INSIGHTS INTO ACUTE

Santi Trimarchi ISBN: 978-94-6108-288-6 Lay-out and printed by: Gildeprint Drukkerijen - Enschede, the Netherlands INSIGHTS INTO ACUTE AORTIC DISSECTION

Inzichten in acute aortadissectie (met een samenvatting in het Nederlands)

Proefschrift

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof.dr. G.J. van der Zwaan, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op donderdag 3 mei 2012 des middags te 4.15 uur

door

Santi Trimarchi

geboren op 14 april 1965 te Santa Teresa di Riva (ME), Italië Promotoren: Prof. dr. F.L. Moll

Co-promotoren: Dr. J.A. van Herwaarden Dr. B.E. Muhs

This thesis was (partly) accomplished with financial support from Policlinico San Donato IRCCS TAblE Of CONTENTS

Part one: General introduction

Chapter 1: Introduction 7

Part two: Role of Aortic Diameter in acute dissection

Chapter 2: Descending aortic diameter of 5.5 cm or greater is not an accurate 17 predictor of acute type B aortic dissection. Journal Thoracic Cardiovascular 2011 Sep;142(3):e101-7.

Chapter 3: Acute Type B Aortic Dissection in the Absence of Aortic Dilatation. 31 Journal of Vascular Surgery (Accepted)

Part three: Risk stratification for mortality

Chapter 4: Role of age in acute type A aortic dissection outcome: 43 Report from the International Registry of Acute Aortic Dissection (IRAD). Journal Thoracic Cardiovascular Surgery 2010 Oct;140(4):784-9.

Chapter 5: Age-related Decision Making in Complicated Acute Type B 57 Aortic Dissection (Submitted)

Chapter 6: Importance of Refractory Pain and Hypertension in Acute Type B 71 Aortic Dissection: Insights From the International Registry of Acute Aortic Dissection (IRAD). Circulation. 2010 Sep 28;122(13):1283-1289.

Chapter 7: Acute Abdominal Aortic Dissection: Insights from the International 85 Registry of Acute Aortic Dissection (IRAD). Journal of Vascular Surgery 2007 Nov;46(5):913-919. Part four: Evolution of descending aortic dissection (or type b dissection)

Chapter 8: Aortic Expansion after Uncomplicated Acute Type B Aortic Dissection. 99 (Submitted)

Chapter 9: Long-term outcomes of surgical aortic fenestration for complicated 113 acute type B aortic dissections. Journal of Vascular Surgery 2010 Aug;52(2):261-6.

Part five: biomarkers in aortic dissection

Chapter 10: Circulating transforming growth factor-Beta levels in acute 125 aortic dissection. Journal of the American College of . 2011 Aug 9;58(7):775.

Chapter 11: In search of blood tests for thoracic aortic diseases. 129 Annals of Thoracic Surgery 2010 Nov;90(5):1735-42.

Part six: General Discussion of the Thesis

Chapter 12: Summary and General Discussion 147 Chapter 13: Summary in Dutch – Nederlandse Samenvatting 157 Chapter 14: Review Committee 169 Acknowledgements 171 List of publications 177 Curriculum Vitae 187 Chapter 1

Introduction Chapter 1

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

8 Introduction

INTRODUCTION R1 1 R2 Aortic dissection represents one of the most catastrophic and complex cardiovascular diseases. R3 Its origin is related to an intimal tear with course of blood flow into the aortic wall. The tear is R4 due to a weakened aorta or repetitive, excessive forces on the arterial wall. Through the laceration R5 of the aortic intima and inner layers of the media, blood flow divides the aortic lumen into two R6 different lumens, defined as the true and false lumen, and separated by a septum or intimal R7 flap. Most frequently, aortic dissection extend distally causing additional tears in 90% of the R8 cases.1-4 These re-entries allow the blood communication between the two aortic lumens. The R9 proximal tear is located in ascending aorta in about 60% of the dissections, at the level of the R10 proximal descending aorta in about 30% and in the aortic arch in the remaining 10%. 1-4 Such R11 different sites of origin are related to the mechanical forces acting on the aorta during the cardiac R12 cycle, consisting of shear stress and normal stress vectors. During the cardiac cycle the greatest R13 mechanical forces are generated in the ascending due to the convexity, the highest pressurization R14 and greatest distention, then at the level of the origin of the left subclavian , where the aorta R15 is relative fixated. R16 R17 Classification R18 Based on the location of the entry tear, aortic dissections are actually classified as type A when R19 the proximal tear is located in the ascending aorta and as type B when the tear is present after the R20 origin of the left subclavian artery (Stanford classification).5 The other historical classification is R21 the De Bakey classification, who classify patients based on the origin of the intimal tear and the R22 extent of the dissection. DeBakey type I include those patients with both ascending, descending R23 and abdominal aorta involved by the acute dissection, DeBakey type II with aortic involvement R24 limited to the ascending aorta and DeBakey type III when the dissection origins below the R25 ostium of the left subclavian artery involving distally only to the descending aorta (DeBakey type R26 IIIa) or extending to the abdominal section (DeBakey type IIIb).5 R27 Whereas classic dissection is characterized by an intimal flap with two lumens, some patients R28 have only a crescentic thickening of the aortic wall without an entry point, defined as intramural R29 hematoma. R30 The interval time between the onset of symptoms and presentation are also classified as within 2 R31 weeks as acute, between 2 and 6 weeks as subacute and after as chronic.5 R32 R33 Incidence R34 The overall incidence of the acute aortic dissection as a cause of mortality in the population, R35 seems to be around 0.5% per year, with a frequency of about 2.9-4 per 100.000 people a R36 year, about two times higher than aortic rupture.6-8 Recently, a study in the Swedish R37 population showed an increased incidence with 16 per 100.000 men a year.9. The disease affects R38 R39

9 Chapter 1

R1 more commonly the fifth decade of life and the male population, with a frequency varying from R2 2:1 to 5:1, compared to the females.1, 6, 9 Patients with type A are usually younger (25-55 years R3 old) than those with type B, who aged between 60 and 70 years.1, 6-9 R4 R5 Risk factors R6 Arterial hypertension is the clinical sign more frequently present in the history of patients with R7 aortic dissection aged over 45 years.1 Known associated conditions with dissections are several R8 connective tissue disorders, such as Marfan, Ehlers Danlos, Loyes-Dietz, Turner’s and Noonan R9 syndromes.10 In particular, Marfan syndrome, seems to be the cause of the acute event in about R10 3 5% of all dissections due to a genetic defect in the synthesis of the fibrillin microfibrillar.10, 11 R11 Other congenital cardiovascular diseases are often associated with aortic dissection, like R12 coarctation and the presence of bicuspid aortic valves.10-14 As well, a positive family history R13 of thoracic aneurysm is a relevant risk factor, while the role of the aortic diameter as primary R14 cause of acute dissection has been recently issue of debate.15 In the adulthood, the medial cystic R15 necrosis of the aorta has been considered the primary cause of dissection for long time. It’s an R16 idiopathic degeneration of the tunica media, determined by the alteration of smooth muscle R17 cells in the tunica media. Predisposing cause or determinant is also atherosclerosis, which causes R18 almost exclusively distal forms of dissection, pregnancy, cocaine abuse, strenuous activities and R19 severe emotional stress, some inflammatory diseases such as syphilis and giant cell arteritys and R20 trauma. Other less frequent causes are systemic lupus erythematosus, osteogenesis imperfecta, R21 chronic nephropathic cystinosis, pheochromocytoma and hypercortisolism. Acute dissections R22 may be also due to iatrogenic causes such as interventional cardiac or radiological procedures, R23 and during cardiac operations, i.e. arterial cannulations, aortic clamping or previous aortotomy. R24 R25 Symptoms R26 Chest pain is most commonly the initial symptom of acute dissection, present in approximately R27 90% of cases, with a particular location depending on the affected aortic segment.16-18 In acute R28 type A dissections chest pain may be confined to the anterior chest region, while in type B R29 dissection the pain is usually located in the interscapular region. The pain is referred as sudden, R30 severe, unremitting, often with very specific characteristics of migration along the course of the R31 involved aorta and its branches.1 Dissections rarely occur without pain, as forms of type A in R32 patients already affected by ascending .18, 19 R33 Acute aortic dissections type A and B may present and/or evolve as complicated or uncomplicated. R34 In type A, most frequent complications are represented by pericardial effusion, aortic rupture R35 with cardiac tamponade and dissection involvement of the origin of the coronary vessels, R36 reported in about 7% of cases and often associated with acute myocardial infarction.20-22 R37 Additional complications in type A dissection are cardiac arrest and acute failure which R38 may be caused by sudden valvular aortic regurgitation due to incompetence of the sino-tubular R39

