously treated patients with some aortic narrowing to check for a gradient across the stenosis and for hypertension during exercise testing (154).
Untreated coarctation of the aorta has a dismal prognosis, with 80% of patients dying from complications associated with the coarctation. Approximately one quarter will die from AoD or rupture, one quarter will die from heart failure, one quarter will die from intracranial hemorrhage, and the remainder will die from other complications.
Surgical options, depending on the lesion, include subclavian artery patch aortoplasty, patch aortoplasty, bypass of the coarctation, tube graft replacement, aneurysm replacement, 2-stage combined bicuspid valve surgery, and arch and descending aorta replacement or ascending aorta–to–descending aorta bypass (3). Endovascular balloon dilatation and stent placement has been used successfully and is becoming a less invasive alternative to conventional open surgical procedures (155). Occasionally, the adult aorta may be redundant and kinked opposite the ligamentum arteriosum without any pressure gradient, the so-called pseudocoarctation. Aneurysms that require surgical treatment may develop proximal and distal to the kinked area (156).
6.4 Right Aortic Arch
A right-sided aortic arch is present in approximately 0.5% of the population and rarely requires surgical repair. However, some patients present with dysphagia or asthma-like symptoms with expiratory wheezing. CT or MR readily diagnoses the problem of either tracheal compression or esophageal compression with the esophagus enlarged and filled with gas above the level of the arch. Felson and Palayew described 2 types (157).
In Type I, the great vessels come off the right-sided arch in a manner that is a mirror image to normal anatomy; however, compression of the esophagus or trachea is caused by an enlarged aorta where it crosses the vertebral bodies or by the vascular ring formed by the atretic ductus arteriosus. In Type II, the aberrant left subclavian artery comes off the descending aorta and typically runs posterior to the trachea and compresses the trachea. As with an aberrant right subclavian artery, the proximal segment of the subclavian artery may enlarge and form a Kommerell diverticulum. Other variations are also seen but are exceedingly rare. A separate trunk may arise from the arch, and from this trunk, the innominate, carotid, and subclavian arteries arise. Another rare finding is a right-sided arch with a stump of left-sided aorta joining the right-sided arch after it crosses into the left chest. The stump of the left-sided aorta gives off a branch to the left subclavian artery. The aorta with right-sided arches is also abnormal and typically very fragile and prone to AoD, rupture, or aneurysm formation. Surgical repair involves resection of the aorta and, if needed, reimplantation or bypass of the aberrant left subclavian artery.
7 Inflammatory Diseases Associated With Thoracic Aortic Disease
7.1 Recommendations for Takayasu Arteritis and Giant Cell Arteritis
See Table 8.
1. Initial therapy for active Takayasu arteritis and active giant cell arteritis should be corticosteroids at a high dose (prednisone 40 to 60 mg daily at initiation or its equivalent) to reduce the active inflammatory state. (158,159) (Level of Evidence: B)
2. The success of treatment of patients with Takayasu arteritis and giant cell arteritis should be periodically evaluated to determine disease activity by repeated physical examination and either an erythrocyte sedimentation rate or C-reactive protein level. (160,161) (Level of Evidence: B)
3. Elective revascularization of patients with Takayasu arteritis and giant cell arteritis should be delayed until the acute inflammatory state is treated and quiescent. (162) (Level of Evidence: B)
4. The initial evaluation of Takayasu arteritis or giant cell arteritis should include thoracic aorta and branch vessel computed tomographic imaging or magnetic resonance imaging to investigate the possibility of aneurysm or occlusive disease in these vessels. (Level of Evidence: C)
1. It is reasonable to treat patients with Takayasu arteritis receiving corticosteroids with an additional anti-inflammatory agent if there is evidence of progression of vascular disease, recurrence of constitutional symptoms, or re-elevation of inflammatory marker. (158) (Level of Evidence: C)
7.2 Takayasu Arteritis
Takayasu arteritis, also known as pulseless disease, is an idiopathic vasculitis of the elastic arteries, involving the aorta and its branches. Initially described in Japan, the disease is found worldwide. In the United States, a review of cases in Olmsted County, Minn, reported a rate of 2.6 cases per million persons (158). In the United States, the disease affects all ethnic and racial groups in proportion to the census with a moderate Asian overrepresentation. The disease affects women approximately 10 times more often than men. Most commonly diagnosed in the third decade (i.e., the 20s) of life, the disease has been found in children and adults in the fifth decade (i.e., the 40s). Two specific disease distributions have been reported: Japanese and Indian (167,168).
In the Japanese distribution, the thoracic aorta and great vessels are most commonly affected. In contrast, in the Indian type, the disease most commonly affects the abdominal aorta and the renal arteries (28). The pathogenesis of Takayasu arteritis remains poorly defined. A T-cell–mediated panarteritis, the disease proceeds from adventitial vasa vasorum involvement inward. The antigen for the localized inflammatory process is undefined but likely specific as the T cells undergo clonal expansion. The outcome process of destruction and fibrotic repair depends on the dominant pathophysiologic process: destruction yields aneurysms while fibrosis causes stenosis.
The diagnosis of Takayasu arteritis may be made using the 1990 American College of Rheumatology criteria: 1) age of onset younger than 40 years, 2) intermittent claudication, 3) diminished brachial artery pulse, 4) subclavian artery or aortic bruit, 5) systolic blood pressure variation of greater than 10 mm Hg between arms, and 6) angiographic (CT, MR) evidence of aorta or aortic branch vessel stenosis (163) (Figure 16). When 3 of the criteria are manifest, the sensitivity and specificity for diagnosis are 90.5% and 97.8%, respectively. Laboratory testing may aid in diagnosis. Markers of inflammation, such as C-reactive protein and erythrocyte sedimentation rate, are elevated in approximately 70% of patients in the acute phase and 50% in the chronic phase of disease (158).
The clinical manifestations of the disease typically develop in 2 phases: acute and chronic. Acutely, the inflammation associated with Takayasu arteritis causes a host of constitutional, or “B,” symptoms, such as weight loss, fatigue, night sweats, anorexia, and malaise (158). More chronically, once the vascular process has endured, patients report symptoms referable to the organs involved. In the largest US experience, more than half of all patients experienced upper extremity claudication, half had symptoms associated with cerebrovascular insufficiency (vision loss, lightheadedness, stroke), and a third reported carotid artery pain (158). In an Indian series, hypertension as a result of renal artery involvement was the most common presenting sign (169).
The aorta itself may develop either aneurysm or stenosis. In a Japanese series of 116 patients with Takayasu arteritis, nearly 32% of the patients had aortic aneurysm formation (170). Most commonly, aneurysm formation developed in the descending aorta, followed by the abdominal, then ascending aortic segments. In the National Institutes of Health series of 60 patients with Takayasu arteritis, 23% had aortic aneurysm formation (158). Aneurysms most commonly formed in the aortic arch or root, abdomen, and then other thoracic segments. Stenosis of the aorta is more common than is aneurysm formation, occurring in 53% of patients in the National Institutes of Health series. Any segment of the aorta may be involved, but the abdominal aortic segment is affected more than 70% of the time if stenosis is found.
Treatment of Takayasu arteritis begins with inflammation reduction with corticosteroids. Steroids are typically started at high dose, 40 to 60 mg daily at initiation to lower the erythrocyte sedimentation rate or C-reactive protein to normal, and are required for 1 to 2 years to ensure proper disease treatment (169). Despite the prolonged regimen, nearly half of the patients will relapse during tapering, requiring additional immunosuppression. Second-line agents that have been used include methotrexate, azathioprine, and antitumor necrosis factor-alpha agents (158,169). Unfortunately, markers of inflammation are imperfect barometers of disease activity. Disease progression has been shown to occur in the setting of normal marker levels (158).
Revascularization for aortic stenosis or aneurysm occurs for the same indications as in noninflammatory disorders: secondary organ vascular insufficiency or risk of rupture. There are no randomized trials of percutaneous or surgical intervention in this disease (158,171–173). Nonrandomized reports have shown that revascularization of either variety may be appropriate, with one caveat. The risk of graft failure is higher in patients with active local inflammation (162). Moreover, the presence of aneurysmal disease itself may be problematic. One report documented a 12% incidence of anastomotic aneurysms over 2 decades of follow-up related to the presence of aneurysms at surgery (174).
7.3 Giant Cell Arteritis
GCA, also known as temporal arteritis, is an elastic vessel vasculitis involving the aorta and its secondary and tertiary branches. Distinguishing it from Takayasu arteritis, GCA affects patients above the age of 50 years, with an incidence peaking in the eighth decade of life (175). The disease affects women in a 3:2 ratio to men and has a predilection for those of northern European ancestry (29). In the United States, epidemiologic investigation reports a prevalence of 20 cases per 100 000 persons. The incidence is higher in Scandinavian nations but lower in southern Europe, suggesting a genetic predisposition in certain populations (176).
The clinical presentation of GCA is varied, requiring a heightened suspicion by clinicians for early diagnosis. Half of the patients report constitutional symptoms, such as weight loss, night sweats, malaise, and fever (177). Because of the predilection for secondary and tertiary thoracic branches of the aorta, cranial symptoms are common. Scalp tenderness and headache are present in two thirds of patients and in up to 90% of patients with biopsy-proved disease (177). Jaw claudication is common and affects half of the patients, 20% develop visual changes, and other neurologic symptoms such as stroke or neuropathy occur in nearly one third (29). Visual changes are particularly important to notice, because early treatment may prevent permanent blindness. Patients may report diplopia, amaurosis fugax, or blurriness prior to blindness. Polymyalgia rheumatica characterized by a generalized inflammatory state with proximal muscle involvement is found in nearly half of patients with GCA (29). Patients with polymyalgia rheumatica report muscular pain and stiffness, particularly on initiation of movement.
Extracranial vascular involvement is less common in GCA than in Takayasu arteritis, occurring in 25% of patients. In a 50-year study of Olmsted County, Minn, that included 168 patients with GCA, aortic aneurysm/dissection was found in 18% of the subjects, whereas large-artery stenosis was noted in 13% of patients (178). No patient had stenosis of the aorta. Aortic aneurysm formation represents an important marker. Although aneurysm formation per se does not reduce survival compared with the GCA cohort as a whole, AoD in the setting of an aneurysm reduces survival to an average of 1.1 years (178). Similarly, aortic aneurysm rupture or dissection caused two thirds of deaths in a series of patients with GCA in California (179).
The American College of Rheumatology diagnostic criteria for GCA include 1) age older than 50 years, 2) recent-onset localized headache, 3) temporal artery pulse attenuation or tenderness, 4) erythrocyte sedimentation rate greater than 50 mm/h, and 5) an arterial biopsy demonstrating necrotizing vasculitis (164). Three or more criteria confer a sensitivity and specificity above 90% for the disease. With intracranial disease, temporal artery biopsies are diagnostic in up to 80% of cases (180). The rate of positivity declines with initiation of glucocorticoid therapy, but this should not delay treatment to avoid GCA complications. Biopsies performed within 7 days of steroid initiation retain a high diagnostic yield (181).
The pathophysiology of GCA shares important features with Takayasu arteritis (28). GCA is marked by a T-cell clonal expansion suggesting a specific antigenic response, which currently remains unelucidated. The inflammatory response, which begins in the adventitial layer, is marked by augmented cytokine and MMP production causing granuloma formation. Granuloma formation both shields the vessel from the inciting antigen and causes vessel destruction. The inflammatory environment within the vessel wall with the possible formation of aneurysms or vessel stenosis is histologically identical to that of Takayasu arteritis. Because of the multiyear cyclical nature of disease incidence, some have posited an infectious etiology (29).
Corticosteroids represent the standard in therapy for patients with GCA (182). The typical treatment regimen includes starting prednisone dose of 40 to 60 mg daily, although recent evidence suggests a similar efficacy with 30 to 40 mg daily (183). Therapy is typically required for 1 to 2 years to avoid recurrence, although the dose may be tapered beginning 2 to 3 months after initiation. Patients commonly report feeling much better rapidly but, as with Takayasu arteritis, new vascular involvement may occur in up to half of patients treated with steroids (184). In contrast to Takayasu arteritis, additional immunomodulatory agents do not seem to modulate the disease's progress. Methotrexate studied in a double-blind, placebo-controlled study as an adjunct to prednisone did not reduce morbidity, erythrocyte sedimentation rate level, or cumulative prednisone dose (184a). Revascularization recommendations follow the same pattern as in Takayasu arteritis.
7.4 Behçet Disease
In 1937, Hulusi Behçet described his eponymous syndrome based on a set of 3 symptoms: uveitis, aphthous stomatitis, and genital ulcers. Most common in Turkey, with a prevalence of 80 to 370 cases per 100 000 persons (185,186), the disease is much less common in the United States, with an estimated prevalence of 1 to 3 cases per million persons (187). The diagnostic criteria were established by the International Group for Behçet's disease and require oral ulceration and 2 of these 3 lesions: recurrent genital ulceration, uveitis or retinal vasculitis, or skin lesions, such as erythema nodosum, pseudofolliculitis, or pathergy (188). In addition to these cardinal manifestations, vascular involvement may occur in one third of patients. A small vessel vasculitis commonly associated with the human leukocyte antigen (HLA) B51 allele (187), Behçet disease is 1 of 2 vasculitides that may also involve veins. Venous involvement is most commonly superficial thrombophlebitis, but deep vein thrombosis in the vena cava, varices, and cerebral sinuses has been reported (189). The small vessel involvement of Behçet disease may result in nonvascular complaints, such as erythema nodosum, arthritis, and gastrointestinal involvement with diarrhea, gastrointestinal bleeding, or perforation (187). Treatment of Behçet disease varies based on the manifestation of disease. Systemic corticosteroids are the typical therapy for those with vascular involvement.
Specifically with regard to vascular manifestations of Behçet disease, any artery or vein, large or small, systemic or pulmonary, may be involved by the vasculitic process. Aortic histopathology shows lymphocytic infiltration mixed with histiocytes and eosinophils with giant cells around vasa vasorum of media and adventitia. Destruction of media leads to aneurysm formation and may proceed to pseudoaneurysm formation and rupture. Aneurysm formation may occur in multiple sites and in different sites over a period of follow-up. Aneurysm, stenotic lesions, and occlusion of brachiocephalic arteries may occur with or without aortic involvement. Although aortic involvement is unusual for patients with Behçet vasculitis, aneurysm rupture can be unpredictable and fatal (190,191). With regard to surgical repair, anastomotic pseudoaneurysms often occur (12.9% within 18 months in 1 series) and may be related to ongoing inflammatory changes in the area of anastomotic suture lines (192). Endovascular repair with stent grafts has also been described (193).
7.5 Ankylosing Spondylitis (Spondyloarthropathies)
The group of diseases labeled spondyloarthropathies are linked by the strong association of major histocompatibility complex HLA B-27 and the absence of rheumatoid factor (194,195). Several features are common to the spondyloarthropathies, including sacroilitis, inflammatory arthritis or enthesitis (inflammation of tendon insertions), associations with inflammatory bowel disease or psoriasis, and aortitis and heart block (195).
Ankylosing spondylits is the most common variant and often begins with back pain and stiffness in the second or third decade of life. It affects men 2 to 3 times as often as women, worsens with inactivity, and commonly takes years for the diagnosis to be made (196). The diagnosis requires 4 of the 5 criteria: onset of pain at younger than 40 years, back pain for longer than 3 months, morning stiffness, subtle symptom onset, and improvement with exercise (197). Patients may also report constitutional symptoms, such as malaise or fever. Acute anterior uveitis is reported in up to 40% of patients. Aortic root and aortic valve involvement are reported in up to 80% of patients (198). When involved, the aortic valve may have a nodular appearance, and aortic valvular regurgitation is present in nearly half of the patients (198). Treatment of aortic root expansion and aortic valvular abnormalities is the same as for other conditions.
7.6 Infective Thoracic Aortic Aneurysms
Infection (due to bacterial, fungal, viral, spirochetal, or tubercular organisms) is a rare cause of thoracic aortic aneurysms. Originally named mycotic endarteritis by Osler (199), the terms infected aneurysm or infectious aortitis are now used more commonly, because the majority of etiologic agents are nonfungal. Saccular aneurysms are most common, but infected aneurysms can be fusiform and often even pseudoaneurysms. The ascending thoracic aorta, aortic arch, and descending thoracic aorta can all be affected, as can prosthetic aortic grafts and aortic homografts. Typically, the sites of infected aneurysms are opposite the great vessels in the aortic arch or opposite the visceral ateries in the abdomen. There are several mechanisms by which the aortic infection may arise. First, there may be contiguous spread from adjacent thoracic structures, such a mediastinitis, abscess, infected lymph nodes, infectious pericarditis, empyema, or paravertebral abscess. Second, there may be septic emboli from underlying bacterial endocarditis. Third, there may be hematogenous dissemination of bacteria in the setting of sepsis or intravenous drug abuse. Infection most often arises in a diseased aorta, either in a preexisting aneurysm, at the site of an atherosclerotic plaque, or at the site of some accidental or iatrogenic aortic trauma. Indeed, infected thoracic aortic aneurysms may arise as a late complication of cardiac surgery, often associated with postoperative mediastinitis, typically at the sites of aortic cannulation or anastomotic suture lines (200,201).
Various organisms can infect the aorta, with most infections being bacterial. Staphylococcus aureus and Salmonella are the organisms most commonly identified (202–204). Pneumococcus and Escherichia coli are relatively common gram-positive and gram-negative pathogens, respectively.
Treponema pallidum, the gram-negative spirochete bacterium that causes syphilis, as well as other Treponema species, can cause infected aortitis, with the ascending thoracic aorta most often involved. However, in syphilitic aortitis, thoracic aortic aneurysm does not appear for 10 to 25 years after the initial spirochetal infection. Fungal infections of the aorta, with either Candida or Aspergillus, occur less often (205) and typically occur in the setting of impaired immunity, such as patients with systemic illness, human immunodeficiency virus, or prior organ or bone marrow transplant.
Indeed, patients with impaired immunity are also at increased risk of tuberculous aortitis attributable to mycobacterium tuberculosis. Tuberculous aortitis has until now been exceedingly rare, but the incidence may rise as the prevalence of tuberculosis rises worldwide. Tuberculous aortitis typically affects the distal aortic arch and descending thoracic aorta, likely because the aorta is thought to become infected via direct extension from continuous infected lymph nodes, empyema, or pericarditis (206).
Finally, there appears to be an independent association between human immunodeficiency virus and ascending thoracic aortic dilatation, although its mechanisms are poorly understood (207,208). Moreover, the incidence of frank aneurysms remains extremely low.
8 Acute Aortic Syndromes
Acute aortic syndromes consist of 3 interrelated conditions with similar clinical characteristics and include AoD, IMH, and PAU (209).
8.1 Aortic Dissection
8.1.1 Aortic Dissection Definition
AoD is defined as disruption of the media layer of the aorta with bleeding within and along the wall of the aorta resulting in separation of the layers of the aorta. In the majority of patients (90%), an intimal disruption is present that results in tracking of the blood in a dissection plane within the media. This may rupture through the adventitia or back through the intima into the aortic lumen (Figure 17). This classic dissection results in a septum, or “flap,” between the 2 lumens (Figure 18). The false lumen may thrombose over time (Figure 19). While on noninvasive imaging, 15% of patients with aortic dissection syndromes have an apparent IMH without evidence of an intimal tear, autopsy studies show only 4% have no visible intimal tear; indeed, at the time of surgery a tear is found in most patients (210,211). Occasionally, AoD originates from a small atheromatous ulcer that is difficult to identify. On the other hand, extensive atheromatous disease of the aorta may lead to PAU or a localized IMH.
The true incidence of acute AoD is difficult to define for 2 principal reasons: 1) acute AoD can be rapidly fatal, and when patients die prior to hospitalization, death may be erroneously attributed to another cause and 2) acute AoD is frequently missed on initial presentation, and early mortality among this group may be misclassified as non–dissection related. Population-based studies suggest that the incidence of acute AoD ranges from 2 to 3.5 cases per 100 000 person-years, which correlates with 6000 to 10 000 cases annually in the United States (75,212,215–217). A review of 464 patients from IRAD reported a mean age at presentation of 63 years, with significant male predominance (65%) (47). The prevalence of AoD appears to be increasing, independent of the aging population, as noted by Olsson and colleagues (218), who found the incidence of AoD among Swedish men has increased to 16 per 100 000 men yearly. It may be that 2 to 3 times as many patients die from AoD than from ruptured AAA; approximately 75% of patients with AAA will reach an emergency department alive, whereas for AoD, the prognosis appears to be worse, with 40% dying immediately, 1% per hour dying thereafter, and between 5% and 20% dying during or shortly after surgery (219–221). Furthermore, only 50% to 70% will be alive 5 years after surgery depending on age and underlying etiology (222). Because AoD tends to occur in areas of aneurysmal dilatation, treatment of aneurysms before dissection occurs is important to long-term survival (3) (see Section 8.1).