10 Introduction

junction. When the aortic arch is involved, supra-aortic trunks may be affected in about 1 / 3 R1 of cases, determining neurological signs, such as stroke, coma and brain death in 18% of such 1 R2 patients.23-25 Acute B dissection may be complicated in about 50% of the cases, presenting with R3 shock due to aortic rupture, malperfusion like visceral, renal and/or leg ischemia, paraplegia, R4 refractory pain and refractory hypertension.1, 7 Preoperative complications are mostly predictors R5 of poor outcome. R6 R7 Diagnosis R8 Diagnosis of acute dissection needs to be immediate, in view of its rapid and tragic evolution. R9 Type A dissections can be identified by a trans-thoracic or trans-esophageal echocardiography; R10 the latter in many centers is currently the only preoperative investigation. In patients with R11 clinical suspicion the test of choice is the CT scan performed with iodinate contrast, which shows R12 sensitivity and specificity close to 100%.26, 27 In cases of suspicion of ischemic organ, may be R13 necessary proceed with the arteriography that may represent also the first step of an endovascular R14 therapy. Rarely MRI is adopted in the acute setting, although its relevant sensitivity.27 Differently R15 then acute coronary syndrome, actually there are no specific biomarkers useful for making a R16 standard serologic diagnosis of acute aortic syndromes. R17 R18 Therapy R19 Untreated aortic dissections are associated with prognosis almost always poor. It has been R20 reported that 20-30% of all aortic dissection patients die before hospital admission 9 Thereafter R21 1% of the patients die per hour and 5 to 20% will expire during or after surgery. A similar R22 prognosis had been evidenced previously, emphasizing that about 50% of affected patients died R23 within 48 hours, 85% in month 1, 90% in the first 3 months and 92% in the first year. 28-31 R24 The use of specific antihypertensive therapy in the early hours after the acute event has certainly R25 improved the incidence of immediate death, allowing a time extension needed to complete both R26 the diagnostic and patient transport into the operating room. While the acute type A dissection R27 presents an mandatory indication for surgery, the indications for treating acute type B have R28 been evolving over the years and are still the subject of different evaluations.1, 32 The aim of the R29 treatment of type B dissection is to prevent the aortic rupture, to maintain or restore R30 of the vital organs and prevent progression of the dissection. Therefore it is important to make a R31 quality risk assessment at a early stage, when patients may benefit from medical or endovascular R32 or surgical intervention. Actually, in uncomplicated B patients, medical therapy is considered R33 the therapy of choice, using B-blockers, peripheral vasodilators, such as sodium nitroprusside R34 and nitroglycerin, and Ca channel blockers as Nifedipine. In complicated B patients, the R35 introduction of stent-grafting has opened up the way for less invasive treatments to improve on R36 the dismal results following acute surgery of the descending aorta and reducing the morbidity R37 related to surgical treatment in general. As stent-grafting has been introduced in clinical practise, R38 R39

11 Chapter 1

R1 it has become the first invasive approach to such complicated B patients.33, 34 Currently, the R2 open surgical management seems to be limited to complications such as the progression of R3 the dissection with retrograde involvement of the aortic arch, failure of endovascular stenting R4 or percutaneous fenestration or large reconstructive aortic procedures in young patients with R5 connective tissue disorders.5 R6 R7 Questions R8 1. Is the role of the aortic diameter totally clarified as cause of acute type B aortic dissection? R9 2. How important are the preoperative conditions of acute dissected patients in R10 determining their outcome? R11 3. Are there any useful and available specific biomarkers for making a rapid serologic R12 diagnosis of acute dissection? R13 R14 Objectives and outline of this thesis R15 The main objective of this thesis is to look insight acute aortic dissection in the contemporary era, R16 during which new data modify continuously our knowledge and the consequent management R17 of such patients. R18 Chapter 2 describes as descending aortic diameter of 5.5 cm or greater is not an accurate predictor R19 of acute type B aortic dissection, as previously reported by many papers, included consensus and R20 guidelines. Chapter 3 focuses on those patient who are affected by acute B dissection in the R21 absence of aortic dilatation, reporting on characteristics, history, presentation and results. These R22 two clinical projects, which were developed using the database of the International Registry R23 of Acute Aortic Dissection (IRAD), evidenced as the role of the aortic diameter in acute B R24 dissection is still unclear and needs further investigations. R25 The importance of age in determining management and results of type A and B dissection is R26 showed in Chapter 4 and 5. In-hospital results of patients with acute type B dissection presenting R27 with refractory hypertension and pain and no hemodynamic or ischemic complications are R28 described in Chapter 6. Chapter 7 reports on patients affected by acute abdominal aortic R29 dissection, an uncommon anatomical localization of this acute aortic syndrome. Also these three R30 research projects were elaborated using the IRAD database. They showed as the age is a primary R31 predictor of outcome in these patients, as well the preoperative clinical conditions. Such studies R32 have underlined the importance of risk stratification for predicting the outcome of such patients, R33 being acute dissection a syndrome with a multi-potential clinical scenario, each associated with R34 a different result. R35 Chapter 8 focuses on the predictors of aortic expansion after uncomplicated acute type B R36 dissection which was managed medically in the acute setting. The investigation evidenced that a R37 small aortic diameter at presentation of B dissection is associated with increased aortic expansion R38 during follow-up, while female gender and IMH were associated with less aortic expansion. In R39

12 Introduction

Chapter 9 long-term results of open surgical fenestrations for complicated acute B dissection are R1 described. The experience shows that the acute dissected aortic segment, where the longitudinal 1 R2 aortotomy was performed to manage visceral and renal ischemic complications, do not dilate R3 over time. R4 Circulating transforming growth factor-Beta levels in acute aortic dissection has described in R5 Chapter 10. Chapter 11 reports an update on the contemporary potential biomarkers that R6 might be useful in the serologic diagnosis and follow-up of aortic dissections. R7 Chapter 12 provides a summary of the most important finding and speculation regarding the R8 contents of this thesis. R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