Regarding time from onset of initial symptoms to time of presentation, acute dissection is defined as occurring within 2 weeks of onset of pain; subacute, between 2 and 6 weeks from onset of pain; and chronic, more than 6 weeks from onset of pain.
8.1.2 Anatomic Classification of Aortic Dissection
Anatomically, acute thoracic AoD can be classified according to either the origin of the intimal tear or whether the dissection involves the ascending aorta (regardless of the site of origin). Accurate classification is important as it drives decisions regarding surgical versus nonsurgical management. The 2 most commonly used classification schemes are the DeBakey and the Stanford systems (Figure 20). For purposes of classification, the ascending aorta refers to the aorta proximal to the brachiocephalic artery, and the descending aorta refers to the aorta distal to the left subclavian artery.
The DeBakey classification system categorizes dissections based on the origin of the intimal tear and the extent of the dissection:
• Type I: Dissection originates in the ascending aorta and propagates distally to include at least the aortic arch and typically the descending aorta (surgery usually recommended).
• Type II: Dissection originates in and is confined to the ascending aorta (surgery usually recommended).
• Type III: Dissection originates in the descending aorta and propagates most often distally (nonsurgical treatment usually recommended).
– Type IIIa: Limited to the descending thoracic aorta.
– Type IIIb: Extending below the diaphragm.
The Stanford classification system divides dissections into 2 categories, those that involve the ascending aorta and those that do not.
• Type A: All dissections involving the ascending aorta regardless of the site of origin (surgery usually recommended) (Figures 18 and 20).
• Type B: All dissections that do not involve the ascending aorta (nonsurgical treatment usually recommended). Note involvement of the aortic arch without involvement of the ascending aorta in the Stanford classification is labeled as Type B (Figure 21).
At this time, there is no unanimity regarding which classification system is the ideal one to use. Some of the writing committee members believe that a more pragmatic approach is to refer to the dissection involving the aorta as either proximal or distal to the left subclavian artery. Others of the writing committee do not use this approach. Thus, if a patient has an arch dissection even without ascending aortic involvement, then immediate surgery would be recommended by some, if feasible and the patient is viable. Others on the writing committee would select medical management if the patient has only an arch dissection without proximal extension, malperfusion, or bleeding, as long as repeat imaging demonstrates stability. If there is evidence of malperfusion or bleeding in such a patient, then the writing committee would usually select a surgical approach.
8.1.3 Risk Factors for Aortic Dissection
Risk factors for AoD include conditions that result in aortic medial degeneration or place extreme stress on the aortic wall (Table 9). Two thirds to three quarters of patients have hypertension, which is often uncontrolled. Genetic predisposition (see Section 5) to AoD can occur in the context of a syndrome, such as Marfan syndrome or Loeys-Dietz syndrome, or can be inherited in families in the absence of syndromic features (3). IRAD data showed that of patients under 40 years of age with AoD, 50% had a history of Marfan syndrome (223). Other congenital or genetically based diseases as well as inflammatory conditions associated with a higher risk of AoD are noted in Sections 6.3, 6.4, and 7.
First and foremost, a family history of thoracic aortic aneurysm is an important risk factor. In 2 separate clinical studies, 13% to 19% of patients without an identified genetic syndrome with thoracic aortic aneurysms had first-degree relatives with thoracic aortic aneurysms or AoD (127,130). The term “familial thoracic aortic aneurysm and dissection syndrome” is often applied (see Section 5). In taking a history for thoracic aortic disease, one should be careful to distinguish a history of an abdominal aortic aneurysm from a thoracic aortic aneurysm. Many people, even healthcare providers, mistakenly use the terms AAA or triple A for any aortic aneurysm, regardless of location. Clarifying that the aneurysm was thoracic rather than abdominal affects one's consideration of risk. Also, one must consider the potential underlying diagnosis when a patient reports a family history of “sudden death” or “heart attack” when there was no confirmatory autopsy. If the patient's father, at the age of 45, had sudden onset of chest pain and then died moments later, there is a chance that the death may have been from an acute AoD rather than an acute MI.
The history may reveal syndromic causes of thoracic aortic aneurysm and dissections, especially Marfan, Loeys-Dietz, and vascular Ehlers-Danlos syndromes. In some cases, patients have only some of the features of Marfan or Loeys-Dietz syndrome, rather than the full-blown clinical syndrome, so a history of any phenotypic features, such as mitral valve prolapse or pectus excavatum, should prompt consideration of thoracic aortic aneurysms or dissections (224,225). Bicuspid aortic valve is a strong risk factor for ascending thoracic aortic aneurysms, as well as coarctation of the aorta. In addition, a history of extreme exertion or emotional distress may preceed the onset of pain (226).
8.1.4 Clinical Presentation of Acute Thoracic Aortic Dissection
The clinical presentation of acute AoD spans a spectrum from the overt with classic pain and physical examination findings to the enigmatic as a painless process with few physical manifestations of the disease (Table 10). Given its exceedingly high mortality, clinicians must maintain a high index of suspicion for acute AoD, as noted in Section 8.6 (Figure 22).
126.96.36.199 Symptoms of Acute ThoracicAortic Dissection
Patients with acute aortic syndromes often present in a similar fashion, regardless of whether the underlying condition is AoD, IMH, PAU, or contained aortic rupture. Pain is the most commonly reported presenting symptom of acute AoD regardless of patient age, sex, or other associated clinical complaint (228–235). Pooled data from over 1000 patients in 8 studies found that the pain of acute dissection is perceived as abrupt in onset in 84% of cases (95% CI 80% to 89%) and of severe intensity in 90% of cases (95% CI 88% to 92%) (236). Although classically described as having a tearing or ripping quality, registry data suggest patients are more likely to describe the pain of acute dissection as sharp or stabbing (51% to 64%, respectively) and that report of a migrating quality to pain is highly variable (12% to 55%) (228,236). Pain may subsequently ease or abate, leading to a false reassurance on the part of the patients and physicians.
Pain location and other associated symptoms reflect the site of initial intimal disruption and may change as the dissection extends along the aorta or involves other arteries or organ systems (236). Data from 464 patients enrolled in IRAD found that patients with Type A dissections most frequently present with chest pain (80%), more commonly described as anterior (71%) than as posterior (32%) (228). Although less common, patients with Type A dissection report back pain (47%) and abdominal pain (21%), presumably as a result of antegrade dissection into the descending aorta (228). In contrast, patients with Type B dissections are most likely to present with back pain (64%) followed by chest and abdominal pain (63% and 43%, respectively) (228). Some patients present with abdominal pain in the absence of chest pain or with only painful or numb lower extremities related to end-organ ischemia. In 1 retrospective study of 44 patients ultimately diagnosed with acute thoracic AoD, the location of the patient's pain was highly predictive of the clinician's suspicion for acute AoD; dissection was suspected in 86% of patients who presented with chest and back pain, 45% of those who presented with chest pain alone, and only 8% of those primarily abdominal pain (237).
Although uncommon, acute AoD may present without pain (238–240). In a separate analysis of 977 IRAD patients, 63 patients (6.4%) presented without pain (241,242). This group of patients was noted to be older and more likely to present with syncope, stroke, or congestive heart failure than were patients with painful dissection (241). Patients on steroids and patients with Marfan syndrome may be more prone to present without pain (243).
188.8.131.52 Perfusion Deficits and End-Organ Ischemia
Perfusion deficits as a result of dissection-related obstruction of aortic branch vessels have long been recognized as a common clinical manifestation, resulting in organ complications at initial presentation (Table 11). End-organ involvement in acute thoracic AoD can occur via several mechanisms. Most occlusions are caused by obstruction by the dissection flap, which can either prolapse across a vessel origin without entering it (dynamic obstruction) or directly extend into a vessel (static obstruction) (244) (Figure 18). Other causes include postobstructive arterial thrombosis, embolism to branches of either the true or false lumen, direct compression of an aortic branch artery or adjacent structures by an expanding false lumen (245), rupture or leakage of the false lumen into contiguous structures, and occlusion or dissection of coronary arteries and/or aortic valve distortion leading to heart failure.
Physical examination is insensitive to renal and mesenteric ischemia early in the course of acute AoD. Elevated serum creatinine or refractory hypertension may be due to renal ischemia but may represent the clinical baseline in a patient with poorly documented or inadequately treated prior medical conditions. Serologic markers of mesenteric ischemia may not be present until hours after onset.
Combined data from over 1500 patients in 16 studies found that pulse deficits were present in 31% of cases (95% CI 24% to 39%) and, when present, were strongly suggestive of AoD (positive likelihood ratio 5.7; 95% CI 1.4 to 23) (37) and predict increased risk. Of 513 cases of Type A dissection, patients with perfusion deficits were more likely to present with hypotension, shock, neurologic deficits, and tamponade and were more likely to have higher rates of hospital complications and mortality (41% versus 25%, P=0.0002) (246). Furthermore, overall mortality rates correlated with the number of pulse deficits present, likely as a reflection of the extent of vascular compromise and associated end-organ ischemia (246). Similarly, of 118 patients with Type A acute dissection, limb ischemia (defined as loss of pulse with associated pain and neurologic symptoms) was present in 38 cases (32%) (247). The presence of limb ischemia was associated with an increased likelihood of other end-organ ischemia (i.e., cerebral, visceral, or coronary) and a significant increase in overall mortality (247). Among the 38 patients with limb ischemia, in-hospital mortality was 45% compared with 15% among the 61 patients without organ malperfusion (247).
These studies underscore the clinical importance of an adequate vascular examination to help both identify the disease and stratify risk once the diagnosis is established. Every patient being evaluated for possible acute AoD should have pulses checked in all extremities to identify the presence of perfusion deficits. In patients with acute limb ischemia versus those without, renal and mesenteric malperfusion were nearly 2-fold more frequent and mortality was twice as high, further highlighting the importance of this finding (248).
8.1.5 Cardiac Complications
The heart is the most frequently involved end organ in acute AoD involving the ascending aorta. In distinction to other end-organ pathology, most cardiac complications are a direct result of dissection-related disruption of normal anatomic relationships (215,245).
184.108.40.206 Acute Aortic Regurgitation
Acute aortic regurgitation is the most commonly recognized cardiac complication of Type A dissection (228–234), occurring in 41% to 76% of cases (228–232). Three distinct dissection-related mechanisms for acute aortic valve incompetence have been identified, and they can occur in combination: 1) acute dilatation of the aortic root by an expanding false lumen, resulting in incomplete aortic valve closure; 2) a dissection extending into the aortic root and disrupting aortic valve commissural attachments, resulting in valve leaflet prolapse; and 3) a portion of dissection flap prolaping through the aortic valve in diastole, preventing adequate leaflet closure (235). Clinical manifestations of dissection-related aortic regurgitation span the spectrum from only a hemodynamically insignificant diastolic murmur to congestive heart failure and cardiogenic shock (236,249).
220.127.116.11 Myocardial Ischemia or Infarction
Myocardial ischemia or infarction is an infrequent but serious complication of acute AoD. Registry and review data suggest that ECG evidence of myocardial ischemia was present in up to 19% of patients with acute AoD, whereas pooled data from 988 patients in 9 different studies found that acute MI was present in 7% of cases (95% CI 4% to 14%) (37,47,250). Coronary artery flow can be compromised by an expanding false lumen compressing the proximal coronary or by extension of the dissection flap into the coronary artery ostium (251).
Clinically, a dissection-related cardiac malperfusion syndrome may present with ECG changes that are indistinguishable from those of primary myocardial ischemia or infarction, increasing the likelihood of misdiagnosis and inappropriate therapeutic intervention (252).
18.104.22.168 Heart Failure and Shock
Heart failure is a relatively uncommon complication of AoD, found to occur in approximately 6% of cases (236). In this setting, heart failure may result from acute aortic insufficiency, acute myocardial ischemia or infarction, or cardiac tamponade. Registry data suggest that patients with acute AoD complicated by heart failure are often atypical in their presentation, frequently leading to a delay in diagnosis (236). The largest study to evaluate heart failure in acute AoD included 1069 patients from the IRAD database and found that patients with AoD and concomitant heart failure were more likely to present in shock but were less likely to complain of chest pain and that, when chest pain was present, the pain was more often mild and less often abrupt in onset (236).
22.214.171.124 Pericardial Effusion and Tamponade
Pericardial pathology is a frequent complication of acute Type A AoD and can occur via 2 distinct mechanisms (37,253–256). Most commonly, transudation of fluid across the thin wall of an adjacent false lumen into the pericardial space leads to a hemodynamically insignificant pericardial effusion (256), which is present in about one third of patients (257). Less often, the dissected aorta ruptures directly into the pericardium, leading rapidly to tamponade physiology and hemodynamic compromise (245,258,259). Cardiac tamponade is diagnosed in 8% to 10% of patients presenting with acute Type A AoD and is an ominous clinical predictor of poor outcomes (260), as well as the leading cause of mortality in this group (47,215,231). Consequently, the presence of cardiac tamponade should prompt truly urgent aortic repair (260).
Syncope is a well-recognized dissection-related complaint occurring in approximately 13% of cases (242,261) with multiple potential etiologies, including: 1) cardiac (e.g., severe aortic regurgitation, ventricular outflow obstruction, cardiac tamponade), 2) vascular (e.g., impaired cerebral blood flow and aortic baroreceptor activation); 3) neurologic (e.g., vasovagal in response to pain), and 4) volume-related (e.g., false lumen rupture into the pleural space) causes (240,261–267). Regardless of its etiology, syncope in the setting of AoD increases the risk of near-term adverse events. In a review of 728 cases of acute AoD, patients with a history of syncope were significantly more likely to die than were those without syncope (34% versus 23%, respectively; P=0.01) (242). Additionally, patients who presented with syncope more frequently had associated cardiac tamponade, stroke, decreased consciousness, and evidence of spinal cord ischemia (242).
8.1.7 Neurologic Complications
Acute AoD frequently presents with dissection-related neurologic complications. Pooled data from more than 1300 patients in 13 studies that included both Type A and B dissections found that neurologic symptoms were reported in 17% (37). Neurologic complications may result from hypotension, malperfusion, distal thromboembolism, or nerve compression (251,253,254). In a recent study of 102 patients with Type A AoD, 29% had neurologic symptoms on initial presentation (253); of those with neurologic symptoms, 53% had ischemic stroke (predominantly right hemispheric) and 37% had ischemic neuropathy (described as limb pain with sensory or motor deficit) (253).
Last, although uncommon, acute paraplegia as a result of spinal cord malperfusion has been described as a primary manifestation of thoracic AoD, occurring in 1% to 3% of patients (215,251,253,254,268,269).
Of clinical note, up to 50% of dissection-related neurologic symptoms may be transient and as many as one-third of patients with neurologic symptoms present without complaints of chest pain, complicating appropriate diagnosis and treatment (253,255,256,262,269–271).
8.1.8 Pulmonary Complications
Pleural effusion, the most common pulmonary complication of acute AoD, is noted in 16% of cases at presentation (37); whereas large effusions may result from leaking of blood from the aorta into the pleural space, small effusions are typically a nonhemorrhagic exudate believed to be inflammatory in origin (37,272–275).
Other pulmonary complications of acute AoD include dissection-related compression of the pulmonary artery and development of an aortopulmonary fistula, either of which may present with dyspnea as a prominent symptom (276–278). Hemoptysis, noted in 1 study in 3% of patients presenting with thoracic AoD, may result from compression of lung parenchyma by an expanding false lumen or via direct aneurysmal rupture into the lung, leading to massive hemoptysis and death (215,279,280).
8.1.9 Gastrointestinal Complications
Mesenteric ischemia is the most common gastrointestinal complication of acute AoD (228) and can result from malperfusion or systemic hypotension. It is the most common cause of death among those with Type B AoD. Mesenteric ischemia is associated with abdominal pain, but pain may be nonspecific and out of proportion to the physical examination of the abdomen, so the cause of pain often goes unrecognized early on. Unfortunately, by the time serum markers of bowel ischemia or infarction turn positive, it is often too late to salvage the bowel or the patient. Therefore, it is essential to be viligant for mesenteric ischemia in every patient with acute AoD and associated abdominal pain.
Gastrointestinal hemorrhage is a rare but potentially catastrophic complication of acute AoD (281–283). Dissection-related gastrointestinal bleeding may present with limited bleeding as a result of mesenteric infarction or as massive hemorrhage secondary to an aortoesophageal fistula or false lumen rupture into proximal small bowel (281–283). Although rare, dissection-related gastrointestinal hemorrhage should be in the differential of all patients presenting with bleeding and complaints of thoracic or abdominal pain.
8.1.10 Blood Pressure and Heart Rate Considerations
Blood pressure abnormalities are common in patients presenting with acute thoracic AoD. About half of patients are hypertensive at presentation, with 71% of Type B patients having a systolic blood pressure greater than 150 mm Hg versus only 36% of Type A patients (284–287). Conversely, nearly 20% present with either hypotension or shock (37). Hypotension and shock can result from cardiac tamponade, aortic hemorrhage, severe aortic insufficiency, myocardial ischemia or infarction, true lumen compression by distended false lumen, or an intra-abdominal catastrophe. Of more than 1000 patients with acute AoD, those with hypotension on admission were found more likely to have neurologic complications; myocardial, mesenteric, or limb ischemia; and death (249).
Accurate systemic blood pressure measurement may be complicated by dissection-related occlusion of aortic branch arteries, resulting in erroneously low blood pressure readings in the affected limb. Accordingly, blood pressures may need to be measured in both arms and, at times, both legs to determine the highest central blood pressure.
8.1.11 Age and Sex Considerations
Acute AoD presentation varies with patient age and sex. In 951 IRAD patients, 7% were younger than 40 years. Compared with patients 40 years of age and older, this group was less likely to have a history of hypertension and significantly more likely to have Marfan syndrome, bicuspid aortic valve, or a history of prior aortic surgery (223). Clinically, young patients in this study were more likely to describe pain as abrupt in onset but less likely to be hypertensive at presentation (25% versus 45%) (223). In contrast, a separate study of IRAD data evaluating 550 patients with Type A dissection found that among patients more than 70 years of age (32% of total), typical symptoms (abrupt onset of pain) and signs (murmur of aortic regurgitation or pulse deficits) were significantly less common, suggesting that extra vigilance may be required to identify acute AoD in young and elderly patients (288).
Sex appears to affect the presentation of acute AoD as well. In a study of 1078 patients enrolled in IRAD, 32% were women. Women were older; were less likely to present within 6 hours of symptom onset, to complain of abrupt onset of pain, and to have a pulse deficit; and were more likely to present with either altered mental status or congestive heart failure. Consequently, women were less likely to be diagnosed within 24 hours of symptom onset and had significantly higher in-hospital mortality (30% versus 21%, P=0.001) than men (289).
8.2 Intramural Hematoma
Among acute aortic syndromes, acute dissection is the most common, but approximately 10% to 20% of patients (290–292) with a clinical picture of dissection exhibit an IMH via imaging without identification of blood flow in a false lumen or an intimal lesion. Some believe IMH arises from hemorrhage of the vasa vasorum located within the medial layer of the aorta (293,294), whereas others argue that the hematoma arises from microscopic tears in the aortic intima. The resulting hematoma may then propagate in an antegrade or a retrograde manner, producing symptoms that may be impossible to differentiate clinically from those of a classic AoD (61). IMH has a variable radiologic appearance (Figure 23) according to the area of the aorta involved. In some cases, IMH may be associated with a PAU (see Section 8.2 or 8.3).
Clinically, IMH most commonly occurs in the descending aorta and in older patients. Pain is characteristic of IMH, whereas malperfusion and pulse deficit are much less likely than with classic AoD (295,296).