13 Chapter 1

R1 REfERENCE lIST R2 R3 (1) Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, et al. The R4 International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. R5 JAMA 2000 Feb 16;283(7):897-903. R6 (2) Nienaber CA, Eagle KA. Aortic dissection: new frontiers in diagnosis and management: Part I: R7 from etiology to diagnostic strategies. Circulation 2003 Aug 5;108(5):628-35. R8 (3) Nienaber CA, Eagle KA. Aortic dissection: new frontiers in diagnosis and management: Part II: R9 therapeutic management and follow-up. Circulation 2003 Aug 12;108(6):772-8. R10 (4) Tsai TT, Nienaber CA, Eagle KA. Acute aortic syndromes. Circulation 2005 Dec 13;112(24):3802- R11 13. R12 (5) Hiratzka LF, Bakris GL, Beckman JA, Bersin RM, Carr VF, Casey DE, Jr., et al. 2010 R13 ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM Guidelines for the diagnosis R14 and management of patients with thoracic aortic disease. A Report of the American College R15 of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, R16 American Association for Thoracic Surgery, American College of Radiology,American Stroke R17 Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography R18 and Interventions, Society of Interventional Radiology, Society of Thoracic ,and Society R19 for Vascular . J Am Coll Cardiol 2010 Apr 6;55(14):e27-e129. R20 (6) Clouse WD, Hallett JW, Jr., Schaff HV, Spittell PC, Rowland CM, Ilstrup DM, et al. Acute R21 aortic dissection: population-based incidence compared with degenerative aortic aneurysm R22 rupture. Mayo Clin Proc 2004 Feb;79(2):176-80. R23 (7) Meszaros I, Morocz J, Szlavi J, Schmidt J, Tornoci L, Nagy L, et al. Epidemiology and R24 clinicopathology of aortic dissection. Chest 2000 May;117(5):1271-8. R25 (8) Kurland BF, Elveback LR, Norbrega FT. The community as an epidemiologic laboratory: a R26 casebook of community studies. In: Kessler II, Levin ML, editors. Population study in Rochester R27 and Olmsted County, Minnesota, 1906–1968.Baltimore, MD: Johns Hopkins Press.; 1970. R28 (9) Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection: R29 increasing prevalence and improved outcomes reported in a nationwide population-based study R30 of more than 14,000 cases from 1987 to 2002. Circulation 2006 Dec 12;114(24):2611-8. R31 (10) Svensson LG, Rodriguez ER. Cardiovascular and Vascular disease of the Aorta.Philadelphia, PA: R32 WB Saunders Co; 1997. R33 (11) Januzzi JL, Isselbacher EM, Fattori R, Cooper JV, Smith DE, Fang J, et al. Characterizing the R34 young patient with aortic dissection: results from the International Registry of Aortic Dissection R35 (IRAD). J Am Coll Cardiol 2004 Feb 18;43(4):665-9. R36 (12) Coady MA, Davies RR, Roberts M, Goldstein LJ, Rogalski MJ, Rizzo JA, et al. Familial patterns R37 of thoracic aortic aneurysms. Arch Surg 1999 Apr;134(4):361-7. R38 (13) Larson EW, Edwards WD. Risk factors for aortic dissection: a necropsy study of 161 cases. Am J R39 Cardiol 1984 Mar 1;53(6):849-55.

14 Introduction

(14) Bonderman D, Gharehbaghi-Schnell E, Wollenek G, Maurer G, Baumgartner H, Lang IM. R1 Mechanisms underlying aortic dilatation in congenital malformation. Circulation 1 R2 1999 Apr 27;99(16):2138-43. R3 (15) Elefteriades JA. Indications for aortic replacement. J Thorac Cardiovasc Surg 2010 Dec;140(6 R4 Suppl):S5-S9. R5 (16) DeBakey ME, McCollum CH, Crawford ES, Morris GC, Jr., Howell J, Noon GP, et al. R6 Dissection and dissecting aneurysms of the aorta: twenty-year follow-up of five hundred twenty- R7 seven patients treated surgically. Surgery 1982 Dec;92(6):1118-34. R8 (17) Hirst AE, Jr., JOHNS VJ, Jr., KIME SW, Jr. Dissecting aneurysm of the aorta: a review of 505 R9 cases. Medicine (Baltimore) 1958 Sep;37(3):217-79. R10 (18) Miller DC. Surgical treatment of aneurysms and dissections involving the transverse aortic R11 arch. In: Jamieson SW, Shumway NE, editors. Operative Surgery-.: R12 Butterworths; 1986. p. 538. R13 (19) DeSanctis RW, Doroghazi RM, Austen WG, Buckley MJ. Aortic dissection. N Engl J Med 1987 R14 Oct 22;317(17):1060-7. R15 (20) Cambria RP, Brewster DC, Gertler J, Moncure AC, Gusberg R, Tilson MD, et al. Vascular R16 complications associated with spontaneous aortic dissection. J Vasc Surg 1988 Feb;7(2):199-209. R17 (21) Hirata K, Wake M, Kyushima M, Takahashi T, Nakazato J, Mototake H, et al. Electrocardiographic R18 changes in patients with type A acute aortic dissection. Incidence, patterns and underlying R19 mechanisms in 159 cases. J Cardiol 2010 Sep;56(2):147-53. R20 (22) Shapira OM, Davidoff R. Images in cardiovascular medicine. Functional left main coronary R21 artery obstruction due to aortic dissection. Circulation 1998 Jul 21;98(3):278-80. R22 (23) Erbel R, Oelert H, Meyer J, Puth M, Mohr-Katoly S, Hausmann D, et al. Effect of medical and R23 surgical therapy on aortic dissection evaluated by transesophageal echocardiography. Implications R24 for prognosis and therapy. The European Cooperative Study Group on Echocardiography. R25 Circulation 1993 May;87(5):1604-15. R26 (24) Tsai TT, Isselbacher EM, Trimarchi S, Bossone E, Pape L, Januzzi JL, et al. Acute type B aortic R27 dissection: does aortic arch involvement affect management and outcomes? Insights from the R28 International Registry of Acute Aortic Dissection (IRAD). Circulation 2007 Sep 11;116(11 R29 Suppl):I150-I156. R30 (25) Umana JP, Lai DT, Mitchell RS, Moore KA, Rodriguez F, Robbins RC, et al. Is medical therapy R31 still the optimal treatment strategy for patients with acute type B aortic dissections? J Thorac R32 Cardiovasc Surg 2002 Nov;124(5):896-910. R33 (26) Sebastia C, Pallisa E, Quiroga S, varez-Castells A, Dominguez R, Evangelista A. Aortic dissection: R34 diagnosis and follow-up with helical CT. Radiographics 1999 Jan;19(1):45-60. R35 (27) Small JH, Dixon AK, Coulden RA, Flower CD, Housden BA. Fast CT for aortic dissection. Br R36 J Radiol 1996 Oct;69(826):900-5. R37 (28) Anagnostopoulos CE, Prabhakar MJ, Kittle CF. Aortic dissections and dissecting aneurysms. Am R38 J Cardiol 1972 Aug;30(3):263-73. R39

15 Chapter 1

R1 (29) Lindsay J, Jr., Hurst JW. Clinical features and prognosis in dissecting aneurysm of the aorta. A R2 re-appraisal. Circulation 1967 May;35(5):880-8. R3 (30) Svensson LG, Crawford ES, Hess KR, Coselli JS, Safi HJ. Dissection of the aorta and dissecting R4 aortic aneurysms. Improving early and long-term surgical results. Circulation 1990 Nov;82(5 R5 Suppl):IV24-IV38. R6 (31) Wheat MW, Jr., Palmer RF, Bartley TD, SEELMAN RC. Treatment of dissecting aneurysms of R7 the aorta without surgery. J Thorac Cardiovasc Surg 1965 Sep;50:364-73. R8 (32) Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Mehta RH, et al. Contemporary R9 results of surgery in acute type A aortic dissection: The International Registry of Acute Aortic R10 Dissection experience. J Thorac Cardiovasc Surg 2005 Jan;129(1):112-22. R11 (33) Buffolo E, da Fonseca JH, de Souza JA, Alves CM. Revolutionary treatment of aneurysms R12 and dissections of descending aorta: the endovascular approach. Ann Thorac Surg 2002 R13 Nov;74(5):S1815-S1817. R14 (34) Leurs LJ, Bell R, Degrieck Y, Thomas S, Hobo R, Lundbom J. Endovascular treatment of thoracic R15 aortic diseases: combined experience from the EUROSTAR and United Kingdom Thoracic R16 Endograft registries. J Vasc Surg 2004 Oct;40(4):670-9. R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

16 Chapter 2

Descending Aortic Diameter ≥ 5.5 is not an Accurate Predictor of Acute Type B Aortic Dissection

Journal Thoracic Cardiovascular Surgery 2011 Sep;142(3):e101-7.