Imaging criteria of IMH are based on the appearance of fresh thrombus in the aortic wall. These include crescentic or circular thickening of the aortic wall with maximal thickness greater than or equal to 7 mm on TEE without intimal flap or tear or longitudinal flow in the false lumen. The thickened wall has a higher tissue density than unenhanced blood on CT and is without enhancement after contrast on the CT/MR (290,296–298). When the term IMH is used strictly, no intimal defect such as a tear or an ulcer is present. But in practice, the term is used loosely to mean a thrombosed false lumen regardless of a small intimal defect. The distinction is further blurred by the facts that the intimal defect may be subtle and difficult to exclude and that some patients with IMH begin with a CT scan that shows a thrombosed false lumen with no apparent intimal defect and then over the course of 1 or 2 months develop 1 or more distinct ulcerlike communications (299,300). Because of this overlap in imaging findings, it is difficult and perhaps somewhat arbitrary to base treatment on the appearance of the CT snapshot of the aorta in its disease progression.
The natural history of IMH is variable. The hematoma may entirely resolve (10%) (301), it may convert to a classic dissection, or the aorta may enlarge and potentially rupture. The clinical behavior of IMH varies according to the location and mimics that of classic AoD. IMH involving the ascending aorta has a high, early risk of complication and death with medical treatment alone, and surgery is usually indicated. IMH involving the descending aorta may be treated with blood pressure control, and the use of beta blockers has been shown to improve the long-term survival rate (296). Conversion of the IMH to a more classic picture of dissection occurs in 3% to 14% of cases involving the descending aorta (297,298,302,303) and in 11% to 88% of cases involving the ascending aorta (302–305), with that figure increasing with increased length of follow-up. Progressive increase in aortic diameter has been demonstrated by serial imaging studies (298,306). In 1 study, the mortality after 2 years of patients with acute proximal Type A IMH versus that of patients with classic dissection was not significantly different (307). Another group found improved actuarial survival rates at 1, 2, and 5 years in patients with IMH versus classic dissection: 90%, 90%, and 90% versus 67%, 66%, and 62%, respectively, for Type A (304), and 100%, 97%, and 97% versus 83%, 79%, and 79%, respectively, for Type B (297). Song et al also described increased risk for complications or mortality for patients with IMH involving the ascending aorta when ascending aortic diameter is greater than 4.8 cm or IMH thickness is greater than 11 mm (308).
8.3 Penetrating Atherosclerotic Ulcer
PAU refers to an atherosclerotic lesion with ulceration that penetrates the internal elastic lamina and allows hematoma formation within the media of the aortic wall (309). This lesion sets the stage for development of IMH, AoD, or frank vessel rupture (309) (Figure 24). Anatomically PAUs develop in aortic segments where atherosclerotic changes are most common and therefore are localized to the descending thoracic aorta in over 90% of cases (310). When viewed tangentially, the classic appearance of the lesion is a mushroom-like outpouching of the aortic lumen with overhanging edges, resembling a gastric ulcer, as depicted on a barium study (Figure 24). The typical patient is elderly (usually over 65 years of age) and has hypertension and diffuse atherosclerosis, who presented with chest or back pain but without signs of aortic regurgitation or malperfusion. Less commonly, patients presented only with signs of distal embolization (311). Asymptomatic patients may also be found with aortic lesions that are indistinguishable, by imaging criteria, from PAUs (310,311).
Two entities can mimic PAUs. A branch artery pseudoaneurysm represents a small collection of flowing blood within an otherwise thrombosed aortic false lumen, which is created by an injury to a small branch artery during the propagation of the IMH (312). These are usually incidental findings, are distinguished from ulcers by the apparent absence of a communication with the aortic lumen by CT, and do not usually require specific treatment. A dissection entry or reentry tear may develop in an area of IMH as detected by several imaging studies over a period of several months (297,299,300,306).
8.4 Pseudoaneurysms of the Thoracic Aorta
Pseudoaneurysms of the thoracic aorta are frequently related to deceleration or torsional trauma to the aorta from motor vehicle accidents, falls, and sports injuries (3,313–319). Aortic pseudoaneurysms are relatively rare, with posttraumatic pseudoaneurysms having an incidence of 3% to 4% after blunt trauma (51). Other pseudoaneurysms may occur following aortic surgery, catheter-based interventions, or penetrating trauma. Pseudoaneurysms often have a slim “neck” that leads to the “aneurysm” that corresponds to points of penetration and containment (320). Aortic infections (mycotic aneurysms) and penetrating ulcers may also result in pseudoaneurysms (see Section 9.2.2). Penetrating injuries are usually repaired immediately whenever recognized and feasible (317,318,321).
8.5 Traumatic Rupture of the Thoracic Aorta
A UK survey of all motor vehicle accident fatalities found that approximately 20% of patients had an autopsy finding of a ruptured aorta, emphasizing the importance of traumatic rupture of the aorta (TRA). In the United States, there are around 40 000 motor vehicle deaths annually, and it is likely that around 8000 of the victims had TRA. It is estimated that only 9% to 14% of patients with TRA reach a hospital alive and only 2% ultimately survive. In this survey, 29% were involved with frontal impact crashes and 44% were involved with side impact crashes (318,319).
Parmley and colleagues (318) noted the correlation of high risk of early death and the sites of TRA on the basis of autopsies in 275 deaths from unrelated aortic rupture. In 45%, the tear was at the aortic isthmus; 23%, in the ascending aorta; 13%, in the descending aorta; 8%, in the transverse aorta; 5%, in the abdominal aorta; and 6%, multiple sites.
Examination of the patient usually reveals signs similar to those of coarctation of the aorta with arm blood pressure higher than leg blood pressure, delay between radial versus femoral artery pulsation, and a harsh interscapular murmur (3). Evidence of polytrauma is, however, common.
The best method for detection of a TRA is debated. A chest x-ray with a nasogastric tube in position has 80% sensitivity for suggesting TRA by showing displacement of the nasogastric tube by the hematoma. However, signs of hemomediastinum are more often false positive than true positive (40). Even when present, mediastinal blood is less likely to be due to arterial/aortic injury than to less-consequential venous bleeding. A biplane contrast aortogram may fail to detect the tear until the development of a pseudoaneurysm. TEE may be used, but if dilatation has not occurred, the diagnosis may still be in doubt. CT is used but is not absolutely certain to establish the diagnosis. In questionable cases, intravascular ultrasound can also be used (3,322,323) (Figure 8). Realistically, the imaging sequence often depends on the stability of the patient and the need for the diagnosis of concomitant injuries. Sometimes, this may even fail to detect the tear, and the study may have to be repeated at a later date to detect the tear.
8.6 Evaluation and Management of Acute Thoracic Aortic Disease
8.6.1 Initial Evaluation and Management
126.96.36.199 Recommendations for Estimation of Pretest Risk of Thoracic Aortic Dissection
1. Providers should routinely evaluate any patient presenting with complaints that may represent acute thoracic aortic dissection to establish a pretest risk of disease that can then be used to guide diagnostic decisions. This process should include specific questions about medical history, family history, and pain features as well as a focused examination to identify findings that are associated with aortic dissection, including:
a. High-risk conditions and historical features (47,127,223,261) (Level of Evidence: B):
• Marfan syndrome, Loeys-Dietz syndrome, vascular Ehlers-Danlos syndrome, Turner syndrome, or other connective tissue disease.
• Patients with mutations in genes known to predispose to thoracic aortic aneurysms and dissection, such as FBN1, TGFBR1, TGFBR2, ACTA2, and MYH11.
• Family history of aortic dissection or thoracic aortic aneurysm.
• Known aortic valve disease.
• Recent aortic manipulation (surgical or catheter-based).
• Known thoracic aortic aneurysm.
b. High-risk chest, back, or abdominal pain features (37,47,215,223,261,264,288) (Level of Evidence: B):
• Pain that is abrupt or instantaneous in onset.
• Pain that is severe in intensity.
• Pain that has a ripping, tearing, stabbing, or sharp quality.
c. High-risk examination features (37,47,253,257,261,324) (Level of Evidence: B):
• Pulse deficit.
• Systolic blood pressure limb differential greater than 20 mm Hg.
• Focal neurologic deficit.
• Murmur of aortic regurgitation (new).
2. Patients presenting with sudden onset of severe chest, back, and/or abdominal pain, particularly those less than 40 years of age, should be questioned about a history and examined for physical features of Marfan syndrome, Loeys-Dietz syndrome, vascular Ehlers-Danlos syndrome, Turner syndrome, or other connective tissue disorder associated with thoracic aortic disease. (223) (Level of Evidence: B)
3. Patients presenting with sudden onset of severe chest, back, and/or abdominal pain should be questioned about a history of aortic pathology in immediate family members as there is a strong familial component to acute thoracic aortic disease. (223) (Level of Evidence: B)
4. Patients presenting with sudden onset of severe chest, back, and/or abdominal pain should be questioned about recent aortic manipulation (surgical or catheter-based) or a known history of aortic valvular disease, as these factors predispose to acute aortic dissection. (Level of Evidence: C)
5. In patients with suspected or confirmed aortic dissection who have experienced a syncopal episode, a focused examination should be performed to identify associated neurologic injury or the presence of pericardial tamponade (seeSection 8.1.6). (Level of Evidence: C)
6. All patients presenting with acute neurologic complaints should be questioned about the presence of chest, back, and/or abdominal pain and checked for peripheral pulse deficits as patients with dissection-related neurologic pathology are less likely to report thoracic pain than the typical aortic dissection patient (253) (seeSection 8.1.7). (Level of Evidence: C)
188.8.131.52 Laboratory Testing
Several plasma markers have been investigated for their utility in the evaluation of acute AoD. Plasma smooth muscle myosin heavy chain protein, D-dimer, and high-sensitivity C-reactive protein have shown diagnostic promise, although a lack of large prospective studies precludes a recommendation regarding their use (325–328).
Elevation of D-dimer levels occurs with intravascular activation of the coagulation cascade and secondary fibrinolysis and in conditions such as venous thromboembolism, sepsis, disseminated intravascular coagulation, malignancies, recent trauma or surgery, and acute MI and following fibrinolytic therapy. The ACEP has published guidelines regarding the use of certain D-dimer assays to rule out pulmonary embolism in low-risk patients (329).
Regarding the potential role of plasma D-dimer levels to screen for AoD, significant elevations of D-dimer were seen in all 24 patients with documented acute AoD involving either the ascending or descending thoracic aorta regardless of time from presentation, ranging from 1 to 120 hours (325). A meta-analysis of 11 studies (330) noted that the pooled sensitivity of D-dimer in 349 patients with documented acute AoD was 94% (95% CI 91% to 96%) with specificity ranging from 40% to 100%. Two patients had limited ascending aortic IMH without intimal flap and had negative D-dimer assays (331).
Some authors (325,332) recommend that D-dimer assays be performed in all patients where clinical suspicion exists, to help identify those who do not require definitive imaging studies. However, the efficacy and safety of this strategy have not been tested in a large clinical trial, and several caveats should apply. The negative likelihood ratio provided by the most sensitive D-dimer assay is not of sufficient magnitude to provide useful information in high-risk individuals and therefore cannot be used to “rule out” the disease in this group. Clinical scoring systems to identify the true pretest probability for AoD in individual patients have not been developed or validated, thus limiting an accurate determination of the true “posttest” probability associated with a negative D-dimer result. Finally, there are reports of negative D-dimer assays associated with ascending aortic IMH or a thrombosed false lumen, such that further studies are needed regarding the sensitivity of D-dimer levels to detect the presence of IMH or PAU (331). Given these issues, the writing committee cannot recommend serum D-dimer screening for all patients being evaluated for AoD.
Where a high level of suspicion for acute AoD exists, laboratory testing aimed at presurgical screening (blood count, serum chemistries, coagulation profiles, and blood type and screen) may reduce preoperative delays.
184.108.40.206 Recommendations for Screening Tests
1. An electrocardiogram should be obtained on all patients who present with symptoms that may represent acute thoracic aortic dissection.
a. Given the relative infrequency of dissection-related coronary artery occlusion, the presence of ST-segment elevation suggestive of myocardial infarction should be treated as a primary cardiac event without delay for definitive aortic imaging unless the patient is at high risk for aortic dissection. (37,47,333) (Level of Evidence: B)
2. The role of chest x-ray in the evaluation of possible thoracic aortic disease should be directed by the patient's pretest risk of disease as follows:
a. Intermediate risk: Chest x-ray should be performed on all intermediate-risk patients, as it may establish a clear alternate diagnosis that will obviate the need for definitive aortic imaging. (Level of Evidence: C)
b. Low risk: Chest x-ray should be performed on all low-risk patients, as it may either establish an alternative diagnosis or demonstrate findings that are suggestive of thoracic aortic disease, indicating the need for urgent definitive aortic imaging. (Level of Evidence: C)
3. Urgent and definitive imaging of the aorta using transesophageal echocardiogram, computed tomographic imaging, or magnetic resonance imaging is recommended to identify or exclude thoracic aortic dissection in patients at high risk for the disease by initial screening. (42–46,67,73) (Level of Evidence: B)
1. A negative chest x-ray should not delay definitive aortic imaging in patients determined to be high risk for aortic dissection by initial screening. (Level of Evidence: C)
220.127.116.11 Recommendations for Diagnostic Imaging Studies
1. Selection of a specific imaging modality to identify or exclude aortic dissection should be based on patient variables and institutional capabilities, including immediate availability. (Level of Evidence: C)
2. If a high clinical suspicion exists for acute aortic dissection but initial aortic imaging is negative, a second imaging study should be obtained. (212) (Level of Evidence: C)
18.104.22.168 Recommendations for Initial Management
1. Initial management of thoracic aortic dissection should be directed at decreasing aortic wall stress by controlling heart rate and blood pressure as follows:
a. In the absence of contraindications, intravenous beta blockade should be initiated and titrated to a target heart rate of 60 beats per minute or less. (Level of Evidence: C)
b. In patients with clear contraindications to beta blockade, nondihydropyridine calcium channel-blocking agents should be used as an alternative for rate control. (Level of Evidence: C)
c. If systolic blood pressures remain greater than 120 mm Hg after adequate heart rate control has been obtained, then angiotensin-converting enzyme inhibitors and/or other vasodilators should be administered intravenously to further reduce blood pressure that maintains adequate end-organ perfusion. (Level of Evidence: C)
d. Beta blockers should be used cautiously in the setting of acute aortic regurgitation because they will block the compensatory tachycardia. (5) (Level of Evidence: C)
1. Vasodilator therapy should not be initiated prior to rate control so as to avoid associated reflex tachycardia that may increase aortic wall stress, leading to propagation or expansion of a thoracic aortic dissection. (Level of Evidence: C)
22.214.171.124 Recommendations for Definitive Management
1. Urgent surgical consultation should be obtained for all patients diagnosed with thoracic aortic dissection regardless of the anatomic location (ascending versus descending) as soon as the diagnosis is made or highly suspected. (Level of Evidence: C)
2. Acute thoracic aortic dissection involving the ascending aorta should be urgently evaluated for emergent surgical repair because of the high risk of associated life-threatening complications such as rupture. (47) (Level of Evidence: B)
3. Acute thoracic aortic dissection involving the descending aorta should be managed medically unless life-threatening complications develop (e.g., malperfusion syndrome, progression of dissection, enlarging aneurysm, inability to control blood pressure or symptoms). (285,288,334–337) (Level of Evidence: B)
Early identification of acute thoracic dissection is challenging. During the initial evaluation, the correct diagnosis of AoD has been made in only 15% to 43% of patients initially thought to have the disease (215,237,252). Factors that may impede accurate diagnosis of AoD include the following:
1. Acute AoD is often believed to be a rare disease (2.9 to 3.5 cases per 100 000 person-years) (215,217), whereas the incidence of acute MI is several orders of magnitude greater (more than 200 cases per 100 000 person-years) (338). Data previously cited in this guideline from UHC suggest that acute AoD is not “rare.” Some common explanations that clinicians give for this underlying belief include the following:
a. It is difficult for clinicians to effectively separate patients with AoD from the multitude of other patients who present to emergency departments and primary care physicians with common complaints that are much more often not due to acute AoD.
b. Front-line medical providers may have little direct experience with acute AoD and are unlikely to be aware of the subtleties of its presenting signs and symptoms.
c. Institutional multidisciplinary pathways developed for other emergencies such as ST-elevation myocardial infarction (STEMI) or acute stroke have not generally existed for acute AoD.
2. In contrast to other more common cardiovascular emergencies, acute AoD may occur in younger patients. Accurate identification of acute AoD; however, requires that clinicians recognize on a routine basis that the disease may present in younger patients. Case reports exist of children as young as 3 years of age presenting with acute AoD (105).
3. Acute AoD may present with a wide range of unusual manifestations that do not conform with classic “textbook findings.” (339–341)
4. There are no well-studied, rapidly available, and effective screening tests for acute AoD.
8.6.2 Evaluation and Management Algorithms
The provided algorithms guide the initial evaluation of patients whose presentations are concerning for AoD and the management of patients in whom the diagnosis of acute thoracic AoD is confirmed. Although clinicians may ultimately choose to deviate from the pathway for patient-specific reasons, the algorithms provide a framework with which to quickly diagnose (Figure 25) and provide early management (Figure 26) of AoD. This decision model is supported by several large studies that indicate a targeted history and physical examination are likely to identify the vast majority of patients who present with acute AoD, suggesting that adequate screening need not be time intensive or technology dependent (47,215,264). Using a target history and physical examination, patients can be placed into 1 of 3 categories: 1) those with immediately apparent acute AoD requiring emergent surgical evaluation and expedited aortic imaging, 2) those whose presentation is concerning for acute AoD and in the absence of a clear alternative diagnosis require expedited aortic imaging, and 3) those whose clinical presentation is not initially suggestive of acute AoD but may benefit from aortic imaging in the absence of a likely alternative diagnosis at the completion of the initial evaluation.
Several high-risk conditions (Figure 25, T2–1) greatly increase the likelihood that presenting complaints that could be a result of acute AoD.
Pain (Figure 25, T2–2) is the most commonly reported presenting symptom of acute AoD regardless of patient age, sex, or other associated clinical complaint (37,47,215,223,242,264,289). Pain described as abrupt or instantaneous in onset, that is severe in intensity, or that has a ripping or tearing quality establishes a high pretest probability for AoD (47,261).
The combination of 2 or more high-risk features (Figure 25, T3) is strongly suggestive of acute AoD.
The presence of a single high-risk feature (i.e., high-risk condition, pain, or physical examination) may trigger immediate concern for acute AoD; however, other diagnostic considerations may exist. Pain with high-risk features, although suggestive of acute AoD, may occur as a result of an alternate disease process (Table 12). For patients who are determined to have an intermediate probability of acute AoD on the basis of initial bedside assessment, Figure 25, T4, provides a pathway for further evaluation.
For patients presenting with new ST-segment elevations on the initial ECG and without high-risk AoD features, immediate coronary angiography and reperfusion therapy (i.e., thrombolysis or percutaneous coronary intervention) are indicated (333). However, if coronary angiography is performed and no culprit coronary lesion is identified, then Figure 25, T6, provides a pathway to dedicated aortic imaging. As approximately 40% of chest films in acute AoD lack a widened mediastinum, and as many as 16% are normal, the absence of radiographic abnormalities does not exclude the diagnosis of AoD (37,47) (Figure 25, T7).
Missed or delayed diagnosis of acute AoD is most commonly ascribed to an incorrect working diagnosis of acute coronary syndrome, a condition that may require a prolonged time interval to correctly identify (i.e., serial cardiac biomarkers) and whose management with antiplatlet and antithrombin agents may cause harm to the patient with AoD. For patients with an intermediate-risk profile for acute AoD and who do not have diagnostic STEMI but who are being evaluated for a possible acute coronary syndrome, aortic imaging may detect AoD prior to the administration of antiplatelet and antithrombin agents (Figure 25, T8).
Unexplained hypotension is present in approximately 20% of patients with acute AoD (37,47). Similarly, a widened mediastinum on chest x-ray strongly suggests the need for additional definitive diagnostic aortic imaging, particularly in patients without a clear alternative explanation for their presenting complaint (342) (Figure 25, T9).
Some patients with acute AoD present without any high-risk features, making early diagnosis difficult. If a clear alternative diagnosis is not established after the initial evaluation, then obtaining a diagnostic aortic imaging study, particularly in patients with advanced age (older than 70 years), syncope, focal neurologic deficit, or recent aortic manipulation (surgery or catheter based), should be considered (47,242,253,288,343) (Figure 25, T10).