Santi Trimarchi,1 MD,Frederik H. W. Jonker, MD,2 Stuart Hutchison, MD,3 Eric M. Isselbacher, MD,4 Linda A. Pape, MD,5 Himanshu J. Patel, MD,6James B. Froehlich, MD, MPH,6 Bart E. Muhs, MD, PhD,2 Vincenzo Rampoldi, MD,1 Viviana Grassi, MD,1 Arturo Evangelista, MD,7 Gabriel Meinhardt, MD,8Joshua Beckman, MD,9 Truls Myrmel, MD,10 Reed E. Pyeritz, MD,11 Alan T. Hirsch, MD,12 Thoralf M. Sundt III, MD,13 Christoph A. Nienaber, MD,14 and Kim A. Eagle, MD 6

1 Policlinico San Donato IRCCS, San Donato Milanese, 2 Yale University School of Medicine, New Haven Conn, USA 3 University of Calgary, Calgary, Alberta, Canada 4Massachusetts General Hospital, Boston, Mass, USA 5 University of Massachusetts Hospital, Worcester,Mass 6 University of Michigan, Ann Arbor, Michigan, USA 7 Hospital General Universitary Vall d’Hebron, Barcelona, Spain 8 Robert-Bosh Krakenhaus, Stuttgart, Germany 9 Brigham and Women’s Hospital, Boston, Masss, USA 10 Tromso University Hospital, Tromso, Norway 11 University of Pennsylvania, School of Medicine, Philadelphia, Pa, USA 12 Minneapolis Heart Institute, Minneapolis, Minn, USA 13 Mayo Clinic,m Rochester, Minn,USA 14 University of Rostock, Rostock, Germany Chapter 2

R1 AbSTRACT R2 R3 Objectives R4 The risk of acute type B aortic dissection (ABAD) is thought to increase with descending thoracic R5 aortic diameter. Currently, elective repair of the descending thoracic aorta is indicated for an R6 aortic diameter ≥ 5.5cm. We sought to investigate the relationship between aortic diameter and R7 ABAD, and the utility of aortic diameter as a predictor of ABAD. R8 R9 Methods R10 We examined the descending aortic diameter at presentation of 613 cases of ABAD enrolled R11 in the International Registry of Acute Aortic Dissection (IRAD) between 1996 and 2009, and R12 analyzed the subset of ABAD patients with an aortic diameter < 5.5cm. R13 R14 Results R15 The median aortic diameter at the level of ABAD was 4.1 cm (range 2.1 to 13.0 cm). Only 18.4% R16 of ABAD patients in IRAD had an aortic diameter ≥ 5.5 cm. Cases with Marfan syndrome R17 represented 4.3%, and had a slightly larger aortic diameter than non-Marfans (4.68 cm vs. 4.32 R18 cm, p=.121). Complicated ABAD was more common among patients with an aortic diameter R19 ≥ 5.5 cm (52.2% vs. 35.6%, p<.001), and the in-hospital mortality for patients with an aortic R20 diameter < 5.5 cm and ≥ 5.5 cm were 6.6% and 23.0% (p<.001), respectively. R21 R22 Conclusions R23 The vast majority of patients with ABAD present with a descending aortic diameters < 5.5 cm R24 prior to dissection, and do not fall within the guidelines for elective descending thoracic aortic R25 repair. Aortic diameter measurements do not appear to be a useful parameter to prevent aortic R26 dissection, and other methods are needed to identify patients at risk for ABAD. R27 R28 Ultramini abstract R29 The descending aortic diameter at presentation of 613 cases of acute type B aortic dissection R30 (ABAD) enrolled in IRAD between 1996 and 2009 was examined. Only 18.4% of patients R31 had an aortic diameter ≥ 5.5 cm, so the vast majority of ABAD patients do not fall within the R32 guidelines for elective descending thoracic aortic repair. R33 R34 R35 R36 R37 R38 R39

18 Aortic diameter of 5.5 cm or greater is not a predictor

INTRODUCTION R1 R2 Acute type B aortic dissection (ABAD) is a serious cardiovascular condition that is associated R3 1-4 R4 with high mortality and morbidity rates, and its incidence has increased over the last decades. 2 The risks of ABAD and aortic rupture are thought to increase with increasing descending R5 thoracic aortic diameter.5-7 For thoracic aortic aneurysms with a diameter larger than 6 cm, R6 dissection occurs in about 3.7% per year, and dissection, rupture and/or death in about 15% R7 per year.5-7 The risks of elective thoracic aortic repair for preventing acute dissection or rupture R8 are considerable as well, and the risks of elective surgery have to be balanced against the risks of R9 the natural course of descending thoracic aortic aneurysms. Establishing a diameter threshold R10 value that discriminates low from high risk of acute dissection is therefore crucial. Although the R11 optimal diameter threshold is still under debate, current indications are to treat at a descending R12 thoracic aortic diameter of above 5.5 or 6.0 cm.5-9 R13 However, investigating the descending thoracic aortic diameter among patients with ABAD R14 enrolled in the International Registry of Acute Aortic Dissection (IRAD), we observed that a R15 substantial part of these patients do not seem to have markedly dilated aortic diameters. We R16 sought to assess the utility of aortic diameter as a predictor of ABAD by investigating the subset R17 of patients with ABAD that have an aortic diameter < 5.5 cm in IRAD, for which elective R18 descending thoracic aortic repair would not be recommended. R19 R20 R21 METHODS R22 R23 Patient selection R24 The inception and structure of IRAD has been described previously.10 All patients presenting R25 with ABAD that were enrolled in the International Registry of Aortic Dissection between 1996 R26 and September 2009 were selected for analysis. In total, we identified 613 patients with ABAD R27 for whom data regarding the aortic diameter were available. Acute type B aortic dissection R28 was defined as any dissection that originated at the descending thoracic aorta in a patient R29 that presented within 14 days after onset of symptoms. Only patients where the descending R30 aortic diameter at presentation was available were included for this evaluation; traumatic aortic R31 dissections were excluded. Patients were categorized according to aortic diameter < 5.5 cm, and ≥ R32 5.5 cm, and demographics, clinical presentation, management and outcomes of the two groups R33 were compared. R34 R35 R36 R37 R38 R39