Multidetector CT with contrast, TEE, and MR all provide acceptable diagnostic accuracy for the diagnosis of acute AoD. In patients with hemodynamic instability, requiring close monitoring, bedside TEE is preferred to avoid moving the patient out of the acute care environment (Figure 25, T11).
The most recent comparative study with nonhelical CT, 0.5 Tesla MR and TEE showed 100% sensitivity for all modalities, with better specificity of CT (100%) than for TEE and MR (44). A recent meta-analysis that evaluated the diagnostic accuracy of TEE, helical CT, and MR for suspected AoD found that all 3 imaging techniques provided equally reliable diagnostic values (46). Accordingly, selection of an imaging modality is influenced by individual patient variables and institutional capabilities.
The diagnosis of acute AoD cannot be excluded definitively based on the results of a single imaging study. Although TEE, CT, and MR are all highly accurate for the evaluation of acute AoD; false-negative studies can and do occur (47) (Figures 9 and 15). If a high clinical suspicion exists for acute AoD but initial aortic imaging is negative, strongly consider obtaining a second imaging study (Figure 25, T12).
8.6.3 Initial Management
Once the diagnosis of AoD or one of its anatomic variants (IMH or PAU) is obtained, initial management is directed at limiting propagation of the false lumen by controlling aortic shear stress while simultaneously determining which patients will benefit from surgical or endovascular repair (Figure 26).
126.96.36.199 Blood Pressure and Rate Control Therapy
Aortic wall stress is affected by the velocity of ventricular contraction (dP/dt), the rate of ventricular contraction, and blood pressure. Initial medical stabilization using beta blockers controls these 3 parameters by reducing heart rate and blood pressure to the lowest amounts that will still maintain adequate end-organ perfusion (61). Reasonable initial targets are a heart rate less than 60 bpm and a systolic blood pressure between 100 and 120 mm Hg (61).
Intravenous propranolol, metoprolol, labetalol, or esmolol is an excellent choice for initial treatment. In patients who have a potential contraindication to beta blockade (e.g., those with asthma, congestive heart failure, or chronic obstructive pulmonary disease), esmolol may be a viable option given its extremely short half-life. Use of labetalol, which is both an alpha- and beta-receptor antagonist, offers the advantage of potent heart rate and blood pressure control from a single agent, potentially eliminating the need for a secondary vasodilator. In patients who are unable to tolerate beta blockade, nondihydropyridine calcium channel antagonists (verapamil, diltiazem) offer an acceptable, although less-established, alternative (61). Beta blockers, verapamil, or diltiazem for rate control in patients with significant aortic regurgitation may be problematic because of deleterious affects on reflex tachycardia.
188.8.131.52 Additional Antihypertensive Therapy
It is frequently difficult to reduce blood pressure to optimum levels (285–287,344–346). In 1 study, patients required a median of 4 different antihypertensive drugs (347). In addition to beta blockade, vasodilators may be required to control blood pressure. Intravenous sodium nitroprusside is the most established agent and offers the advantage of being rapidly titratable (61). Nicardipine (348), nitroglycerin, fenoldopam, and various other intravenous antihypertensive agents are appropriate for this situation. Vasodilator therapy without prior beta blockade may cause reflex tachycardia and increased force of ventricular contraction leading to greater aortic wall stress and potentially causing false lumen propagation (42). Following initial stabilization with intravenous antihypertensives, most patients will require long-term antihypertensive treatment including the use of a beta blocker plus additional classes of agents. Angiotensin-converting enzyme inhibitors or angiotension receptor blockers may retard aortic dilatation and their use may be indicated as outlined in Section 184.108.40.206.
220.127.116.11 Pain Control
Adequate pain control is essential in the setting of acute AoD to decrease sympathetic mediated increases in heart rate and blood pressure. Appropriate use of intravenous opiate anal-gesia will help augment the effects of rate control and vasodilator agents.
Medical management options for all forms of dissection-related hypotension are limited. Volume administration titrated to improvement of blood pressure is a reasonable first approach. Vasopressors can be added, if needed, to maintain adequate perfusion but have the potential to cause further false lumen propagation. Inotropic agents are likely to increase the force and rate of ventricular contraction and therefore increase sheer stress on the aortic wall.
Pericardiocentesis for dissection-related hemopericardium has been associated with recurrent pericardial bleeding and associated mortality (307,349,350). Several articles from the Asian literature suggest that pericardiocentesis may be safe in the setting of acute Type A IMH (351,352). Other cardiac complications that may result in hypotension include severe dissection-related aortic regurgitation, true lumen obstruction by a compressing false lumen, and acute MI. All require definitive operative management. Hypotension or shock in the setting of AoD may also result from contained rupture of the false lumen into adjacent structures (i.e., pleural space or mediastinum), a scenario that also mandates immediate operative intervention.
Ultimately, hypotension or shock in the acute AoD patient suggests the need for immediate operative management. For patients with hemopericardium and cardiac tamponade who cannot survive until surgery, pericardiocentesis can be performed by withdrawing just enough fluid to restore perfusion.
18.104.22.168 Determining Definitive Management
In the clinically stable patient, the decision for surgical versus medical management of patients with acute AoD is based primarily on the location of the dissection as described by the Stanford and DeBakey classification systems (61,353) (see Section 8.1.2). A prompt cardiac surgical consultation provides the best management resource, regardless of location of the AoD, as it is impossible to predict which complications may develop or when they may occur.
8.6.4 Recommendation for Surgical Intervention for Acute Thoracic Aortic Dissection
1. For patients with ascending thoracic aortic dissection, all of the aneurysmal aorta and the proximal extent of the dissection should be resected. A partially dissected aortic root may be repaired with aortic valve resuspension. Extensive dissection of the aortic root should be treated with aortic root replacement with a composite graft or with a valve sparing root replacement. If a DeBakey Type II dissection is present, the entire dissected aorta should be replaced. (Level of Evidence: C)
When a Type A AoD involves the aortic root, resuspension of the valve with preservation of the aortic sinuses and excision of the sinuses and resuspension of the valve within a polyester graft are suitable options. If the aortic root is dilated, or if there is extensive dissection and disruption of the aortic sinuses, replacement with a composite graft is necessary.
8.6.5 Endovascular Interventions
Endovascular stent grafts are not approved for AoD involving the ascending aorta or aortic arch. Endovascular stent grafts used for descending thoracic aortic dissection is discussed in Section 22.214.171.124.1. Indications for either surgical or endovascular interventions are discussed in Section 9.
8.6.6 Principles of Treatment for Intramural Hematoma and Penetrating Atherosclerotic Ulcer
The goals of treatment are to prevent aortic rupture or progression to classic AoD, allow patient stabilization before urgent surgery, and reduce complexity of unavoidable aortic surgery. Aggressive medical treatment usually includes, particularly in symptomatic patients, beta blockers and other antihypertensive therapy. Indications for open or endograft treatment are based on the anatomic features of the lesion, clinical presentation and course, patient comorbidities, and anatomic constraints related to endograft technology. Treatment by endografts or open aortic reconstruction can be discussed in the context of 3 overlapping aortic lesions: intimal defect without IMH, intimal defect with IMH, and IMH without an intimal defect.
126.96.36.199 Intimal Defect Without Intramural Hematoma
These are localized lesions and may involve a limited segment of the aorta. They are often an incidental finding. By imaging criteria, they include uncomplicated aortic ulcers, blebs, and eccentric or saccular aneurysms of the aorta. They are treated as saccular aneurysms based on their maximum diameter and clinical feature (212,306,354). These lesions can be treated with open reconstruction and are the most suitable of the 3 groups for treatment by endografts, if in the descending thoracic aorta. They involve a limited segment, which can easily be excluded from the circulation, as long as there is an adequate distance from a critical branch artery. When these limited dissections involve the ascending aorta, emergency surgery is indicated as for other types of AoD because rupture or cardiac tamponade can occur (212,306,354).
188.8.131.52 Intimal Defect With Intramural Hematoma
The intimal defect again presents a target lesion for endovascular treatment in the descending thoracic aorta, but the associated IMH involves a longer segment of aorta than the first category. If the patient becomes asymptomatic in response to aggressive medical treatment, it may be possible to delay endovascular or open reconstruction until the IMH has reabsorbed and organized. (Of note, some writing committee members have observed healing of IMH such that immediate reconstruction was not required but have continued to follow that small number of patients closely.) Two considerations affect the length of aorta bordering the intimal defect, which is to be included in the segment targeted for treatment. Evidence of adjacent atheromatous wall should favor more extensive treatment of the aorta with longer endografts, because radiographic imaging underestimates shallow ulcerated atheromas, and the ulcer typically arises in a bed of atheromatous intima. Treatment with longer endografts provides a safety margin against undertreating the intimal defect. The second consideration is the extent of associated IMH. The self-expanding endograft may tear through the intimal surface into underlying thrombosed false lumen. When treatment of this lesion in the acute stage is clinically necessary (e.g., persisting pain, evidence for expansion or rupture, compromise of critical branches), it is preferable to anchor the endograft in the noninvolved wall above and below the intimal defect.
184.108.40.206 Recommendation for Intramural Hematoma Without Intimal Defect
1. It is reasonable to treat intramural hematoma similar to aortic dissection in the corresponding segment of the aorta. (Level of Evidence: C)
As noted earlier, some authors suggest treating IMH as an AoD in the corresponding aortic territory. Others recommend invasive treatment regardless of location or aortic diameter (354). However, small patient series, incomplete anatomic description of case material, and lack of explicit anatomic or clinical guidelines indicating open or endovascular aortic repair make it difficult to generalize from the literature.
The absence of an intimal defect, which can serve as a target lesion, presents a diagnostic as well as treatment challenge. Intimal tears can be extremely subtle, depending on the size of the intimal tear and the amount of intramedial thrombus, which can sometimes fill the cavity flush with the aortic lumen. The intimal tear may be remote in the aorta despite leaking into the chest. IMH in a normal caliber aorta without an apparent intimal tear precludes limited treatment of a target lesion. There are no data supporting prophylactic implantation of endografts covering the entire descending aorta, yet in unusual circumstances one may be forced to propose such treatment. IMH in an aneurysmal aorta presents a particularly urgent problem, because this complication may be a precursor to aneurysm rupture. Although the literature gives no compelling guidelines for treatment, the writing committee believes that treatment of IMH corresponding to treatment of AoD in the corresponding segment of the aorta is reasonable.
8.7 Treatment for the Management of Traumatic Aortic Rupture
The management of blunt TRA is evolving (3,318,319,355,356). On the basis of the report by Parmley and colleagues (318), most surgeons have recommended immediate surgical repair. However, when other serious traumatic injuries are present including head injuries and long bone or pelvic fractures, immediate surgery may not be feasible or may be dangerous. Multiple studies appear to show that if careful blood pressure control is used, many patients can be treated initially conservatively and then undergo operation once their other injuries have been stabilized (3,321,357–359). In a review by Svensson et al (359) of 44 patients initially treated with careful blood pressure control who subsequently had delayed open surgery, there were no operative deaths. This approach has also been reported as being safe by Pate and colleagues and others (3,321,357,358). Thus, in selected patients at high risk for other injuries and bleeding, delayed repair of traumatic contained rupture of the aorta may be an option.
The open surgical repair of TRA has evolved over time. In a meta-analysis of 596 TRA patients by Svensson et al (359), the highest mortality rate was noted with cardiopulmonary bypass (16.7%), the rate was less with shunts (11.4%), and the rate was lowest with a simple “clamp and sew” approach (5.8%, P<0.01). There was no difference in the risk of postoperative paralysis. Subsequently, von Oppell and colleagues (360) reviewed 1742 patients and found the risk of death was 18.2% with cardiopulmonary bypass, 11.9% for distal perfusion with atriofemoral bypass, 12.3% for shunts, and 16% for the “clamp and sew” method. The respective paralysis rates were 2.4%, 1.7%, 11.1%, and 19.2%, respectively. The key factor in open repairs has been to keep the total aorta cross-clamp time to as short a period as possible, especially less than 45 minutes (3,359,361–364).
The latest evolution in managing TRA is the use of endovascular deployed graft covered stents. Although endovascular stent grafting has not been prospectively studied for this clinical scenario, US Food and Drug Administration–approved devices are being used “off label,” with considerable success reported based on retrospective studies. In a collected series of 284 patients reported in the literature, Lettinga-van de Poll and colleagues reported the procedure-related mortality was 1.5%, 6.7% had endoleaks, and 14.4% had procedure-related complications (358,365–370). In a multicenter study of 30 patients with 100% implantation success, 6% to 7% of patients died, 1 had a stroke (3.3%), and 1 had partial stent collapse (3.3%) (370). Reporting bias of favorable results may be an issue regarding interpretation of the safety and efficacy of this approach.
The problems with endovascular grafting for TRA have included the need to cover the left subclavian artery; the acute sharp angle of the distal aortic arch, particularly in young patients; and the lack of sufficiently small prostheses for use in young patients. The size and angle problem can result in the “bird beak” deformity, where the proximal edge of the stent is not in contact with the aortic wall and can result in lifting or collapse of the stent (366) (Figure 27). Similarly, when the stent graft is larger than the aortic diameter, enfolding of the stent and collapse can occur. On cross-sectional views, the stent graft has the appearance of a diagrammatic heart.
Over time, it is hoped newer iterations of endografts will be developed that are better able to accommodate the angulation of the distal arch and are smaller. It is unlikely that a prospective randomized study will be performed for this because of the small number of patients who make it to any surgical center and because initial results with endografting of TRA, with exception of the problems listed earlier, have been reasonable. The Expert Opinion Committee of the Society of Thoracic Surgeons and the American Association of Thoracic Surgeons suggested that both acute and chronic ruptures be considered for treatment with endografts. Some measures of caution must be taken because these are young patients who may be subjected to cumulative radiation exposure with multiple CT scans and because the long-term durability of endovascular stent grafts is not known (371).
9 Thoracic Aortic Aneurysms
Most thoracic aortic aneurysms are caused by degenerative disease resulting in dilatation of the aorta (Figure 28). The incidence of thoracic aortic aneurysms is estimated to be increasing and there are around 10.4 cases per 100 000 person-years (372).
Risk factors for development of thoracic aortic aneurysms include hypertension, smoking, and chronic obstructive pulmonary disease. In addition, several genetic syndromes with a predisposition for thoracic aortic aneurysms have been identified and are listed in Section 5. Thoracic aortic aneurysms are also associated with bicuspid aortic valve (see Section 6.1) and other congenital cardiovascular anomalies (see Section 6) and inflammatory diseases (see Section 7). Some thoracic aortic aneurysms are due to an inheritance of a predisposition for the disease, termed familial thoracic aortic aneurysm syndrome (see Section 5.1.6), and still others are idiopathic.
Many patients with a thoracic aortic aneurysm are asymptomatic and diagnosed by chest x-ray or CT scan obtained for other reasons. An aneurysm may cause compressive symptoms on adjacent structures including hoarseness, from left recurrent laryngeal nerve stretching; stridor, from tracheal or bronchial compression; dyspnea, from lung compression; dysphagia, from esophageal compression; and plethora and edema, from superior vena cava compression. Aortic valve regurgitation may develop due to aortic root or ascending aortic dilatation and result in heart failure. Neck and jaw pain may occur with aortic arch aneurysms, whereas back, interscapular, and/or left shoulder pain may occur with descending thoracic aortic aneurysms. Embolization of atherosclerotic debris with end-organ symptoms may occur. Finally, acute syndromes including dissection or rupture without dissection may occur with potentially catastrophic outcomes as described in Section 8.5.
Thoracic aortic aneurysms may involve different segments of the aorta. The ascending thoracic aorta and/or root is most commonly involved, with the descending aorta involved less often. Involvement of the aortic arch occurs in only 10%. The etiology, natural history, and treatments differ somewhat for aneurysms in each location. In Marfan syndrome, aneurysms typically arise in the aortic root, a process often referred to as annuloaortic ectasia. Because the leaflets of the aortic valve are suspended within the root, successful repair of the aortic root may require performance of a valve-sparing root repair or, in some cases, a composite aortic graft.
The average rate of expansion of thoracic aortic aneurysms is estimated to be 0.10 to 0.42 cm/y (373–375). Medical and surgical treatment considerations and selection criteria are noted in Section 9.2.
A leaking or ruptured aneurysm (see Section 220.127.116.11) may also present as chest pain with hypotension due to hemorrhage into the left or right pleural space or pericardium (376); an aortoesophageal fistula may manifest as gastrointestinal hemorrhage (377). An unusual manifestation reported is a hemoptysis from a ruptured ascending aortic aneurysm eroding into the left lung bronchus (378).
Further anatomic classifications refer to segments of the descending thoracic aorta and thoracoabdominal aorta divided into subsections according to the extent of the disease that is replaced at the time of surgery. These extents have an important influence on the risk of death and complications after surgery or stenting (Figure 29) (see Section 18.104.22.168).
9.1 General Approach to the Patient
9.1.1 Recommendation for History and Physical Examination for Thoracic Aortic Disease
1. For patients presenting with a history of acute cardiac and noncardiac symptoms associated with a significant likelihood of thoracic aortic disease, the clinician should perform a focused physical examination, including a careful and complete search for arterial perfusion differentials in both upper and lower extremities, evidence of visceral ischemia, focal neurologic deficits, a murmur of aortic regurgitation, bruits, and findings compatible with possible cardiac tamponade. (339–341) (Level of Evidence: C)
The physical findings of thoracic aortic diseases may be subtle indirect manifestations of uncommon underlying genetically predisposing conditions. In evaluating the evidence base for physical examination of patients with thoracic aortic disease, there are no controlled or blinded experimental research studies that have stratified patients into different treatment categories based on physical findings.
For all patients with thoracic aortic disease, the first and foremost issue is to identify those who are acutely at risk for catastrophic harm as early as possible. Establishing a set of “triggers” or “red flags” may serve as alerts to either exclude or identify life-threatening thoracic aortic disease.
Given the growing awareness of an extensive variety of diseases associated with nonemergent thoracic aortic disease, it is important to be aware of the many different physical findings associated with extracardiovascular etiologies particularly those of genetic origin (see Section 5).
22.214.171.124 Coronary Artery Disease
The frequency of coexisting CAD varies widely among patient subgroups with thoracic aortic disease as does the etiology of the coronary artery abnormalities, if present. Patients with atherosclerotic aneurysms of the descending aorta are at elevated risk for coronary atherosclerosis, particularly if they have multiple atherosclerotic risk factors. While patients with Type A dissection or annuloaortic ectasia may be protected from atherosclerosis (383), patients with Takayasu arteritis may occasionally have inflammatory coronary involvement with coronary aneurysms (less than 10%) (384,385). Similarly, an occasional patient with GCA may have coronary artery involvement (386,387). If ascending aortic surgery is being considered, with or without aortic valve surgery, then identification of the coronary anatomy and any underlying CAD is important for planning the best operation.
Embolization of material thrombus, atheromatous debris, or cholesterol crystals may affect any distal arterial bed (see Section 11.3). Embolization may occur in patients with thoracic aortic aneurysms or atheromas and in those who have undergone angiography, major vessel surgery, or thrombolytic therapy (388–397). Clinical consequences of such embolization vary considerably, from being completely asymptomatic to presenting with acute multiorgan failure, including progressive renal failure or cutaneous involvement, with a mortality rate as high as 70% to 90% (398).
126.96.36.199 Associated Renal Ischemia
Renal complications of thoracic aortic disease may be acute, subacute, and chronic (399). Patients may present with severe abdominal or flank pain, hematuria, fever, nausea, or a combination of these signs and symptoms (400).
188.8.131.52 Associated Mesenteric Ischemia
Patients with acute intestinal ischemia have severe abdominal pain that is initially out of proportion to physical findings (401). Hours to days later, peritonits and sepsis correlate with intestinal perforation. Findings suggestive of intestinal ischemia as well as specific arterial or venous obstruction require further surgical or vascular specialist evaluation (402–406).
184.108.40.206 Associated Peripheral Ischemia
Acute limb ischemia results in pain, pallor, paraesthesias, and paralysis (4). Noninvasive vascular diagnostic testing (e.g., ankle- and toe-brachial indices, segmental pressure measurements, pulse volume recordings, duplex ultrasound imaging, and Doppler waveform analysis) may document ischemia with additional use of angiographic imaging when necessary (407).