19 Chapter 2

R1 Data collection R2 Data were collected using standardized forms with >290 clinical variables, including R3 demographics, medical history, clinical presentation, clinical findings, imaging studies, medical R4 and surgical management, and in-hospital outcomes. The maximum descending aortic diameters R5 were measured by computed tomography (CTA), transesophageal echocardiography (TEE), R6 magnetic resonance imaging (MRI), and/or angiography at the time of presentation with ABAD. R7 If patients underwent multiple imaging studies, diameters from the study that identified the R8 largest aortic diameter were selected for analysis. The maximum aortic diameter was measured at R9 cross-sectional images perpendicular to the long axis of the descending aorta. In IRAD, all aortic R10 diameter measurements were obtained after aortic dissection had occurred. Data were collected R11 at presentation or by retrospective review, and were submitted to the IRAD coordinating center R12 at the University of Michigan. Submitted forms are checked for face validity and analytical R13 internal validity. R14 R15 Data analysis R16 Data analysis was performed with the use of SPSS statistical analysis software (SPSS Inc., Chicago, R17 Ill.). The Chi-square test was used for comparing categorical variables between patients with an R18 aortic diameter <5.5 cm and ≥5.5 cm; the Student’s t-test was used for comparing continuous R19 variables between both groups. Summary statistics are presented as frequencies and percentages R20 for categorical variables and mean ±SD for continuous variables. In all cases, missing data were R21 not defaulted to negative, and denominators reflect only cases reported. Forward stepwise logistic R22 regression analysis was performed with variables that reached P< 0.20 on univariate analysis to R23 investigate independent predictors of in-hospital mortality at aortic diameter < 5.5 cm. A P value R24 < .05 was considered significant. The authors had full access to and took full responsibility for the R25 integrity of the data. All authors have read and agree to the manuscript as written. R26 R27 R28 RESUlTS R29 R30 Aortic Diameters R31 The mean maximum diameter at the level of ABAD was 4.40 ±1.35 cm (range 2.1 to 13.0 cm) R32 and the median descending thoracic aortic diameter was 4.1 cm (figure 1). Among all 613 ABAD R33 patients, 81.6% had an aortic diameter < 5.5 cm, which is the current threshold for elective R34 descending thoracic aortic repair (figure 1), and 40.3% had an aortic diameter < 4.0 cm. Only R35 18.4% of ABAD patients in IRAD had an aortic diameter ≥ 5.5cm. R36 R37 R38 R39

20 Aortic diameter of 5.5 cm or greater is not a predictor

R1 R2 R3 R4 2 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 Figure 1. Descending aortic diameter of patients with acute type B aortic dissections in IRAD R16 Among all 613 patients with acute type B aortic dissection, 81.6% had an aortic diameter < 5.5 cm, which is the current threshold for elective descending thoracic aortic repair R17 R18 Demographics and Clinical History R19 Patients with an aortic diameter ≥ 5.5 cm were more likely to have experienced a prior aortic R20 dissection (19.1% vs. 7.7%, p<.001) and/or a known aortic aneurysm (39.6% vs. 16.8%, R21 p<.001). Marfan’s syndrome was present in 6.4% of patients with an aortic diameter ≥ 5.5cm, R22 compared with 3.8% in patients with an aortic diameter < 5.5cm (p=.227) (table 1). The mean R23 aortic diameter of patients with Marfan syndrome was 4.68 cm vs. 4.32 cm in non-Marfan R24 patients (p=.121) R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

21 Chapter 2

R1 Table 1. Demographics and clinical history R2 Overall Descending Descending P Value R3 < 5.5 cm > 5.5 cm R4 N (%) 613 (100) 500 (81.6) 113 (18.4) Demographics R5 Age (sd) 63.9 (±14.0) 62.9 (±13.5) 65.2 (±14.1) .111 R6 Age > 70 years 219 (36.6) 168 (34.3) 51 (47.2) .012 R7 Male 409 (66.7) 335 (67.0) 74 (65.5) .758 R8 History Marfan 26 (4.3) 19 (3.8) 7 (6.4) .227 R9 Bicuspid Aortic Valve 9 (1.9) 8 (2.0) 1 (1.3) .997 R10 IMH 116 (19.1) 97 (19.5) 19 (17.1) .561 R11 Hypertension 486 (79.8) 392 (78.7) 94 (84.7) .157 R12 Atherosclerosis 207 (34.3) 166 (33.5) 41 (38.0) .380 Diabetes 42 (7.0) 31 (6.3) 11 (10.2) .154 R13 Known aortic aneurysm 127 (21.0) 83 (16.8) 44 (39.6) <.001 R14 Prior aortic dissection 59 (9.8) 38 (7.7) 21 (19.1) <.001 R15 Prior aortic repair 84 (14.1) 63 (13.0) 21 (19.3) .089 R16 Prior CABG 28 (4.8) 23 (4.8) 5 (4.6) .943 R17 Prior Cath/Angiography 42 (9.1) 37 (9.6) 5 (6.5) .392 Iatrogenic Dissection 12 (2.1) 12 (2.5) 0 (0) .136 R18 R19 IMH: intramural hematoma, CABG: coronary artery bypass grafting. R20 R21 Presentation and Imaging Results R22 Patients with an aortic diameter < 5.5 cm presented more frequently hypertensive (70.1% vs. R23 47.3%, p<.001), and less frequently normotensive (26.1% vs. 45.5%, p<.001) or in hypovolemic R24 shock (0.6% vs. 3.6%, p=.008). Complicated ABAD, defined as patients presenting with pre- R25 operative mesenteric ischemia/infarction and/or, acute renal failure, limb ischemia, spinal cord R26 ischemia, shock, recurrent pain, refractory pain, or refractory hypertension, were more common R27 among patients with an aortic diameter ≥ 5.5 cm (52.2% vs. 35.6%, p<.001, table 2). Peri-aortic R28 hematoma was more frequently diagnosed in this group (28.0% vs. 13.7%, p<.001). R29 As expected, patients with an aortic diameter < 5.5 cm were more likely to have a normal chest R30 X-ray (29.1% vs. 10.6%, p<.001), while a widened mediastinum, an abnormal aortic contour, R31 displacement of the aorta and pleural effusion were more frequent on chest X-rays of patient R32 with an aortic diameter ≥ 5.5 cm (table 2). Partial false lumen thrombosis was more common in R33 patients with enlarged aortas (53.0% vs. 33.0%, p<.001). R34 Mean aortic arch diameter was 3.61 cm (± 1.24), and was significantly larger in patients with R35 aortic diameter ≥ 5.5 cm (3.49 vs. 4.16, p < 0.001). The site of origin of the ABAD was more R36 commonly at the level of the left subclavian artery for those patients having a smaller aorta than R37 5.5 cm (58.7% vs. 44.0%, p=0.005), while for those ≥ 5.5 cm it was, more likely to be in the R38 descending aorta (20.9% vs. 33.9%, p=0.002) (table 2). R39