9.1.2 Differential Diagnosis
Symptoms are most commonly related to pain or discomfort. Particularly large thoracic aneurysms may be associated with chest discomfort. Rarely, dysphagia (dysphagia lusoria) or dyspnea is present, usually related to congenital distal arch lesions, such as aberrant right subclavian artery and Kommerell diverticulum or Felson and Palayew Type I or II right-sided aortic arch lesions (408).
History of fevers may be related to inflammatory disease or mycotic aneurysms. Occasionally, with chronic dissection and leaking aneurysms, the reabsorption of blood may be associated with fever or jaundice.
220.127.116.11 Physical Findings
Most physical findings are not specific for thoracic aortic disease. Other findings may be related to genetic syndromes and connective tissue disorders (see Section 5) or inflammatory diseases (see Section 7). Findings associated with coarctation of the aorta include brachial-femoral pulse delay and murmurs.
9.1.3 Considerations for Imaging
Because most cases of chronic thoracic aortic disease are asymptomatic and difficult to detect on physical examination, the clinician must have a low threshold for screening for thoracic aortic disease. CT or MR is required to adequately visualize the affected aorta. There has been no cost–benefit analysis of screening in these populations (see Section 18.1).
9.2 General Medical Treatment and Risk Factor Management for Patients With Thoracic Aortic Disease
9.2.1 Recommendation for Medical Treatment of Patients With Thoracic Aortic Diseases
1. Stringent control of hypertension, lipid profile optimization, smoking cessation, and other atherosclerosis risk-reduction measures should be instituted for patients with small aneurysms not requiring surgery, as well as for patients who are not considered to be surgical or stent graft candidates. (Level of Evidence: C)
The patient's general status should be optimized where possible. Respiratory illness, a common comorbid problem, may be improved by stopping smoking, clearing bronchitis, and exercising regularly with walking. Because other atherosclerotic disease is usually present in patients with thoracic aortic aneurysms or atheroma, risk-reduction measures as outlined in other guidelines are appropriate (409). Additional medical management rationales are noted in Table 13.
Patients who are not candidates for operative intervention include those whose aneurysms or other aortic pathology do not meet the criteria for surgical intervention and those in whom the criteria are met but who are considered inoperable, most commonly because of coexisting disease. Patients with large aneurysms who are considered inoperable may benefit from stringent control of risk factors (see Section 3.2) to potentially slow the rate of growth and reduce the risk of rupture or dissection. Recommendations for periodic imaging are noted in Section 14.
18.104.22.168 Recommendations for Blood Pressure Control
1. Antihypertensive therapy should be administered to hypertensive patients with thoracic aortic diseases to achieve a goal of less than 140/90 mm Hg (patients without diabetes) or less than 130/80 mm Hg (patients with diabetes or chronic renal disease) to reduce the risk of stroke, myocardial infarction, heart failure, and cardiovascular death. (415–419) (Level of Evidence: B)
2. Beta adrenergic–blocking drugs should be administered to all patients with Marfan syndrome and aortic aneurysm to reduce the rate of aortic dilatation unless contraindicated. (88) (Level of Evidence: B)
1. For patients with thoracic aortic aneurysm, it is reasonable to reduce blood pressure with beta blockers and angiotensin-converting enzyme inhibitors (412) or angiotensin receptor blockers (89,413) to the lowest point patients can tolerate without adverse effects. (88,410,411) (Level of Evidence: B)
2. An angiotensin receptor blocker (losartan) is reasonable for patients with Marfan syndrome, to reduce the rate of aortic dilatation unless contraindicated. (89,90) (Level of Evidence: B)
Treatment of hypertension to reduce end points of MI, stroke, and death is well established with many randomized clinical trials (420). In the Jikei Heart Study, Japanese patients who received valsartan along with other antihypertensive therapy had a significantly lower rate of cardiovascular morbidity and mortality compared with patients treated without valsartan. Reductions noted in particular included lower incidence of stroke, transient ischemic attack (TIA), angina pectoris, and heart failure. Moreover, pertinent to this guideline, there was a significant reduction in the incidence of AoD in the valsartan-treated patients, which contributed to the reduction in overall cardiovascular morbidity and mortality (413).
Currently, beta adrenergic blockade serves as the foundation of the medical regimen because of demonstrated inhibition of aneurysm expansion in patients with Marfan syndrome. Shores and colleagues (88) randomized 70 patients with Marfan syndrome to propranolol or placebo in a open-label study demonstrating an attenuated rate of expansion over the 10-year follow-up. Dietz and colleagues (91) demonstrated that angiotensin receptor blocker therapy reduces aneurysm expansion in animal models of Marfan syndrome. This group has also recently demonstrated that angiotension receptor blocker therapy slowed the rate of progression of progressive aortic root dilatation in a preliminary study of 18 pediatric patients with Marfan syndrome (89). Both beta blockade and angiotensin II receptor blockade therapy are being further investigated in a randomized trial for patients with Marfan syndrome (90).
Lifestyle modifications of diet, weight reduction for overweight or obese patients, moderation of alcohol consumption, and aerobic exercise are standard approaches to treat hypertension (421), but pharmacological therapy is usually required for patients with thoracic aortic diseases.
22.214.171.124 Recommendation for Dyslipidemia
1. Treatment with a statin to achieve a target LDL cholesterol of less than 70 mg/dL is reasonable for patients with a coronary heart disease risk equivalent such as noncoronary atherosclerotic disease, atherosclerotic aortic aneurysm, and coexistent coronary heart disease at high risk for coronary ischemic events. (422–425) (Level of Evidence: A)
The National Cholesterol Education Program ATP III recommends that patients with noncoronary atherosclerosis be treated like patients with established coronary heart disease (426). Atherosclerosis in the aorta, like atherosclerosis in any noncoronary vascular bed, markedly increases the risk of MI and stroke. As a result of this high-risk status (greater than 20% event rate in 10 years), the goal for hypolipidemic therapy is an LDL level less than 100 mg/dL. Initial therapy in these patients should be a statin. After the National Cholesterol Education Program ATP III guidelines were released in 2001, the Heart Protection Study reported in 2002 that patients with atherosclerosis and a total cholesterol level greater than 135 mg/dL benefited from the addition of simvastatin 40 mg/d (427). The RR reductions remained even when LDL started at less than 100 mg/dL. In concert with data from patients with acute coronary syndromes, the more recent ACC/AHA Guidelines for the Management of Patients With Peripheral Arterial Disease also gave a Class IIa recommendation suggesting the use of a statin to achieve a target LDL of less than 70 mg/dL for patients at very high risk of ischemic events is reasonable (4).
There are experimental data demonstrating a delayed development of atherosclerosis and prevention of aneurysm development by statins (428–430). However, there are no clinical outcomes data that justify their use acutely or suggest that statins prevent expansion after thoracic aortic aneurysms have developed.
126.96.36.199 Recommendation for Smoking Cessation
1. Smoking cessation and avoidance of exposure to environmental tobacco smoke at work and home are recommended. Follow-up, referral to special programs, and/or pharmacotherapy (including nicotine replacement, buproprion, or varenicline) is useful, as is adopting a stepwise strategy aimed at smoking cessation (the 5 A's are Ask, Advise, Assess, Assist, and Arrange).(431–432b) (Level of Evidence: B)
There are no randomized or prospective trials that have investigated the effect of smoking cessation on thoracic aortic disease. Patients with thoracic aortic aneurysm who smoke have double the rate of aneurysm expansion (433). Aneurysm expansion and rupture after Type B dissection are not affected by cigarette smoking (434). Smoking cessation reduces the rate of MI and death in patients with noncoronary atherosclerosis (435). Patients who smoke require close follow-up in conjunction with medical and other support to achieve complete smoking cessation.
9.2.2 Surgical and Endovascular Treatment by Location of Disease
188.8.131.52 Ascending Aorta and Aortic Sinuses
184.108.40.206.1 Recommendations for Asymptomatic Patients With Ascending Aortic Aneurysm
1. Asymptomatic patients with degenerative thoracic aneurysm, chronic aortic dissection, intramural hematoma, penetrating atherosclerotic ulcer, mycotic aneurysm, or pseudoaneurysm, who are otherwise suitable candidates and for whom the ascending aorta or aortic sinus diameter is 5.5 cm or greater, should be evaluated for surgical repair. (371) (Level of Evidence: C)
2. Patients with Marfan syndrome or other genetically mediated disorders (vascular Ehlers-Danlos syndrome, Turner syndrome, bicuspid aortic valve, or familial thoracic aortic aneurysm and dissection) should undergo elective operation at smaller diameters (4.0 to 5.0 cm depending on the condition; seeSection 5) to avoid acute dissection or rupture. (81,114,143,371,436–439) (Level of Evidence: C)
3. Patients with a growth rate of more than 0.5 cm/y in an aorta that is less than 5.5 cm in diameter should be considered for operation. (Level of Evidence: C)
4. Patients undergoing aortic valve repair or replacement and who have an ascending aorta or aortic root of greater than 4.5 cm should be considered for concomitant repair of the aortic root or replacement of the ascending aorta. (Level of Evidence: C)
1. Elective aortic replacement is reasonable for patients with Marfan syndrome, other genetic diseases, or bicuspid aortic valves, when the ratio of maximal ascending or aortic root area (π r2) in cm2 divided by the patient's height in meters exceeds 10. (16,143) (Level of Evidence: C)
2. It is reasonable for patients with Loeys-Dietz syndrome or a confirmed TGFBR1 or TGFBR2 mutation to undergo aortic repair when the aortic diameter reaches 4.2 cm or greater by transesophageal echocardiogram (internal diameter) or 4.4 to 4.6 cm or greater by computed tomographic imaging and/or magnetic resonance imaging (external diameter). (78) (Level of Evidence: C)
Aortic diameter is a major criterion for recommending elective operation in asymptomatic patients with aneurysm of the thoracic and thoracoabdominal aorta. This assumes that the risk of operation is low (less than 5%). Currently, aortic diameter perpendicular to the axis of flow as measured by CT is the dimension most often used to determine the size of the enlarged aorta. This recommendation is based on the observation that the risk of an adverse event (rupture, dissection, death) exceeds the risk of elective operation when the maximum diameter exceeds 5.5 to 6.0 cm (374,436,437,440) (Figure 30). Formulas that incorporate height and aortic cross-sectional area have been developed to establish thresholds for operation in shorter patients but are less widely used (16,143).
The morphology and histopathology of thoracic aortic enlargements affect the natural history of aortic diseases, including the risk of rupture or dissection, and thus can influence the decision to intervene (Figures 31 and 32).⇓⇓ Fusiform aneurysms are most common and behave in a relatively predictable manner. Aortic dimension can thus be used as an indication for operation. Saccular aneurysms occur less frequently and may be associated with a greater risk of rupture. Many of these are actually pseudoaneurysms, which can develop after previous trauma or aortic surgery, or with PAUs, and which result in focal disruption or weakening of the layers of the aorta.
Patients with substantial dilatation of the aortic sinuses may develop asymptomatic aortic regurgitation as a result of loss of coaptation of the otherwise normal aortic valve cusps. Patients with associated bicuspid aortic valve disease may have asymptomatic stenosis or regurgitation of the valve. In these patients, the valvular disease may be an indication for operative intervention (5).
220.127.116.11.2 Recommendation for Symptomatic Patients With Thoracic Aortic Aneurysm
1. Patients with symptoms suggestive of expansion of a thoracic aneurysm should be evaluated for prompt surgical intervention unless life expectancy from comorbid conditions is limited or quality of life is substantially impaired. (Level of Evidence: C)
Symptoms associated with thoracic aneurysms usually develop later in the course of enlargement of the aorta and most commonly result from impingement of the aneurysm on adjacent structures. Aneurysms of the ascending aorta and aortic sinuses may result in symptoms related to the aortic regurgitation that develops as a result of the progressive aortic enlargement. Chest or back pain in the presence of an enlarged thoracic aorta is a predictor of aortic rupture (336,441). Patients who develop an acute Type A AoD commonly present with severe chest or back pain.
18.104.22.168.3 Endovascular Grafting for Ascending Aortic Aneurysm
At the time of this writing, endovascular stent grafts have not been approved by the US Food and Drug Administration for treatment of aneurysms or other conditions of the ascending aorta.
22.214.171.124.4 Recommendations for Open Surgery for Ascending Aortic Aneurysm
1. Separate valve and ascending aortic replacement are recommended in patients without significant aortic root dilatation, in elderly patients, or in young patients with minimal dilatation who have aortic valve disease. (Level of Evidence: C)
2. Patients with Marfan, Loeys-Dietz, and Ehlers-Danlos syndromes and other patients with dilatation of the aortic root and sinuses of Valsalva should undergo excision of the sinuses in combination with a modified David reimplantation operation if technically feasible or, if not, root replacement with valved graft conduit. (134,164,442–444) (Level of Evidence: B)
Ascending Aortic Aneurysms: The extent of aortic resection and the need for ancillary procedures are determined by preoperative testing and intraoperative findings. Ancillary procedures that may be performed concurrently include coronary artery bypass graft surgery, valve replacement or repair, repair of cardiac septal defects, closure of vascular fistulas, and ablative therapy for arrhythmias.
For patients with isolated aneurysms confined to the ascending aorta, resection and graft replacement is the most commonly performed and recommended procedure. Alternatively, reduction aortoplasty with or without external reinforcement has only been performed in very limited circumstances (445,446).
For patients with aortic valve stenosis who require valve replacement, the choice of valve substitute is determined by age of the patient, presence of comorbid disease, risk of complications related to anticoagulation and reoperation, and life expectancy (139).
For patients with aortic regurgitation associated with a bicuspid aortic valve, repair of the aortic valve with or without root remodeling or tailoring of the sinotubular junction is preferable if the valve is not severely fibrotic or calcified (99,140). For patients with a dilated aortic root, particularly those with stenotic bicuspid valves, composite valve grafts containing either mechanical or biological valves are implanted.
Ascending aneurysms larger than 4.5 to 5.0 cm require repair or tube graft replacement when aortic valve repair or replacement is the primary indication for operation (5). In elderly patients, ascending aortic aortoplasty when the aortic diameter does not exceed 5.0 cm may be an acceptable alternative.
Aortic Valve and Root: In patients with aortic valve regurgitation and root dilatation, aortic valve repair and root-sparing procedure may be the primary procedure. In patients with Marfan syndrome or with tricuspid aortic valve regurgitation, a modification of the David reimplantation operation may be considered. (94,95,97–99,447) Composite valve grafts with either biological or mechanical valves are an alternative option, particularly for valvular stenosis (99,140,441,448).
126.96.36.199 Recommendations for Aortic Arch Aneurysms
1. For thoracic aortic aneurysms also involving the proximal aortic arch, partial arch replacement together with ascending aorta repair using right subclavian/axillary artery inflow and hypothermic circulatory arrest is reasonable. (222,449,450) (Level of Evidence: B)
2. Replacement of the entire aortic arch is reasonable for acute dissection when the arch is aneurysmal or there is extensive aortic arch destruction and leakage. (222,450) (Level of Evidence: B)
3. Replacement of the entire aortic arch is reasonable for aneurysms of the entire arch, for chronic dissection when the arch is enlarged, and for distal arch aneurysms that also involve the proximal descending thoracic aorta, usually with the elephant trunk procedure. (451–453) (Level of Evidence: B)
4. For patients with low operative risk in whom an isolated degenerative or atherosclerotic aneurysm of the aortic arch is present, operative treatment is reasonable for asymptomatic patients when the diameter of the arch exceeds 5.5 cm. (374) (Level of Evidence: B)
5. For patients with isolated aortic arch aneurysms less than 4.0 cm in diameter, it is reasonable to reimage using computed tomographic imaging or magnetic resonance imaging, at 12-month intervals, to detect enlargement of the aneurysm. (Level of Evidence: C)
6. For patients with isolated aortic arch aneurysms 4.0 cm or greater in diameter, it is reasonable to reimage using computed tomographic imaging or magnetic resonance imaging, at 6-month intervals, to detect enlargement of the aneurysm. (Level of Evidence: C)
Aneurysms of the aortic arch are commonly associated with aneurysmal disease or dissection of the ascending aorta or the adjacent descending thoracic aorta, and the indications for operative intervention in these patients are those for the adjacent aortic segment. This relates to the need for hypothermic cardiopulmonary bypass and an interval of hypothermic circulatory arrest, and to higher operative mortality and stroke rates than those observed following operation for isolated aneurysms of the ascending or descending thoracic aorta (451–459). As with ascending aneurysms, a growth rate of more than 0.5 cm/y in the absence of symptoms could be considered an indication for operation.
Symptoms associated with aortic arch aneurysms such as hoarseness resulting from stretching of the left recurrent laryngeal nerve, dysphagia, dyspnea, and chest or back pain are indications for operative intervention for patients with arch aneurysms unless life expectancy is quite limited. Suitability for operative intervention involves similar risk assessment to that for aneurysm or other disorders of the ascending aorta and aortic root.
The innominate, left carotid, and left subclavian arteries may require separate grafting. For short periods of circulatory arrest, the use of retrograde or antegrade brain perfusion has not conclusively been shown to add further brain protection; however, use of the subclavian or axillary artery bypass with a side graft reduces the risk of stroke (449).
188.8.131.52.1 Open Surgery
At present, endovascular stent grafts have not been approved by the US Food and Drug Administration for treatment of aneurysms or other conditions of the aortic arch. For patients with large aneurysms who are not candidates for conventional open operation, experience is accumulating with operative procedures that involve translocation of the brachiocephalic arteries from the aortic arch using branch grafts from the proximal ascending aorta, and placement of an endovascular graft into the distal ascending aorta, the entire aortic arch, and a segment of the adjacent descending thoracic aorta (371,460,461).
Cardiopulmonary bypass with some degree of hypothermia is required for operations that require replacement of the aortic arch. Brain protection can be achieved by profound hypothermia alone, direct antegrade perfusion of 1 or more of the brachiocephalic arteries, or retrograde perfusion using cold oxygenated blood that is infused into the superior vena cava during the arrest period (211,449,462–467) (see Section 14.5.1). The aortic arch is replaced with a synthetic graft. The brachiocephalic arteries are attached to the graft using a patch of the aorta which contains the origins of the 3 vessels or separately using a graft that contains 3 branches. The proximal and distal ends of the aortic graft are attached to normal segments of ascending and descending thoracic aorta.
An “elephant trunk” procedure has been used to reconstruct the arch and then provide a Dacron graft landing zone for endovascular stent graft treatment of descending thoracic aortic aneurysms (Figure 33).
184.108.40.206 Descending Thoracic Aorta and Thoracoabdominal Aorta
220.127.116.11.1 Recommendations for Descending Thoracic Aorta and Thoracoabdominal Aortic Aneurysms
1. For patients with chronic dissection, particularly if associated with a connective tissue disorder, but without significant comorbid disease, and a descending thoracic aortic diameter exceeding 5.5 cm, open repair is recommended. (371,382,468) (Level of Evidence: B)
2. For patients with degenerative or traumatic aneurysms of the descending thoracic aorta exceeding 5.5 cm, saccular aneurysms, or postoperative pseudoaneurysms, endovascular stent grafting should be strongly considered when feasible. (371,469) (Level of Evidence: B)
3. For patients with thoracoabdominal aneurysms, in whom endovascular stent graft options are limited and surgical morbidity is elevated, elective surgery is recommended if the aortic diameter exceeds 6.0 cm, or less if a connective tissue disorder such as Marfan or Loeys-Dietz syndrome is present. (371) (Level of Evidence: C)
4. For patients with thoracoabdominal aneurysms and with end-organ ischemia or significant stenosis from atherosclerotic visceral artery disease, an additional revascularization procedure is recommended. (470) (Level of Evidence: B)
At the time of publication of this document, 3 endovascular stent grafts have been approved by the US Food and Drug Administration only for aneurysms involving the descending thoracic aorta. Although the feasibility and safety of endovascular stent grafting of the descending aorta have been demonstrated for other pathologic conditions including acute and chronic Type B AoD, IMH, PAU, acute traumatic aortic transsection, and pseudoaneurysms, these conditions are currently considered “off label.”
There are no published randomized trials that compare the outcomes of endovascular stent grafting with conventional open operation or nonoperative management. Thus, recommendations for use are based principally on observational studies and nonrandomized comparisons of cohorts of patients.
18.104.22.168.2 Endovascular Versus Open Surgical Approach
The potential advantages of endovascular grafting over open operation include the absence of a thoracotomy incision and the need for partial or total extracorporeal circulatory support and clamping of the aorta, as well as lower hospital morbidity rates and shorter length of hospital stay.
Endovascular grafting may be of particular value in patients with significant comorbid conditions (older age, substantial cardiac, pulmonary and renal dysfunction) who would be considered poor or noncandidates for open surgery. Patients who are not considered candidates for open surgery but who have undergone endovascular grafting have substantially poorer long-term outcomes than patients who are reasonable candidates for open operation and are treated with endografts (471). Furthermore, intervention (endovascular stent graft or open surgical graft replacement) for a descending aneurysm has real risks of mortality and morbidity, including the risk of spinal cord ischemic injury. All physicians should work collaboratively among specialities during the initial decision-making steps to determine via consensus whether a particular patient's pathology, risk factors, and projected natural history if treated medically justify an intervention on the descending thoracic aorta, either a stent graft or an open procedure.
There are no data that conclusively demonstrate that the prevalence of spinal cord ischemic injury (lower extremity paralysis or paresis) is less for endovascular approaches than for open surgical repair. Similarly, there are no firm data to indicate that overall costs of medical care are lower with endovascular procedures. Although the costs of the initial hospitalization may be lower because of reduced operative time and a shorter length of stay, these benefits may be negated by the cost of the devices, the need for subsequent interventions, and the cost and dissatisfaction of repeated imaging studies, which are necessary in the postoperative period (472,473).
Some patients are not suitable candidates for endovascular grafting procedures. Absence of suitable “landing zones” above and below the aneurysm (usually 2 to 3 cm of normal diameter aorta without circumferential thrombus) is a contraindication. A width of the aorta at the landing zones that exceeds the recommended width for the largest available endovascular grafts (generally 10% to 15% larger than the width of the aorta) is also a contraindication.
Lack of vascular access sites to insert the relatively large-bore sheaths that are necessary for deployment of the grafts is also a contraindication. Severe atherosclerosis and intraluminal thrombus of the aorta may increase the risk of embolic stroke during manipulation of guidewires and catheters and represents a relative contraindication (474).
22.214.171.124.3 End-Organ Preservation During Thoracic Endograft Implantation
Because aneurysmal disease can involve any portion of the aorta, organ preservation during repair of either aneurysmal disease or dissection is an important part of the implant procedure. Aneurysms involving the aortic arch pose a significant risk to cerebral and upper extremity blood flow with endovascular repair. The need to cover either the left common carotid or the innominate artery to treat arch aneurysmal disease is infrequent. Intentional coverage of the left subclavian artery is more common, occurring in approximately 50% of thoracic endograft implants. Most patients tolerate coverage of the left subclavian artery without upper extremity ischemia (475–478), but recently several groups have suggested that these patients may be at higher risk of perioperative stroke and spinal cord ischemic injury (479–481). In the Talent VALOR trial, the need for intentional preimplant left subclavian artery bypass was only 5.2% (482). To minimize the risk associated with intentional coverage of the left subclavian artery, it is recommended that the patency of the contralateral right subclavian and vertebral arteries be determined preoperatively by CT, MR, or invasive angiography. Additionally, verification that the vertebral arteries communicate at the basilar artery by either transcranial Doppler or angiography is also recommended. If these steps are taken to ensure that the contralateral posterior circulation is intact, the need to perform a left subclavian artery bypass postoperatively to prevent symptomatic arm claudication or vertebral basilar insufficiency is infrequent.
Spinal cord ischemia leading to paralysis is one of the serious potential complications of the thoracic endograft procedure. Intercostal arteries are intentionally covered by the endograft. There is evidence that the risk of spinal cord ischemia may be greater when treatment involves coverage of most of the descending thoracic aorta (e.g., from left subclavian to celiac artery) (483). Additionally, patients who have had previous repair of abdominal aneurysm (either open or endovascular) are at increased risk for spinal cord ischemia and paraplegia (484). In 1 study, the risk of spinal cord ischemia was 10% to 12% in patients with previous abdominal aneurysm repairs and 2% in patients who did not have previous abdominal aneurysm repairs (485). In circumstances where it is necessary to cover most of the descending thoracic aorta or if the patient has had a previous abdominal aneurysm repair, cerebrospinal fluid (CSF) pressure monitoring and drainage are an important strategy to minimize the risk of paraplegia (486).
Treatment of TAAs with an endograft may require intentional coverage of the celiac and/or superior mesenteric arteries to achieve a seal at the distal attachment site. In these instances, a superior mesenteric-to-celiac artery bypass graft or an aorta-to-superior mesenteric and/or celiac artery bypass graft has been used as a first-stage debranching procedure before proceeding with the endograft implant (Figure 34). However, risk of the debranching operation may be no lower than conventional open repair, and therefore these operative approaches are performed selectively (487). Fenestrated grafts are in development and are undergoing clinical trials. Roselli et al (488) published initial results in the first 73 patients undergoing in situ endovascular repair of extensive TAAs using a branch endograft with promising results in a high-risk population. With careful attention to preservation of blood flow to the mesenteric vessels, the incidence of mesenteric ischemia with endograft implants is approximately 3% (482). Most instances of mesenteric ischemia are the result of emboli rather than malperfusion owing to coverage or occlusion of a mesenteric vessel.
AoDs pose a complex situation for intestinal, renal, or lower extremity perfusion, as the branches of the aorta in the abdominal cavity may be perfused from either the true or false lumen. Often, both the true and false lumens are patent and some of the visceral, renal, or lower extremity vessels are fed by one channel and the remainder by the other. Thus, the operator must consider how blood flow reaches vital organs in the abdominal cavity before considering treatment of an AoD with an endograft. In most cases, where descending thoracic dissections are treated with an endograft, the important treatment principle is to cover the proximal entry tear obliterating the false lumen. However, if the false lumen supplies blood flow to the visceral vessels, blood flow to these organs may be compromised after endograft implantation. Stenting of the vessels at risk from the true lumen or establishing flow communication between true and false lumen more distally with a fenestration procedure may prevent such compromise. In cases of acute Type B AoD treated with endografts, coverage of the proximal entry site into the false lumen may result in healing of the tear downstream with restoration of blood flow from the true lumen without the need for adjunctive fenestration procedures (490).
126.96.36.199.4 Periprocedural Complications of Endograft Procedures
The acute complications of thoracic endograft procedures are related to problems with access and with device implantation. Vascular access is a substantial issue with thoracic endograft procedures. The sheath required to deploy endovascular grafts is size 20 French or greater and can be up to size 25 French. Many patients have femoral arteries that are too small to accommodate such large sheaths. Use of the iliac artery or the aorta for access is required in approximately 15% of the patients. Infection at the access site is not an infrequent complication. Bleeding complications are less frequent. Average blood loss in the Talent VALOR trial was 371 mL, and the need for transfusion in the STARZ TX2 clinical trial (study results on the Zenith TX2 Endovascular Graft for the Thoracic Aortic Aneurysms) was 3% (469,482). Thromboemboli to mesenteric, renal, or lower extremity vessels can also occur (491).
Major adverse events related to the device occur in 10% to 12% of patients in the initial 30-day perioperative period. Stroke was observed in 2.5% of the patients in the STARZ TX2 clinical trial (469) and 3.6% in the Talent VALOR trial (482). It has been reported to be as high as 8% in other studies (474,492). Most strokes occur as a result of atheroemboli from the transverse aortic arch with manipulation of the guidewires and device in the arch vessels (474). Paraplegia varied from 1.3% in the STARZ TX2 clinical trial to 1.5% in the Talent VALOR trial to 3% in the TAG trial (469,482,493).
Cardiac complications, principally MI, occur rarely (2% to 4%) (469,482). Cardiac tamponade or rupture is rare and may be caused by either the stiff guidewires that are required to deliver the devices or excessive afterload created by the balloons that are inflated to produce a seal of the graft to the aortic wall. Ventricular tachycardia or ventricular fibrillation has been reported (469). Pulmonary complications include postoperative pneumonias, which occur in less than 5% of patients. Acute renal failure requiring dialysis is also uncommon, reported to occur in 1.3% of patients in the Talent VALOR trial (482). Device implant infection is exceedingly rare and has been observed mainly in situations where the device was implanted in an infected medium emergently (i.e., mycotic aneurysms or aortoenteric fistulae) (482).
Endoleak is defined as the persistence of blood flow outside the lumen of the endoluminal graft but within the aneurysm sac, based on imaging. Endoleaks are classified based on the source of blood flow (Table 14).
Perforation or dissection of the aorta at the implantation site is infrequent but usually reported with stent systems that have uncovered or bare proximal attachment stents (495–499). Involution or infolding of the endografts may occur and has been most frequently reported with grafts applied to a relatively small aorta where oversizing can be substantial, especially in trauma cases (366,500,501). Infoldings can also occur when there is inadequate conformity of the device to the aortic wall in a region of marked curvature or “beaking” (see Figure 27). This leads to collapse of the endograft, gross endoleak, and the potential for acute occlusion of flow to the descending thoracic aorta. It can be prevented (or managed after it occurs) with implantation of a reinforcing stainless steel stent at the proximal leading edge of the implant (500).
Late complications of thoracic aortic endografting include endoleaks, continued aneurysm growth, metal fatigue and stent fracture and kinking, detachment, migration, perforation, and infection of the implanted device. Endoleaks occur in 10% to 20% of patients (482). The frequency is greatest in the first month following implantation and declines over the ensuing 5 years (502). The most frequent type of endoleak is Type IA, or proximal attachment site failure (see Section 8.6) (469,482,503–505). Migration occurs infrequently—at 1-year follow-up in 0.4% of cases in the TAG clinical trial and 3.9% of cases in the Talent VALOR trial (469,482,506).
Late perforations of the aorta by the endograft can occur and are common with the use of oversized grafts and/or grafts with bare or uncovered stents (507,508). Most late device complications such as endoleak and migration are treatable with implantation of graft extensions. The overall need for repeat interventions to maintain endograft integrity is 6% to 7%, most commonly in the form of implantation of an extension cuff. Conversions to open operation are rarely needed (1% to 2%) (476,480,505,509,510). Late complications (9 to 24 months) include stent fractures, which are often asymptomatic (511). There are case reports of fractures leading to endograft dysfunction, endoleak, migration, and/or embolization (512). Endograft infection, usually as a result of an ongoing systemic infection, or as a result of infected structures adjacent to the graft (513,514), is uncommon, but when it occurs, it is very difficult to eradicate without explantation of the device and, without explantation, can lead to aortic rupture (515).
The experience with the use of endografts for the treatment of acute descending AoD is limited. A meta-analysis of 609 published cases suggests that when endografts are used, mortality is slightly higher (5.3%) and late conversions to open procedures are more frequent (2.5%) than data reported for treatment of aneurysms. The risk of major complications including stroke, paraplegia, and aneurysm rupture appears to be similar in both conditions (516). A complication unique to endograft treatment of acute Type B dissections is conversion of the dissection to a retrograde Type A dissection, creating a surgical emergency (480,517). Until and unless this life-threatening complication is understood and eliminated, endograft treatment of acute Type B dissections should be undertaken at institutions with a team qualified to perform open aortic surgery.
In the absence of Level A or Level B evidence regarding the relative efficacy of open and endovascular procedures for treatment of descending aortic aneurysms, no firm recommendations can be made regarding the optimal method of treatment. Among comparable patients treated with either open or endovascular procedures, the midterm results can be anticipated to be equivalent. The early mortality advantage of endovascular procedures may be lost during follow-up such as that seen with endovascular treatment of AAAs (518,519). The long-term durability of endovascular stent grafts is uncertain; currently, available grafts may have a durability of no more than 10 years. Younger patient age may be a relative contraindication to endografting.
There are no data to indicate that endovascular stent grafting should be performed in patients with asymptomatic descending aortic aneurysms that are less than 5.5 cm in diameter, because the risk of operation (approximately 5%) exceeds the risk of rupture or dissection (371) (Table 15). Undoubtedly, as new iterations of devices are introduced, these guidelines will change.
188.8.131.52.5 Open Surgical
Diseases of the aorta that require extensive thoracoabdominal surgical or interventional approaches fall into 3 large groups: 1) degenerative aneurysms, 2) AoD resulting in subsequent aneurysms or visceral ischemia, and 3) occlusive disease of the visceral arteries. Selection of patients for repair is based on symptoms, risk of death from rupture, and end-organ ischemia, provided associated comorbidity does not prevent surgical repair. In patients with lower chest or upper abdominal pain, CT or MR is performed to determine if the patient has a contained rupture, leak, or an aneurysm impinging on surrounding structures that may be causing pain. The perioperative risk of death is approximately 80% with emergency surgery; a few patients will recover without a major complication limiting quality of life, and few will be long-term survivors because of comorbid disease. In patients with pain from compression of surrounding structures, if comorbid disease is not a factor, results are considerably better with a 10% to 20% risk of death. In patients undergoing elective surgery, the risk of death is 3% to 10%, depending on the extent of repair.
The extent of repair for TAA is classified by the Crawford types: Type I is a repair that extends from the proximal descending aorta above T6 to the renal arteries; Type II, the highest risk group, extends from the proximal descending aorta above T6 to below the renal arteries; Type III extends from the distal descending aorta below T6 to below the diaphragm for variable extents; and Type IV extends from the diaphragm and involves most of the abdominal aorta. This classification correlates with the risk of paralysis, renal failure, and death.
Open surgical repair of TAA improved over the past 15 years, particularly on preventing postoperative paralysis. Historically, one of the most serious complications was paralysis, either paraparesis or paraplegia, caused by spinal cord injury whose risk is for Type I TAA repair, 15%; for Type II, 30%; Type III, 7%; and for Type IV, 4% (381). Current risk of paralysis is 3% to 5%, depending somewhat on the extent of repair (381,382,520–526).
The risk of renal failure may be reduced by preoperative day admission, fluid hydration starting the day before surgery, and the addition of bicarbonate to the hydration regimen and hypothermia protection during the operative procedure.
184.108.40.206.6 End-Organ Preservation During Open Thoracoabdominal Repairs
For thoracoabdominal aortic repairs done in combination with proximal repairs of the aortic arch, the key vital end organs to protect are the heart and brain. Flooding the field with CO2 appears to be beneficial (527,528). When the aortic arch requires repair, hypothermic circulatory arrest is usually required. Protection of the brain involves ensuring that calcium plaques or atheromata are not disturbed to prevent brain embolization. The temperature and where it is best measured, at which circulatory arrest is commenced, are debated but most large series have recommended circulatory arrest at a temperature below 20°C (211).
Perfusion of the celiac artery or superior mesenteric artery does not appear to be needed to protect the abdominal organs in most patients where moderate hypothermia (30° to 32°C) is used. Protection of the left lung during left thoracotomy repairs is important to avoid lacerations and bleeding into the parenchyma. Deflation of the left lung may be of benefit. However, disruption of the visceral pleura with attendant complications of bleeding and air leak may be unavoidable, particularly if adhesions are present due to prior surgery or inflammatory changes.
220.127.116.11.7 Aortic Dissection With Malperfusion
Renal, mesenteric, lower extremity, or cord malperfusion accompanies up to one third of acute AoD and roughly doubles mortality (248,529,530). In the case of Type A AoD with malperfusion, there is controversy over whether patient outcomes are improved by first repairing the aorta and then treating persistent malperfusion (531) or by first correcting the malperfusion and then repairing the aorta (247,532). The general consensus is to first repair the aorta, because this will correct malperfusion in most patients. In the case of Type B AoD, there is a general consensus that medical management should be supplemented by open or endovascular intervention when there is evidence for renal, mesenteric, lower extremity, or cord malperfusion (248,490,533).
10 Special Considerations in Pregnant Patients With Aortic Disease
10.1 Effects of Pregnancy on the Aorta
Physiologic effects of pregnancy may have a profound effect not only on aortic stress but potentially on arterial ultra structure as well. The pregnant state is characterized by increases in maternal blood volume, heart rate, blood pressure, stroke volume, and cardiac output (534,535). Taken together, the combined effects lead to greater arterial wall tension as well as intimal shear forces. These changes begin in the first and second trimesters but are most notable in the third trimester and peripartum period. Whether arterial wall weakening itself occurs during pregnancy remains controversial. Arterial dissection and/or rupture may occur with the highest incidence in the third trimester (approximately 50%) and peripartum period (33%). In one of the few prospective studies of pregnant patients with Marfan syndrome, 4.4% of carefully monitored patients developed aortic dissection (534). In unmonitored patients, the risk is likely higher.
10.2 Epidemiology of Chronic and Acute Aortic Conditions in Pregnancy
Marfan syndrome, Ehlers-Danlos syndrome, and other non-Marfan aortic disease may manifest during pregnancy. Although clearly rare, it has been estimated that half of AoD and/or ruptures in women younger than 40 years of age have been associated with pregnancy (536). Most dissections occur in the ascending aorta, although dissection or rupture of virtually any artery in the body has been described. In addition, pregnancy-related expansion of the sinotubular junction may lead to aortic valve insufficiency.
10.3 Counseling and Management of Chronic Aortic Diseases in Pregnancy
10.3.1 Recommendations for Counseling and Management of Chronic Aortic Diseases in Pregnancy
1. Women with Marfan syndrome and aortic dilatation, as well as patients without Marfan syndrome who have known aortic disease, should be counseled about the risk of aortic dissection as well as the heritable nature of the disease prior to pregnancy. (74,91) (Level of Evidence: C)
2. For pregnant women with known thoracic aortic dilatation or a familial or genetic predisposition for aortic dissection, strict blood pressure control, specifically to prevent Stage II hypertension, is recommended. (Level of Evidence: C)
3. For all pregnant women with known aortic root or ascending aortic dilatation, monthly or bimonthly echocardiographic measurements of the ascending aortic dimensions are recommended to detect aortic expansion until birth. (Level of Evidence: C)
4. For imaging of pregnant women with aortic arch, descending, or abdominal aortic dilatation, magnetic resonance imaging (without gadolinium) is recommended over computed tomographic imaging to avoid exposing both the mother and fetus to ionizing radiation. Transesophageal echocardiogram is an option for imaging of the thoracic aorta. (Level of Evidence: C)
5. Pregnant women with aortic aneurysms should be delivered where cardiothoracic surgery is available. (Level of Evidence: C)
1. Fetal delivery via cesarean section is reasonable for patients with significant aortic enlargement, dissection, or severe aortic valve regurgitation. (91) (Level of Evidence: C)
1. If progressive aortic dilatation and/or advancing aortic valve regurgitation are documented, prophylactic surgery may be considered. (537) (Level of Evidence: C)
In this regard, risk of major aortic complications during pregnancy appears to be low when the aortic root diameter is less than 4.0 cm (538). Such individuals may have one or more successful pregnancies. For patients with an aortic diameter greater than 4.0 cm and Marfan syndrome, half will have come to prophylactic surgery during pregnancy, will have a rupture, or will have life-threatening growth. Optimal preventive therapy in the pregnant woman with known aortic disease includes use of beta blockers to control heart rate and reduce shear stresses, particularly in the third trimester and peripartum period. Both angiotensin-converting enzyme inhibitors and angiotensin receptor blockers are contraindicated during pregnancy.
10.4 Evaluation and Management of Acute Aortic Syndromes During Pregnancy
Pregnant women with Marfan syndrome, bicuspid aortic valve and ascending aneurysms, Ehlers-Danlos syndrome, and non-Marfan familial thoracic aortic aneurysm and dissection may present with acute aortic syndromes at any point during the pregnancy but are particularly prone to do so in the last trimester, during delivery, or in the early postpartum period. Such women may have no knowledge of their underlying aortic condition until presentation with their acute aortic problem.
Obviously, acute AoD poses a huge risk for both the mother and the unborn child. Optimal treatment parallels that of nonpregnant women but with the added complication of when and how to deliver the child. Optimal care includes involvement with a high-risk maternal-fetal team along with an aortic specialty team capable of medical, percutaneous, and surgical aortic treatment. For Type A AoD occurring during the first or second trimester, urgent surgical repair with aggressive fetal monitoring is preferred. Fetal loss during hypothermia and cardiopulmonary bypass is common. When dissection complicates the third trimester, urgent cesarean section followed by aortic repair appears to offer the best chance for survival for the unborn child and the mother. For acute arch or Type B AoD, medical therapy is preferred unless percutaneous stent grafting or open surgery is mandated by malperfusion, aortic rupture, or subacute aortic leaking (539).