22 Aortic diameter of 5.5 cm or greater is not a predictor

Table 2. Presenting symptoms and Imaging findings R1 N (%) Overall Descending Descending P Value R2 < 5.5 cm > 5.5 cm R3 Presenting Hemodynamics R4 Hypertensive 396 (65.9) 344 (70.1) 52 (47.3) <.001 2 Normotensive 178 (29.6) 128 (26.1) 50 (45.5) <.001 R5 Hypotensive 20 (3.3) 16 (3.3) 4 (3.6) .842 R6 Shock 7 (1.2) 3 (0.6) 4 (3.6) .008 R7 Complicated 237 (38.7) 178 (35.6) 59 (52.2) .001 R8 CXR Findings R9 Normal 140 (25.5) 129 (29.1) 11 (10.6) <.001 Widened mediastinum 249 (46.2) 190 (43.5) 59 (57.8) .009 R10 Abnormal aortic contour 244 (46.2) 187 (43.8) 57 (56.4) .022 R11 Displacement/calc. of aorta 47 (9.1) 30 (7.2) 17 (17.2) .002 R12 Pleural effusion 84 (16.1) 61 (14.5) 23 (23.0) .037 R13 Additional imaging findings Peri-aortic hematoma 90 (16.3) 62 (13.7) 28 (28.0) <0.001 R14 Patent false lumen 230 (47.9) 199 (50.1) 31 (37.3) .034 R15 Partial false lumen thrombosis 175 (36.5) 131 (33.0) 44 (53.0) .001 R16 Complete thrombosis 75 (15.6) 67 (16.9) 8 (9.6) .099 R17 Mean descending aorta diameter 4.40 (±1.35) 3.90 (±0.74) 6.62 (±1.22) <.001 Site origin LSA 335 (56.0) 287 (58.7) 48 (44.0) .005 R18 Site origin desc aorta 135 (22.6) 98 (20.0) 37 (33.9) .002 R19 Site origin abdominal 21 (3.5) 18 (3.7) 3 (2.8) .634 R20 CXR: chest X-rays, LSA: left subclavian artery. Complicated includes those with mesenteric R21 ischemia, acute renal failure, limb ischemia, spinal cord ischemia, shock, periaortic hematoma, R22 recurrent pain, refractory pain, or refractory hypertension. R23 R24 Management and Outcomes R25 Medical management was offered to 72.0% of patients with an aortic diameter < 5.5 cm compared R26 with 59.5% of patients with an aortic diameter ≥ 5.5 cm (p=.009). Surgical management was R27 utilized more frequently in patients with an aortic diameter ≥ 5.5 cm (39.6% vs. 20.6%, p<.001), R28 while patients with an aortic diameter < 5.5 cm were more likely to undergo endovascular R29 management (7.4% vs. 0.9%, p=.008). R30 Increasing diameter was associated with increasing risk of in-hospital mortality (figure 2). Overall, R31 the in-hospital mortality was 6.6% for patients with an aortic diameter < 5.5 cm, and 23.0% R32 for patients with an aortic diameter ≥ 5.5cm (p<.001). The mean diameter of non-survivors was R33 5.22 cm, compared with 4.32 cm for survivors (p<.001). Among patients managed medically, R34 the in-hospital mortality was 4.4% vs. 25.4% for patients with an aortic diameter < and ≥ 5.5cm, R35 respectively (p<.001). The mortality did not differ significantly after surgical or endovascular R36 management between both groups (table 4). R37 R38 R39

23 Chapter 2

R1 Independent predictors for mortality among patients with an aortic diameter < 5.5 cm were: R2 hypotension and/or shock (OR 8.6, p=.001), acute renal failure (OR 3.6, p=.014), mesenteric R3 ischemia/infarction (OR 5.4, p=.012) and periaortic hematoma (OR 3.4, p=.019, table 4). R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 Figure 2. In-hospital mortality related to the descending aortic diameter R22 The in-hospital mortality was 6.6% for patients with an aortic diameter < 5.5 cm, and 23.0% for patients with an aortic diameter ≥ 5.5cm (p<.001) R23 R24 Table 3. In-hospital Management and Outcomes R25 N (%) Overall Descending Descending P Value R26 < 5.5 cm > 5.5 cm R27 Definitive Management Medical 426 (69.7) 360 (72.0) 66 (59.5) .009 R28 Surgery 147 (24.1) 103 (20.6) 44 (39.6) <.001 R29 Endovascular 38 (6.2) 37 (7.4) 1 (0.9) .008 R30 In-hospital Mortality 53 (9.6) 30 (6.6) 23 (23.0) <.001 R31 Medical 29 (7.7) 14 (4.4) 15 (25.4) <.001 Surgery 20 (14.5) 12 (12.2) 8 (20.0) .240 R32 Endovascular 4 (11.1) 4 (11.4) 0 (0) .732 R33 R34 R35 R36 R37 R38 R39

24 Aortic diameter of 5.5 cm or greater is not a predictor

Table 4. Independent predictors of death in patients with an aortic diameter < 5.5cm R1 Variable Odds ratio 95% CI P Value R2 Hypotension/shock 8.60 2.30 – 32.04 .001 R3 Acute renal failure 3.56 1.28 – 9.87 .014 R4 Mesenteric ischemia 5.42 1.45 – 20.28 .012 2 Peri-aortic hematoma 3.40 1.22 – 9.49 .019 R5 R6 CI: confidence interval R7 R8 DISCUSSION R9 R10 The risks of aortic dissection or rupture are thought to increase with increasing aortic diameter, R11 and for thoracic aortic aneurysms larger than 6 cm, the annual risk of aortic dissection, rupture R12 and/or death is about 15%.5-7 The natural history analysis of Elefteriades and colleagues revealed R13 a sharp hinge point of risk for descending thoracic aortas of 7 cm or more, where aortic dissection R14 or rupture becomes more likely. Current guidelines therefore recommend elective descending R15 thoracic aortic intervention in cases of an aortic diameter larger than 5.5 cm or 6.0, in an effort R16 to prevent life-threatening complications.5-9 R17 Our IRAD data show that over 80% of patients with acute type B aortic dissection in IRAD R18 had an aortic diameter < 5.5 cm at presentation. On this basis, elective surgery would not have R19 been indicated to prevent aortic B dissection based on current guidelines. Elefteriades showed R20 that the annual incidence of thoracic aortic rupture gradually increased from 0% to 3.6% in R21 patients with aortic diameters of 3.5 cm and 6.0 cm, respectively, while the incidence of aortic R22 dissection only increased from 2.2% to 3.7% for patients with similar aortic measurements.5 R23 Therefore, the relationship between increasing aortic diameter and the risk of aortic rupture is R24 far more linear than that for aortic dissection. This calls to mind the notion that fundamentally, R25 rupture and dissection presumably have different triggers, and current science is inadequate for R26 risk assessment and prevention. The median diameter of ABAD patients in IRAD was 4.1 cm, R27 and it was surely even smaller when the acute event occurred, since the aortic measurements R28 reported in the registry were obtained after acute dissection. These aortic diameters are also R29 considerably smaller than those reported for ruptured descending thoracic aortas, which usually R30 range between 6 and 8 cm.5-8, 11-13 R31 Although acute type B aortic dissection seems to occur at considerably smaller size than the R32 current threshold for elective descending thoracic aortic repair, lowering the cut-off for aortic R33 intervention is certainly not the solution for this problem. It is likely that many affected individuals R34 in the world have a descending aorta larger than 4 cm in diameter. Yet, we believe that only a R35 small fraction will ultimately dissect. The medical, emotional and financial risk of intervening on R36 this population would be substantial and unpredictable. Before eventually genetic, proteomic, R37 and/or imaging profiles help to identify patients at high risk, a conservative approach is justified. R38 R39