11 Aortic Arch and Thoracic Aortic Atheroma and Atheroembolic Disease
11.1 Recommendations for Aortic Arch and Thoracic Aortic Atheroma and Atheroembolic Disease
1. Treatment with a statin is a reasonable option for patients with aortic arch atheroma to reduce the risk of stroke. (540) (Level of Evidence: C)
1. Oral anticoagulation therapy with warfarin (INR 2.0 to 3.0) or antiplatelet therapy may be considered in stroke patients with aortic arch atheroma 4.0 mm or greater to prevent recurrent stroke. (Level of Evidence: C)
11.2 Clinical Description
Aortic arch atheroma is a risk factor for ischemic stroke based on autopsy (541,542), TEE (543–548), and intraoperative ultrasonographic studies (549) (Figure 35). In particular, plaques 4 mm or greater in thickness proximal to the origin of the left subclavian artery are associated with stroke and constitute one third of patients with otherwise unexplained stroke (542). These patients, even on antiplatelet therapy, carry a risk of recurrent ischemic stroke as high as 11% at 1 year, and the risk of a new vascular event (ischemic stroke, MI, peripheral event, and vascular death) is 20%, 36%, and 50% at 1, 2, and 3 years, respectively (550). The RR of new ischemic stroke was 3.8 (95% CI 1.8 to 7.8, P≤0.002), and that of new vascular events was 3.5 (95% CI 2.1 to 5.9, P≤0.001), independent of carotid stenosis, atrial fibrillation, peripheral artery disease, or other risk factors (550). Other studies showed that aortic arch plaques were independent predictors of recurrent strokes, MI and vascular death (551–553). Patients with noncalcified plaques were at higher risk for recurrent vascular events (554).
Regarding the natural history of aortic arch atheroma, Sen et al (555) noted progression in 29% and regression (defined as an increase or decrease in plaque thickness by 1 grade or greater, respectively) in 9%. Montgomery et al (556) reported 30 patients with moderate-to-severe aortic plaque noted on initial bi/multiplanar TEE as part of a workup for cardiac or an embolic event. Over a mean of 1 year, progression was reported in 23% and regression in 10%. Pistavos et al (557) used monoplanar TEE in 16 patients with familial hypercholesterolemia taking pravastatin to show progression in 19% and regression in 38% over 2 years. Geraci and Weinberger (558), using supraclavicular B-mode ultrasonography of the proximal aortic arch in 89 patients evaluated for transient neurologic symptom, noted a progression rate of 19% and a regression rate of 18% over a mean of 7.7 months (range 3 to 18 months). Sen et al (559) confirmed that in patients with stroke/TIA, aortic arch atheroma progression over 12 months is associated with more vascular events.
11.3 Risk Factors
Risk factors for the development of aortic atheroma include age, sex, heredity, hypertension, diabetes mellitus, hyperlipidemia, sedentary lifestyle, smoking, and endothelial dysfunction. Other factors include elevated levels of inflammatory markers (i.e., serum C-reactive protein), homocysteine, or lipoprotein (560,561). Risk factors for embolic complications include inflammation, shear forces of hypertension, plaque hemorrhage aneurysm formation, and iatrogenic manipulation. The likelihood of embolization is also increased with complex aortic plaque, defined as plaque that contains mobile thrombi or ulcerations or is 4 mm or greater in thickness (562). Plaques with a larger lipid core, a predominance of macrophages, a thin fibrous cap, and a lack of calcification are more “vulnerable” to disruption or rupture. Calcified plaques are more stable and less likely to result in embolic syndromes (562,564–565).
Methods of imaging the aortic arch to detect and/or measure plaque include:
Transesophageal Echocardiography. TEE can provide information of plaque mobility, ulceration, and composition (566), as well as details on the anatomic relationship of the plaque to the origin of the great vessels (69) with excellent interobserver and intraobserver reliability (567). Limitations of TEE in patients with stroke include the need for conscious sedation, patient cooperation for swallowing the probe, and risk of structural damage (566). A small portion of the ascending aorta is masked by the tracheal air column near the origin of the innominate artery, with an estimated 2% of plaques being missed (568). Multiplanar probes may reduce tracheal shadowing seen with monoplanar and biplanar probes (569).
Transthoracic Echocardiography. TTE can usually image the aortic root and proximal ascending aorta but cannot adequately assess aortic arch plaque (570,571).
Epiaortic Ultrasonography. Epiaortic imaging is useful to detect aortic arch plaque in the operative setting when the transducer may be placed directly over the aortic arch. The information may be used to select operative techniques such as off-pump coronary artery bypass grafting to avoid cannulation or cross-clamping of the aorta and reduce risk of perioperative strokes (572,573).
Contrast Aortography. The risk of invasive angiography and the need for contrast injection and radiation make contrast aortography less useful to assess aortic arch plaque in patients with stroke (574).
Magnetic Resonance Imaging. MR has been validated against TEE for detection and measurement of aortic arch plaque (80% overall agreement) (568,575). It has limited use in patients who are obese, have metallic implants, or are claustrophobic. Contrast MR angiography underestimates the plaque thickness but can identify morphologic features including calcification, fibrocellular tissue, lipid, thrombus, and features used to detect plaque stability and may be used to monitor aortic arch plaque progression and regression (572).
Computed Tomography. CT can reliably detect and measure protruding aortic arch plaques (568,576) and is the test of choice for detecting vascular calcification. Nonenhanced dual-helical CT may underestimate the amount of noncalcified plaque and mobile thrombus that presumably is at high risk for embolization (568). In conjunction with positron emission tomography it can be used to localize fluorodeoxyglucose uptake by the plaque, identifying active plaques and unstable plaques (577), but its clinical utility has yet to be established.
There is no definitive therapeutic regimen for this high-risk patient group because no randomized trial has been completed.
11.5.1 Anticoagulation Versus Antiplatelet Therapy
Mobile aortic atheroma have been noted to disappear during anticoagulant therapy (546) or with the use of a thrombolytic agent (578). However, there is concern about the use of warfarin in patients with aortic atheroma because of the theoretical risk of plaque hemorrhage resulting in atheroemboli syndrome (i.e., blue toes, renal failure, intestinal infarction) (579). Anticoagulation has been associated with worsening (580,581), as well as improvement of an aortic thrombus on anticoagulation in a patient with the atheroemboli syndrome (582). Cholesterol emboli have been documented on skin, muscle, and renal biopsy samples in patients with aortic arch atheroma seen on TEE (553,583). However, the risk of clinical atheroemboli syndrome during warfarin therapy in such patients appears to be low (only 1 episode in 134 patients according to the SPAF [Stroke Prevention in Atrial Fibrillation] trial) (562).
Three reports shed light on the potential benefit of warfarin in patients with aortic arch atheroma. The first described 31 patients with mobile lesions in the aorta on TEE (585) where a higher incidence of vascular events was seen in patients who were not treated with warfarin compared with those treated with warfarin (at the discretion of the referring physicians) (45% versus 5%). In the SPAF randomized trial of patients with “high-risk” nonvalvular atrial fibrillation, the risk of stroke at 1 year in 134 patients with complex aortic plaque was found to be reduced from 15.8% (11 events) in those treated with fixed low-dose warfarin plus aspirin (INR 1.2 to 1.5) to only 4% (3 events) in those treated with adjusted-dose warfarin (INR 2.0 to 3.0), a 75% RR reduction (P=0.02) for patients with atheromas who received “therapeutic range” anticoagulation (562). A third observational study reported on 129 patients with aortic arch atheroma on TEE performed to look for a source of cerebral or peripheral embolization (586). Treatment with oral anticoagulation, aspirin, or ticlopidine was not randomly assigned. There was a significant reduction in the number of embolic events in patients with plaques greater than or equal to 4 mm who received oral anticoagulants (0 events in 27 patients versus 5 events in 23 patients treated with antiplatelet agents) (OR 0.06, 95% CI 0.003 to 1.2, P=0.016). For patients with mobile lesions, there was a significant reduction in mortality while on anticoagulants, although the trend toward fewer embolic events did not reach statistical significance in this group.
These 3 reports suggest that warfarin is not harmful in patients with aortic arch atheroma and may reduce stroke rates. However, these studies are not randomized trials of treatment for patients with atheromas, and the numbers are relatively small. The current ARCH (Aortic Arch Related Cerebral Hazard) trial is an open-label trial where patients with aortic arch atheroma (4 mm or greater) and nondisabling stroke are being assigned to oral anticoagulation (target INR 2.0 to 3.0) versus aspirin (75 mg/d) plus clopidogrel (75 mg/d) and followed longitudinally for recurrence of vascular events.
11.5.2 Lipid-Lowering Agent
There are no randomized trials to support specific lipid-lowering drug therapy for a patient with stroke caused by atheroembolism. However, 2 randomized studies of low-dose and higher-dose statin in patients with aortic and/or carotid plaques showed significant regression in plaque seen on MR, which in 1 study was related to LDL cholesterol level but not statin dosage (587) and, in the other study, was related to both LDL lowering and the statin dosage (588). It seems likely that statin therapy decreases the risk of stroke. Mechanisms for this effect may involve pleiotropic effects of statins, including plaque regression, plaque stabilization, decreased inflammation, and inhibitory effects on the coagulation cascade at different levels.
No randomized trial on the use of statins in patients with severe aortic plaque has been published. However, in an observational study of 519 patients with severe aortic plaque on TEE (540), statin use was associated with an RR reduction for ischemic stroke of 59%. Statins have reduced both primary and secondary stroke in a variety of patient populations (589,590). Hence, a majority of patients with stroke and TIA with identified aortic plaque already have other stronger indications for statin therapy (409). Recommendations for statins are noted for other manifestations of atherosclerotic diseases (409) (see Section 18.104.22.168). Clinical trials are needed to determine the effects of statins in patients with severe aortic atheroma and risk of atheroembolism.
11.5.3 Surgical and Interventional Approaches
Aortic arch endarterectomy has been attempted for patients with thromboembolism originating from aortic arch atheroma. Although successful in a handful of case reports, this procedure resulted in a relatively high rate (34.9% with endarterectomy versus 11.6% without endarterectomy) of perioperative stroke and mortality when it was performed to limit stroke during cardiac surgical procedures requiring cardiopulmonary bypass (coronary bypass surgery and valve surgery) (545). Covered stents offer the potential advantage of shielding severely diseased aortic segments to prevent further embolization. However, periprocedural embolization may occur during diagnosis or interventional endovascular manipulations. There is insufficient evidence to recommend prophylactic endarterectomy or aortic arch stenting for purposes of stroke prevention.
12 Porcelain Aorta
Vascular calcification occurs in the media and represents a central component of atherogenesis, typically signaling longstanding inflammation. The amount of calcification directly associates with the extent of atherosclerosis, and the presence of aortic calcium predicts the presence of coronary heart disease (443).
With severe atherosclerosis of the aorta, calcification may be severe and diffuse, causing an eggshell appearance seen on chest x-ray or CT (444). Also termed “porcelain aorta,” this finding is usually noted during operation for coronary heart or valvular heart disease at the time of surgery. The calcification interferes significantly with cannulation of the aorta, cross-clamping, and placement of coronary bypass grafts, increasing the risk of stroke and distal embolism significantly (Figure 36).
In these patients, direct manipulation of the aorta may result in an unrepairable aortic injury and/or distal embolization. Surgeons have used several techniques to reduce adverse neurologic events in these patients: internal aortic balloon occlusion (as apposed to aortic cross-clamping), a “no-touch” technique to avoid the ascending aorta, alternate locations for cannulation or coronary bypass graft anastomoses, replacement of the ascending aorta, and intra-aortic filtration of atherosclerotic debris (591–599).
13 Tumors of the Thoracic Aorta
Neoplasms of the thoracic aorta are usually secondary and related to contiguous spread of adjacent primary malignancies, particularly lung and adjacent primary malignancies, or subsequent metastases, particularly lung and esophagus (600–603).
Primary neoplasms of the thoracic aorta are rare. A review of the literature between 1873 and 2002 collated a total of 53 thoracic and 10 thoracoabdominal tumors with most lesions protruding into the aortic lumen (604) (Table 16).
Metastatic disease is often demonstrated at the time of diagnosis of primary aortic neoplasms, so that constitutional symptoms of malaise, fatigue, weight loss, and nausea may be the presenting complaints. Other presentations can include distal arterial embolization (with histopathologic examination showing neoplasm or identified by imaging techniques during a search for an embolic source) (604–606). AoD may originate in the area of the neoplasm or the aortic occlusion (607). Resection and reconstruction of the segment of aorta containing the neoplasm have been described, but because most patients present with metastatic disease, overall prognosis is poor (608).
14 Perioperative Care for Open Surgical and Endovascular Thoracic Aortic Repairs
14.1 Recommendations for Preoperative Evaluation
1. In preparation for surgery, imaging studies adequate to establish the extent of disease and the potential limits of the planned procedure are recommended. (Level of Evidence: C)
2. Patients with thoracic aortic disease requiring a surgical or catheter-based intervention who have symptoms or other findings of myocardial ischemia should undergo additional studies to determine the presence of significant coronary artery disease. (Level of Evidence: C)
3. Patients with unstable coronary syndromes and significant coronary artery disease should undergo revascularization prior to or at the time of thoracic aortic surgery or endovascular intervention with percutaneous coronary intervention or concomitant coronary artery bypass graft surgery. (Level of Evidence: C)
1. Additional testing is reasonable to quantitate the patient's comorbid states and develop a risk profile. These may include pulmonary function tests, cardiac catheterization, aortography, 24-hour Holter monitoring, noninvasive carotid artery screening, brain imaging, echocardiography, and neurocognitive testing. (Level of Evidence: C)
2. For patients who are to undergo surgery for ascending or arch aortic disease, and who have clinically stable, but significant (flow limiting), coronary artery disease, it is reasonable to perform concomitant coronary artery bypass graft surgery. (Level of Evidence: C)
1. For patients who are to undergo surgery or endovascular intervention for descending thoracic aortic disease, and who have clinically stable, but significant (flow limiting), coronary artery disease, the benefits of coronary revascularization are not well established. (609–611) (Level of Evidence: B)
Preoperative evaluation usually includes imaging studies necessary to establish the extent of disease, the limits of the planned procedure, and the clinical risks attendant to the procedure. When the writing committee was polled regarding the extent of usual preoperative laboratory testing, a variety of approaches emerged. In some centers, extensive testing includes pulmonary function tests (particularly for smokers and those with Marfan syndrome), Holter monitoring, and carotid duplex scans. In some centers, brain imaging and neurocognitive testing are performed in patients with aortic arch disease for whom arch repair or replacement requiring a period of deep hypothermic circulatory arrest or low pump flow is planned (463). Other centers obtain fewer preoperative tests and individualize such testing as cardiac catheterization, 24-hour Holter monitoring, brain imaging, and neurocognitive studies for patients to establish baseline states and risk.
In many centers, where the diagnosis of acute AoD is either made or highly suspected, patients are immediately taken to surgery, and TEE is performed in the operating room to either establish or confirm the diagnosis. Most of the writing committee believes that the delay to obtain coronary angiography was potentially dangerous unless patients had a history of coronary artery bypass graft surgery or a high likelihood of coexisting CAD.
Most patients undergoing elective aortic root and ascending aortic surgery can be admitted the day of surgery. However, some of the writing committee members routinely admit patients the day before surgery primarily for fluid hydration (using 5% dextrose/0.5 normal saline with additional potassium and sodium bicarbonate at 100 to 120 mL/h), particularly those who are to have extensive open surgery for arch, descending thoracic, or thoracoabdominal aortic disease. Preoperative use of acetylcysteine (600 mg by mouth at night and in the morning, or 500 mg in 500 mL of normal saline solution over 3 hours before CT or surgery) has also been described (612,613). However, the effectiveness of these strategies has not been tested in a clinical trial. Placement of thoracic epidural catheters for postoperative analgesia or lumbar spinal drains for CSF drainage is performed on the day prior to surgery in some centers. Despite lack of evidence, there is concern that neuraxial hemorrhage is more likely if blood returns through the placement needle on the day of surgery.
14.1.1 Preoperative Risk Assessment
MI, low cardiac output, respiratory failure, renal failure, and stroke are the principal causes of mortality and morbidity after operations on the thoracic aorta, and preoperative assessment of these organ systems prior to elective operation is essential (381,441,614–617), especially in patients with a history of MI, angina pectoris, or symptoms of heart failure and those older than 40 years. Patients with valvular heart disease are evaluated with echocardiography and cardiac catheterization.
Elderly patients with thoracic aortic disease are likely to have coexisting CAD. The benefits of prior or concomitant coronary revascularization are controversial. Several studies suggested that prior coronary bypass graft surgery was of benefit to patients undergoing major vascular surgery to reduce cardiovascular mortality (618–622).
More recent clinical trials comparing outcomes of patients with stable CAD treated with optimal medical therapy versus revascularization have not shown any significant reductions in cardiovascular events associated with revascularization strategies (609–611). Major thoracic aortic surgery falls into the highest-risk group for cardiac morbidity and mortality (623), prompting some writing committee members to aggressively screen for and treat coexisting CAD, but the effectiveness of such a strategy in patients with stable CAD has not been clearly established. For patients with unstable CAD, left main stenosis, or 3-vessel disease, revascularization is generally warranted prior to or concomitant with the thoracic aortic procedure. The use of drug-eluting stents for single- or double-vessel disease may be problematic because required antiplatelet therapies may significantly raise the risk of bleeding with the thoracic aortic procedure and withholding antiplatelet therapies clearly increases the risk of stent thrombosis.
History of smoking and presence of chronic pulmonary disease are important predictors of postoperative respiratory complications and are frequently present in patients with thoracic aortic disease (624). Pulmonary function tests and arterial blood-gas analyses help to risk-stratify patients with chronic pulmonary disease. If reversible restrictive disease or excessive sputum production is present, antibiotics and bronchodilators should be administered. Cessation of smoking is advisable.
Preoperative renal dysfunction is the most important predictor of acute renal failure after operations on the thoracic aorta (617,625,626). Preoperative hydration and avoidance of hypotension, low cardiac output, and hypovolemia in the perioperative period may reduce the prevalence of this complication.
To minimize the risk of stroke or reversible ischemic neurologic deficits and to determine the potential magnitude of risk, duplex imaging of the carotid arteries and angiography of the brachiocephalic arteries is often performed preoperatively particularly in patients with a history of stroke, TIA, or other risk factors for cerebrovascular disease (627). However, the efficacy of treatment of significant carotid stenosis prior to endovascular or open surgery for thoracic aortic disease has not been evaluated in a randomized clinical trial.
Although older age is a risk factor for increased early and late death after operations on the thoracic aorta (382,441,617,625,627–630), operations can be carried out successfully with satisfactory outcomes in appropriately selected older patients. Emergency operation for aortic rupture or acute dissection is associated with a higher risk of early death after operation compared with elective operation (47).
14.2 Recommendations for Choice of Anesthetic and Monitoring Techniques
1. The choice of anesthetic techniques and agents and patient monitoring techniques should be tailored to individual patient needs to facilitate surgical and perfusion techniques and the monitoring of hemodynamics and organ function. (Level of Evidence: C)
1. Transesophageal echocardiography is reasonable in all open surgical repairs of the thoracic aorta, unless there are specific contraindications to its use. Transesophageal echocardiography is reasonable in endovascular thoracic aortic procedures for monitoring, procedural guidance, and/or endovascular graft leak detection. (631–633) (Level of Evidence: B)
2. Motor or somatosensory evoked potential monitoring can be useful when the data will help to guide therapy. It is reasonable to base the decision to use neurophysiologic monitoring on individual patient needs, institutional resources, the urgency of the procedure, and the surgical and perfusion techniques to be employed in the open or endovascular thoracic aortic repair. (634,635) (Level of Evidence: B)
1. Regional anesthetic techniques are not recommended in patients at risk of neuraxial hematoma formation due to thienopyridine antiplatelet therapy, low-molecular-weight heparins, or clinically significant anticoagulation. (636) (Level of Evidence: C)
2. Routinely changing double-lumen endotracheal (endobronchial) tubes to single-lumen tubes at the end of surgical procedures complicated by significant upper airway edema or hemorrhage is not recommended. (Level of Evidence: C)
Choice of anesthetic technique is dependent on the planned surgical interventions and the patient's comorbid conditions. For placement of endovascular aortic stent grafts, different anesthetic (local, regional, general) techniques have been described, although adequate comparative studies are lacking (637–643). Percutaneous placement of catheters with a limited incision may be tolerated with local anesthesia and sedation. Extensive inguinal dissection or the construction of a femoro–femoral bypass may favor either regional or general anesthesia. If surgical dissection is extended into the retroperitoneum, a higher level of regional anesthesia or general anesthesia is required. If the patient is undergoing local anesthesia or regional anesthesia, adequate intravenous sedation is necessary because of agitation secondary to restlessness and pain from lying in one position for a prolonged period of time.