25 Chapter 2

R1 Predicting acute type B aortic dissection appears to be more difficult than predicting descending R2 thoracic aortic rupture, and other risk factors than increasing aortic diameter may play an R3 important role in the development of acute type B aortic dissection. Identifying the patients R4 with diameters smaller than 5.5 cm that are at risk for developing acute aortic events remains R5 essential. ABAD patients with a diameter smaller than 5.5 cm, presented more frequently with R6 hypertension, while the overall prevalence of pre-existing hypertension prior to dissection was R7 similar in both diameter groups. In this context, poor blood pressure control may play a role in R8 the etiology of ABAD in patients with an aortic diameter smaller than 5.5 cm. R9 In patients with Marfan’s syndrome, the risk of aortic dissection and rupture, in particular of R10 the aortic root, is increased,5, 6, 14 and the thresholds for prophylactic aortic replacement are R11 typically lower for this specific patient group.5, 6, 14-17 In the IRAD ABAD cohort, the incidence R12 of Marfan’s syndrome was not increased among the patients with aortic diameter < 5.5 cm, but R13 the mean aortic diameter tended to be even slightly larger in patients with Marfan’s syndrome R14 than in others (4.68 cm vs. 4.32 cm). Though Marfans did not develop type B aortic dissection R15 at smaller aortic diameters than non-Marfans in IRAD, conclusions are difficult to draw with a R16 limited sample size. R17 It remains unclear if the patients that develop type B aortic dissections in aortas smaller than 5.5. R18 cm, are different from others. IRAD cannot be used to investigate which patients are at risk of R19 developing acute type B aortic dissection, because of the observational nature of this registry and R20 the absence of a disease-free control group. Several inherited connective tissue disorders are known R21 to affect the integrity of the arterial wall, resulting in increased risks of aortic dissection, aneurysm R22 formation and rupture, such as Marfan’s syndrome, Ehlers-Danlos’ syndrome type IV, Turner’s R23 syndrome, and other less well known connective tissue disorders that have been associated with R24 aortic dissection.18-21 Recently, research has focussed on the genetic and proteomic background R25 of thoracic aortic disease, and it seems that up to 20% of patients with thoracic aortic aneurysms R26 or dissections have a family history of thoracic aortic disease without being affected by any R27 syndrome.5, 22, 23 Research projects on aortic diseases, such as the National Registry of Genetically R28 Triggered Thoracic Aortic Aneurysms and Cardiovascular Conditions (GenTAC), have been R29 established to further investigate genetically triggered thoracic aortic disease and the optimal R30 clinical management of these patients.24 Such initiatives may improve the current knowledge on R31 individualized treatment of the descending thoracic aorta with a diameter smaller than 5.5 cm, R32 who are at high risk for acute type B aortic dissection, and require prophylactic thoracic aortic R33 medical treatment and/or intervention to prevent its development. R34 Though the majority of patients in IRAD develop acute type B aortic dissections in aortas smaller R35 than 5.5 cm, complicated ABAD was more common among patients with an aortic diameter R36 ≥ 5.5 cm. In particular periaortic hematoma, suggesting “contained” aortic rupture, was more R37 common among these patients, who were subsequently subjected to surgery. The increased R38 incidence of complicated ABAD among the patients with an aortic diameter larger than 5.5 cm, R39 paralleled with increased mortality in this group.

26 Aortic diameter of 5.5 cm or greater is not a predictor

The present study has several strengths as well as limitations. The IRAD database contains the R1 largest series of patients with acute type B aortic dissection to date, which provided a unique R2 opportunity to investigate the relationship between aortic diameter and ABAD. Aortic size R3 R4 measurements were obtained after acute dissection and thus would tend to overestimate size, 2 which further supports the inaccuracy of aortic diameter as predictor of ABAD. Unfortunately, R5 IRAD does not have a disease-free control group, and therefore the data cannot be used to R6 investigate which patients are at risk of developing acute type B aortic dissection. Furthermore, R7 data regarding blood pressure control prior to the event, which could potentially play an R8 important role in the etiology of dissection, were unavailable in IRAD. R9 R10 R11 CONClUSION R12 R13 The vast majority of patients with ABAD present with a descending aortic diameter smaller R14 than 5.5 cm, and miss the threshold for elective descending thoracic aortic repair. The findings R15 of the present study suggest that descending aortic diameter is not a useful parameter to predict R16 acute type B dissection. In order to prevent acute type B aortic dissection, further natural history R17 studies are needed, as well as research on genetic predisposition for thoracic aortic disease. These R18 may reveal other risk factors for aortic dissection besides increasing aortic diameter, resulting in R19 better medical and/or interventional criteria for prophylactic thoracic aortic repair. R20 R21 funding sources R22 IRAD is supported by grants from the University of Michigan Health System, the Varbedian R23 Fund for Aortic Research, the Mardigian Foundation and Gore Medical Inc (Flagstaff, Arizona). R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

27 Chapter 2

R1 REfERENCES R2 R3 (1) Anagnostopoulos CE, Prabhakar MJ, Kittle CF. Aortic dissections and dissecting aneurysms. Am R4 J Cardiol 1972;30:263-73. R5 (2) Wheat MW, Jr. Acute dissecting aneurysms of the aorta: diagnosis and treatment--1979. Am R6 Heart J 1980;99:373-87. R7 (3) Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection: R8 increasing prevalence and improved outcomes reported in a nationwide population-based study R9 of more than 14,000 cases from 1987 to 2002. Circulation 2006;114:2611-8. R10 (4) Verhoye JP, Miller DC, Sze D, Dake MD, Mitchell RS. Complicated acute type B aortic R11 dissection: midterm results of emergency endovascular stent-grafting. J Thorac Cardiovasc Surg R12 2008;136:424-30. R13 (5) Elefteriades JA. Natural history of thoracic aortic aneurysms: indications for surgery, and surgical R14 versus nonsurgical risks. Ann Thorac Surg 2002;74:S1877-S1880. R15 (6) Davies RR, Goldstein LJ, Coady MA, Tittle SL, Rizzo JA, Kopf GS, et al. Yearly rupture or R16 dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg R17 2002;73:17-27. R18 (7) Coady MA, Rizzo JA, Hammond GL, Kopf GS, Elefteriades JA. Surgical intervention criteria R19 for thoracic aortic aneurysms: a study of growth rates and complications. Ann Thorac Surg R20 1999;67:1922-6. R21 (8) Lobato AC, Puech-Leao P. Predictive factors for rupture of thoracoabdominal aortic aneurysm. J R22 Vasc Surg 1998;27:446-53. R23 (9) Svensson LG, Kouchoukos NT, Miller DC, Bavaria JE, Coselli JS, Curi MA, et al. Expert R24 consensus document on the treatment of descending thoracic aortic disease using endovascular R25 stent-grafts. Ann Thorac Surg 2008;85:S1-41. R26 (10) Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, et al. The R27 International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. R28 JAMA 2000;283:897-903. R29 (11) Juvonen T, Ergin MA, Galla JD, Lansman SL, Nguyen KH, McCullough JN, et al. Prospective R30 study of the natural history of thoracic aortic aneurysms. Ann Thorac Surg 1997;63:1533-45. R31 (12) Crawford ES, Hess KR, Cohen ES, Coselli JS, Safi HJ. Ruptured aneurysm of the descending R32 thoracic and thoracoabdominal aorta. Analysis according to size and treatment. Ann Surg R33 199;213:417-25. R34 (13) Girardi LN, Krieger KH, Altorki NK, Mack CA, Lee LY, Isom OW. Ruptured descending and R35 thoracoabdominal aortic aneurysms. Ann Thorac Surg 2002;74:1066-70. R36 (14) David TE, Armstrong S, Maganti M, Colman J, Bradley TJ. Long-term results of aortic valve- R37 sparing operations in patients with Marfan syndrome. J Thorac Cardiovasc Surg 2009;138:859- R38 64. R39