Retrospective studies indicate that patients having local versus regional or general anesthesia for endovascular stent grafts tend to have lower use of vasoactive agents and lower fluid requirements, shorter intensive care and hospital stays (642), earlier ambulation and gastrointestinal function (637), and lower incidence of respiratory and renal complications (643). In a large multicenter retrospective study of 5557 patients undergoing endovascular aortic repairs (644), 69% received general anesthesia, 25% received regional anesthesia, and 6% received local anesthesia. The incidence of cardiac complications were significantly lower in both the local or regional anesthesia group compared with general anesthesia (1.0% versus 2.9% versus 3.7%), and the incidence of sepsis was significantly lower in the regional anesthesia group compared with general anesthesia (0.2% versus 1.0%). Selection bias and complexity of disease likely affect these results.
14.2.1 Temperature Monitoring
At most centers, temperature is monitored in at least 2 locations that estimate the brain/core (e.g., blood, esophageal, tympanic membrane, nasopharynx) temperature and the visceral (e.g., bladder or rectal) temperature (645).
14.2.2 Hemodynamic Monitoring
Invasive arterial pressure monitoring is required in 1 or more sites depending on the surgical plan for cannulation and perfusion, particularly for thoracoabdominal aortic repairs. Arterial pressure is universally monitored proximal to aortic cross-clamping sites, but there is institutional variability in the monitoring of distal arterial (aortic) pressure, even when distal aortic perfusion is performed.
Central venous cannulation allows measurement of cardiac filling pressures, providing a route for vasoactive drug and fluid administration. Femoral venous catheterization is discouraged by current central line–associated bloodstream infection prevention guidelines, but the literature does not address the subject in thoracoabdominal surgery (646,647). Nevertheless, many experienced centers use short-term catheterization of the femoral vein for volume management during surgery.
Pulmonary artery catheterization is performed routinely in many institutions for thoracic aortic surgery. The literature does not specifically address the subject of thoracic aortic surgery, but the general perioperative literature does not support the use of pulmonary artery catheterization (631).
14.2.3 Transesophageal Echocardiography
TEE is safe (648) and is used to confirm the preoperative diagnoses and detect pericardial or pleural effusions, aortic regurgitation, the extent of dissection, the location of intimal tears, the size and integrity of aneurysms, and the presence of appropriate flow in the true lumen upon commencement of cardiopulmonary bypass. Current American Society of Anesthesiologists and the Society of Cardiovascular Anesthesiologists guidelines for TEE include the following (631):
Category I Indications (supported by the strongest evidence or expert opinion):
• Preoperative use in unstable patients with suspected thoracic aortic aneurysms, dissection, or disruptions that need to be evaluated quickly.
• Intraoperative assessment of aortic valve function in repair of AoDs with possible aortic valve involvement.
Category II Indications (supported by weaker evidence or expert opinion):
• Preoperative assessment of patients with suspected thoracic AoDs, aneurysms, or disruption.
• Intraoperative use during repair of thoracic AoDs without suspected aortic valve involvement.
14.2.4 Transesophageal Echocardiography for Endovascular Repairs of the Descending Thoracic Aorta
TEE can provide views of the aorta and location of guidewires and endografts prior to deployment in relation to the normal and diseased thoracic aorta (649,650). TEE has distinct advantages over angiography by providing exact vessel and lesion sizing and localization and can also be used to estimate endograft size and location. Although not imaged in all patients, large intercostal arteries have been imaged, thus avoiding inadvertent obstruction by the aortic stent graft. After stent graft placement, the presence or absence of endoleaks can be determined by TEE with a high degree of sensitivity and specificity particularly compared with contrast angiography (632,633,650). Finally, because most of these patients have severe concomitant cardiac disease, perioperative TEE allows dynamic assessments of cardiac function.
Intravascular ultrasound is used for endovascular procedures to visualize precise localization of branch arteries and for measurement of aortic and branch artery sizes. Plaques, tears, and saccular aneurysms can also be very accurately demonstrated.
14.3 Airway Management for Descending Thoracic Aortic Repairs
A double-lumen endotracheal tube or various types of endobronchial blockers facilitate surgical exposure (651). For extensive surgery of the left thorax, single-lung ventilation provides better surgical exposure, reduces the need for pulmonary retraction, may decrease the severity of iatrogenic pulmonary contusion, and protects the right lung from contamination by blood and secretions. A large descending thoracic aortic aneurysm may compress or distort the left main bronchus such that left-sided endobronchial tubes must be used with caution. If a right-sided double-lumen endotracheal tube is placed, endoscopic confirmation of tube position (to ensure right upper lobe ventilation) is necessary. Forceful endobronchial tube placement has been associated with thoracic aortic aneurysm rupture. Therefore, using a different tube or lung isolation method may be required when severe airway distortion is encountered.
At the end of surgery, some centers have advocated changing a double-lumen endotracheal tube to a single-lumen tube to facilitate pulmonary toilet and to avoid the complications associated with tube malposition in the intensive care unit. The decision to change the double-lumen endotracheal tube to a single-lumen tube should be made after carefully evaluating the extent of airway edema, as these procedures are associated with significant facial and laryngeal edema. Advanced airway management devices, such as tube exchanges and video laryngoscopy, may be of benefit; however, there is no literature addressing this subject.
14.4 Recommendation for Transfusion Management and Anticoagulation in Thoracic Aortic Surgery
1. An algorithmic approach to transfusion, antifibrinolytic, and anticoagulation management is reasonable to use in both open and endovascular thoracic aortic repairs during the perioperative period. Institutional variations in coagulation testing capability and availability of transfusion products and other prothrombotic and antithrombotic agents are important considerations in defining such an approach. (652) (Level of Evidence: C)
Thoracic aortic surgery is associated with hemorrhage from several interrelated causes, including extensive surgical dissection, arterial and venous bleeding, hypothermia, cardiopulmonary bypass, fibrinolysis, dilution or consumption of coagulation factors, thrombocytopenia, poor platelet function, heparin and other anticoagulants, preoperative antithrombotic therapy, and other congenital and acquired coagulopathies. The extensive blood product and fluid requirements of open thoracic aortic surgical repairs may exceed the total blood volume of the patient in the most complicated cases. Clinical practice guidelines for perioperative blood transfusion and blood conservation in cardiac surgery have been published by the STS and the SCA (652).
These guidelines do not specifically address open or endovascular descending thoracic aortic repairs, but the writing committee supports treatment strategies provided by these guidelines.
Coagulopathies in open descending thoracic aortic and thoracoabdominal repairs mirror many of the abnormalities seen in cardiac and thoracic aortic procedures requiring cardiopulmonary bypass. Illig and colleagues (653) reported significantly decreased euglobulin clot lysis times, elevated tissue plasminogen activator levels, elevated tissue plasminogen activator–to–plasminogen activator inhibitor-1 ratios, and reduced alpha 2-antiplasmin levels within 20 minutes after supraceliac cross-clamping, compared with infrarenal occlusion, consistent with a primary fibrinolytic state. Visceral ischemia may be the condition that initiates fibrinolysis. During supraceliac occlusion, fibrinolysis was attenuated when superior mesenteric artery perfusion was maintained via a shunt (654). Peripheral ischemia may also result in fibrinolysis. Within 30 minutes of the onset of acute peripheral ischemia (with infrarenal aortic occlusion) fibrinolytic activity increased, as demonstrated by an increase in tissue-type plasminogen activity and a decrease in plasminogen activator inhibitor activity. This increase in tissue-type plasminogen activity resulted from release of stores from ischemic vascular tissues (655). Endotoxemia during aortic occlusion may also be associated with fibrinolysis (656).
To counteract fibrinolysis, the use of lysine analogs epsilon-aminocaproic acid and tranexamic acid has been reported in cardiac surgery. The epsilon-aminocaproic acid loading or bolus dose ranged from 75 to 150 mg/kg, with additional dosing from 12.5 to 30 mg/kg/h infused over varying time periods. For tranexamic acid, loading or bolus dose, ranged from 2.5 to 100 mg/kg, with additional dosing from 0.25 to 4.0 mg/kg/h delivered over 1 to 12 hours (657). In a study of 21 adult cardiac surgical patients, the tranexamic acid dosing regimen of 10 mg/kg initial dose, followed by an infusion of 1 mg/kg/h resulted in adequate plasma concentrations defined by in vitro studies to prevent fibrinolysis, with relatively stable drug levels throughout cardiopulmonary bypass (658). Antifibrinolytic therapy for thoracoabdominal aortic surgery with distal perfusion was not associated with decreased bleeding or transfusion in a retrospective cohort of 72 patients (659). The strong evidence derived from other cardiac surgical studies has led to very common use of antifibrinolytic therapy in thoracic aortic surgery, despite the absence of specific evidence in this surgical subset.
14.5 Organ Protection
14.5.1 Recommendations for Brain Protection During Ascending Aortic and Transverse Aortic Arch Surgery
1. A brain protection strategy to prevent stroke and preserve cognitive function should be a key element of the surgical, anesthetic, and perfusion techniques used to accomplish repairs of the ascending aorta and transverse aortic arch. (660–666) (Level of Evidence: B)
1. Deep hypothermic circulatory arrest, selective antegrade brain perfusion, and retrograde brain perfusion are techniques that alone or in combination are reasonable to minimize brain injury during surgical repairs of the ascending aorta and transverse aortic arch. Institutional experience is an important factor in selecting these techniques.(211,615,667–688) (Level of Evidence: B)
1. Perioperative brain hyperthermia is not recommended in repairs of the ascending aortic and transverse aortic arch as it is probably injurious to the brain. (689–691) (Level of Evidence: B)
A brain protection strategy is an essential component of the operative technique for open surgical repairs of the ascending aorta and/or the aortic arch. Moderate or profound hypothermia with periods of circulatory arrest and/or selective antegrade brain perfusion and/or retrograde brain perfusion are the common strategies for achieving brain protection. The experience and outcomes of the operating surgeon and the institution are important considerations in selecting a brain protection strategy.
Achieving brain hypothermia is nearly universally performed using extracorporeal circulation, with temperatures ranging from 12° to 30°C. Retrograde (via jugular vein) brain perfusion is usually performed at a perfusion pressure of 20 to 40 mm Hg at a mildly or profoundly hypothermic temperature. Antegrade brain perfusion is usually performed at a perfusion pressure of 50 to 80 mm Hg and may be instituted by direct cannulation of the brachiocephalic arteries, side-graft anastomosis to the axillary artery, or direct cannulation of a portion of graft material that was anastomosed to the brachiocephalic arteries during a period of hypothermic circulatory arrest. The rewarming of a patient following completion of the repair of the thoracic aorta is usually performed at a measured rate so as not to induce brain hyperthermia.
The reviewed literature describes an evolution of brain protection techniques over the past 2 to 3 decades. Deep hypothermic circulatory arrest emerged as the first technique, but as a sole method of brain protection, it was limited by increasing rates of neurologic morbidity, other adverse outcomes, and mortality as the period of arrest exceeded 25 to 45 minutes (671,692). Deep hypothermic arrest without perfusion adjuncts has been successful, especially when arrest intervals are less than 40 minutes (693). Subsequently, various combinations of retrograde brain perfusion and selective antegrade brain perfusion were developed to extend the “safe period” of interruption of full extracorporeal circulation. Monitoring of brain function and metabolic suppression by electroencephalography, evoked potentials, bispectral index, noninvasive cerebral oximetry, and jugular bulb oxyhemoglobin saturation are additional means used to guide the onset of extracorporeal circulation interruption for repairing the distal ascending aorta and/or aortic arch. Some centers use barbiturates, calcium channel blockers, or steroids for added protection, but no prospective randomized trials have been performed to test the efficacy of pharmacological agents (211,459).
There is controversy regarding the ability of retrograde brain perfusion to support brain metabolic function and to improve neurologic outcomes, including transient postoperative neurologic dysfunction, stroke rates, and mortality (661,671,673–675,694–699). However, this technique can maintain brain hypothermia (700) and has been associated with improved outcomes in the centers where it is used as a primary neuroprotection strategy.
Selective antegrade brain perfusion may be provided by direct cannulation of 1 or more of the brachiocephalic arteries, which permits brain perfusion with minimal periods of interruption. If unilateral cannulation is performed, success may depend on patency of the circle of Willis. Alternatively, unilateral direct or side-graft cannulation of the (usually right) axillary artery permits extracorporeal circulation and cooling without manipulation of the diseased thoracic aorta. This same cannula can then be used for delivering antegrade brain perfusion immediately after the section of aorta from which the brachiocephalic arteries originate is sutured to the graft or immediately after the brachiocephalic vessels are individually anastomosed to a trifurcated graft. The time required to complete these maneuvers requires a relatively shorter period of hypothermic circulatory arrest compared with compete reconstruction of the aortic arch (681,683,701–704). Bilateral brachiocephalic artery cannulation has also been reported (705). The literature is insufficient to determine whether unilateral or bilateral perfusion or complete avoidance of circulatory arrest is associated with improved outcomes. A retrospective analysis by Svensson et al (449) suggested that axillary artery perfusion via a prosthetic side-graft was associated with improved outcomes compared with femoral arterial cannulation.
Finally, direct cannulation of the aortic replacement graft may be used to institute antegrade brain perfusion following a period of circulatory arrest. The variability of techniques among surgical centers makes direct comparison difficult; however, most studies with some type of antegrade arterial brain perfusion report outcomes that are comparable to or better than those using hypothermic circulatory arrest alone or retrograde brain perfusion (660,662,666,685,686,706,707). Furthermore, selective antegrade brain perfusion may reduce the period of brain ischemia and permit less profound hypothermia, which may be associated with good clinical outcomes (676,687,708–712).
Our ability to create evidence-based guidelines from the literature is particularly difficult in the case of brain protection. Changes in surgical technique, perfusion technology, anesthetic and intensive care management, coagulation management, prosthetic graft materials, and the experience of the reporting centers and lack of randomized clinical trials are confounding factors.
14.5.2 Recommendations for Spinal Cord Protection During Descending Aortic Open Surgical and Endovascular Repairs
1. Cerebrospinal fluid drainage is recommended as a spinal cord protective strategy in open and endovascular thoracic aortic repair for patients at high risk of spinal cord ischemic injury. (522,523,713) (Level of Evidence: B)
1. Spinal cord perfusion pressure optimization using techniques, such as proximal aortic pressure maintenance and distal aortic perfusion, is reasonable as an integral part of the surgical, anesthetic, and perfusion strategy in open and endovascular thoracic aortic repair patients at high risk of spinal cord ischemic injury. Institutional experience is an important factor in selecting these techniques. (380,382,714,715) (Level of Evidence: B)
2. Moderate systemic hypothermia is reasonable for protection of the spinal cord during open repairs of the descending thoracic aorta. (525) (Level of Evidence: B)
1. Adjunctive techniques to increase the tolerance of the spinal cord to impaired perfusion may be considered during open and endovascular thoracic aortic repair for patients at high risk of spinal cord injury. These include distal perfusion, epidural irrigation with hypothermic solutions, high-dose systemic glucocorticoids, osmotic diuresis with mannitol, intrathecal papaverine, and cellular metabolic suppression with anesthetic agents. (520,715–717) (Level of Evidence: B)
2. Neurophysiological monitoring of the spinal cord (somatosensory evoked potentials or motor evoked potentials) may be considered as a strategy to detect spinal cord ischemia and to guide reimplantation of intercostal arteries and/or hemodynamic optimization to prevent or treat spinal cord ischemia. (483,634,718,719) (Level of Evidence: B)
Paraparesis and paraplegia are perhaps the most feared complications following repair of the descending thoracic aorta. Although rates as high as 23% have previously been reported, the current incidence is probably somewhere around 2% to 6% (483,521,634,718,720). In any specific patient, however, the likelihood of neurological complications depends highly on individual anatomy, on whether the aorta is dissected or aneurysmal, and on whether the pathology is acute, chronic, or both.
Risk factors for perioperative spinal cord injury include emergency surgery, dissection, extensive disease, prolonged aortic cross-clamp time, aortic rupture, level of aortic cross-clamp, patient age, prior abdominal aortic surgery, and, in particular, hypogastric artery exclusion (143,144,721), as well as a history of renal dysfunction. The risk of paraplegia or paraparesis is minimal if the aortic cross-clamp time is less than 15 minutes (722). Svensson et al (381) reported a 20% risk of neurological injury if aortic cross-clamp time was greater than 60 minutes and less than 10% if aortic cross-clamp time was less than 30 minutes. Although paraplegia or paraparesis has been reported with aortic cross-clamp times of less than 20 minutes (723), others have concluded that aortic cross-clamp times greater than 40 minutes did not result in increases in adverse spinal cord outcome, if distal perfusion was used (382). One study of patients undergoing open or endovascular repair of descending thoracic or thoracoabdominal aortic aneurysms demonstrated a higher risk of spinal cord injury related to the extent of aorta treated but no difference between the operative approaches (724).
Another option for spinal cord protection is deep hypothermic circulatory arrest (630). This has been a useful technique for complex descending thoracic or thoracoabdominal aortic repairs performed via a left thoracotomy approach.
22.214.171.124 Monitoring of Spinal Cord Function in Descending Thoracic Aortic Repairs
In general, the comparatively low current incidence of neurologic complications is attributable to the routine use of multimodal neurophysiologic monitoring such as somatosensory evoked potentials (SSEP) and motor evoked potentials (MEP) in conjunction with neuroprotective strategies noted later (526). Although these neuroprotective strategies continue to evolve and may differ slightly in their implementation from center to center, monitoring of evoked potentials during these cases has become common because it provides the surgeon and anesthesiologist the opportunity to promptly intervene if alterations in monitored potentials indicate neurologic compromise is occurring (524,526,719,725,726).
SSEPs are cerbral cortical electrical potentials recorded with scalp electrodes during electrical stimulation of the posterior tibial or peroneal nerves of the lower extremities, conducted via the lateral and posterior columns of the spinal cord (727). Because SSEP monitoring is less sensitive to anesthetic drugs and paralytic agents may be used, its use is less complex than MEP monitoring. SSEP monitoring is limited because it is only dependent on the integrity of the lateral and posterior columns. The anterior motor column is more likely to be affected by ischemic injury during aortic reconstruction. It is thus possible to sustain an isolated perioperative anterior column injury without changes in SSEPs. In 1 study, SSEP monitoring influenced the surgical strategy in 17 of 63 patients (27%) undergoing descending aortic reconstruction (719). Corrective interventions included partial cardiopulmonary bypass initiated in 1 patient with traumatic aortic rupture; reimplantation of critical intercostal, lumbar, or sacral arteries in 11 patients; suture closure of profusely back-bleeding intercoastal arteries in 1 patient; hastening the proximal suture line in 1 patient; distal clamp repositioning to a more proximal position in 1 patient; and proximal clamp repositioning in 2 patients with left carotid ischemia. The authors reported no cases of unexplained SSEP abnormalities. New immediate paraplegia was observed in 1 patient with sustained SSEP absence, and 2 patients presented with delayed paraplegia despite normal inoperative SSEP.
In contrast, MEPs are performed by stimulating the motor cortex (by either high-voltage short-duration electrical stimulus or magnetic induction) (728) and recording at the level of the spinal cord, peripheral nerves, or muscles (524,526,725–727). Neurogenic MEPs are responses recorded at the peripheral nerves, whereas myogenic MEPs are large biphasic responses recorded over the muscle belly. Because the amplitude of the response is proportional to the number of motor neurons being stimulated, these evoked potentials are very sensitive to neuromuscular blocking and anesthetic agents (729–738).
Monitoring of SSEPs alone has not been demonstrated to improve outcomes in patients undergoing TAA repair. SSEP monitoring is associated with delayed ischemia detection compared with transcranial MEPs (739), as well as high rates of both false-negative and false-positive results (635,739). Lyon et al (740) reported a significantly higher voltage threshold for the generation of a 50 microV amplitude signal at the end of the procedure compared with