28 Aortic diameter of 5.5 cm or greater is not a predictor

(15) Treasure T. Cardiovascular surgery for Marfan syndrome. Heart 2000;84:674-8. R1 (16) Dean JC. Management of Marfan syndrome. Heart 2002;88:97-103. R2 (17) Gott VL, Greene PS, Alejo DE, Cameron DE, Naftel DC, Miller DC, et al. Replacement of the R3 R4 aortic root in patients with Marfan’s syndrome. N Engl J Med 1999;340:1307-13. 2 (18) Groenink M, Rozendaal L, Naeff MS, Hennekam RC, Hart AA, Van Der Wall EE, et al. Marfan R5 syndrome in children and adolescents: predictive and prognostic value of aortic root growth for R6 screening for aortic complications. Heart 1998;80:163-9. R7 (19) Golledge J, Eagle KA. Acute aortic dissection. Lancet 2008;372:55-66. R8 (20) Loeys BL, Schwarze U, Holm T, Callewaert BL, Thomas GH, Pannu H, et al. Aneurysm R9 syndromes caused by mutations in the TGF-beta receptor. N Engl J Med 2006;355:788-98. R10 (21) Pepin M, Schwarze U, Superti-Furga A, Byers PH. Clinical and genetic features of Ehlers-Danlos R11 syndrome type IV, the vascular type. N Engl J Med 2000;342:673-80. R12 (22) Coady MA, Davies RR, Roberts M, Goldstein LJ, Rogalski MJ, Rizzo JA, et al. Familial patterns R13 of thoracic aortic aneurysms. Arch Surg 1999;134:361-7. R14 (23) Albornoz G, Coady MA, Roberts M, Davies RR, Tranquilli M, Rizzo JA, et al. Familial thoracic R15 aortic aneurysms and dissections--incidence, modes of inheritance, and phenotypic patterns. R16 Ann Thorac Surg 2006;82:1400-5. R17 (24) Eagle KA. Rationale and design of the National Registry of Genetically Triggered Thoracic Aortic R18 Aneurysms and Cardiovascular Conditions (GenTAC). Am Heart J 2009;157:319-26. R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

29 Chapter 2

R1 APPENDIX 1 R2 R3 IRAD co-principal investigators R4 Kim A. Eagle, MD, University of Michigan, Ann Arbor, Michigan; Eric M. Isselbacher, MD, R5 Massachusetts General Hospital, Boston, Massachusetts; Christoph A. Nienaber, MD, University R6 of Rostock, Rostock, Germany. R7 R8 IRAD co-investigators R9 Eduardo Bossone, MD, National Research Council, Lecce, Italy; Alan Braverman, MD, R10 Washington University School of Medicine, St. Louis, Missouri; Stefanos Demertzis, MD, R11 Cardiocentro Ticino, Lugano, Switzerland; Giuseppe DiBenedetto, MD, San Giovanni e Ruggi, R12 , Italy; Mark Ehrlich, MD, University of Vienna, Vienna, Austria; Arturo Evangelista, R13 MD, Hospital General Universitari Vall d’Hebron, Barcelona, Spain; Rossella Fattori, MD, R14 University Hospital S. Orsola, Bologna, Italy; James Froehlich, MD, and Thomas Tsai, MD, R15 University of Michigan, Ann Arbor, Michigan; Dan Gilon, MD, Hadassah University Hospital, R16 Jerusalem, Israel; Alan Hirsch, MD, and Kevin Harris, MD, Minneapolis Heart Institute, R17 Minneapolis, Minnesota; G. Chad Hughes, MD, Duke University, Durham, North Carolina; R18 Stuart Hutchison, MD, St. Michael’s Hospital, Toronto, Ontario, Canada; James L. Januzzi, R19 MD, Massachusetts General Hospital, Boston, Massachusetts; Alfredo Llovet, MD, Hospital R20 Universitario 12 de Octubre, Madrid, Spain; Truls Myrmel, MD, Tromsø University Hospital, R21 Tromsø, Norway; Peter Oberwalder, MD, Medical School Graz, Graz, Austria; Patrick O’Gara, R22 MD, and Joshua Beckman, MD, Brigham and Women’s Hospital, Boston, Massachusetts; Jae K. R23 Oh, MD, Mayo Clinic, Rochester, Minnesota; Linda A. Pape, MD, University of Massachusetts R24 Hospital, Worcester, Massachusetts; Reed Pyeritz, MD, University of Pennsylvania School of R25 Medicine, Philadelphia, Pennsylvania; Udo Sechtem, MD, and Gabriel Meinhardt, MD, Robert- R26 Bosch Krankenhaus, Stuttgart, Germany; P. Gabriel Steg, MD, Hôpital Bichat, , France; R27 Toru Suzuki, MD, University of Tokyo, Tokyo, Japan; Santi Trimarchi, MD, IRCCS Policlinico R28 San Donato, San Donato, Italy. R29 R30 Data management and biostatistical support R31 Daniel Montgomery, MS, and Elise Woznicki, University of Michigan, Ann Arbor, Michigan. R32 R33 R34 R35 R36 R37 R38 R39

30 Chapter 3

Acute Type B Aortic Dissection in the Absence of Aortic Dilatation

Journal of Vascular Surgery (Accepted)

Santi Trimarchi,1 MD, Frederik H.W. Jonker,2 MD, PhD, James B. Froehlich,3 MD, Gilbert R. Upchurch,4 MD, Frans L. Moll,5 MD, PhD, Bart E. Muhs,2 MD, PhD, Vincenzo Rampoldi,1 MD, Himanshu J. Patel,3 MD, Kim A. Eagle,3 MD.

On behalf of the International Registry of Acute Aortic Dissection (IRAD) investigators

1 Policlinico San Donato IRCCS, San Donato Milanese, Italy 2 Yale University School of Medicine, New Haven Conn, USA 3 University of Michigan Health System, Ann Arbor, Mich, USA 4 University of Virginia, Charlottesville, Virg, USA 5 University Medical Center Utrecht, the Netherlands Chapter 3

R1 AbSTRACT R2 R3 background R4 Increasing aortic diameter is thought to be an important risk factor for acute type B aortic R5 dissection (ABAD). However, some patients develop ABAD in the absence of aortic dilatation. In R6 this report, we sought to characterize ABAD patients who presented with a descending thoracic R7 aortic diameter < 3.5 cm. R8 R9 Methods R10 We categorized 613 ABAD patients enrolled in the International Registry of Acute Aortic R11 Dissection (IRAD) from 1996 to 2009 according to the aortic diameter < 3.5 cm (group 1), and R12 ≥ 3.5 cm (group 2). Demographics, clinical presentation, management and outcomes of the two R13 groups were compared. R14 R15 Results R16 Overall, 21.2% (n=130) had an aortic diameter < 3.5 cm. Patients in group 1 were younger (60.5 R17 vs. 64.0 years, P=.015) and more frequently female (50.8% vs. 28.6%, P<.001). They presented R18 more often with diabetes (10.9% vs. 5.9%, P=.050), history of catheterization (17.0% vs. 6.7%, R19 P=.001) and CABG (9.7% vs. 3.4%, P=.004). Marfan’s syndrome was equally distributed in the R20 two groups. The overall in-hospital mortality did not differ between group 1 and 2 (7.6% vs. R21 10.1%, P=.39). R22 R23 Conclusions R24 About one fifth of patients with acute type B aortic dissection do not present with any aortic R25 dilatation. These patients are more frequently females and younger, when compared with patients R26 with aortic dilatation. This report is an initial investigation to clinically characterize this cohort, R27 and further research is needed to identify risk factors for aortic dissection in the absence of aortic R28 dilatation. R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

32 Acute Type B Aortic Dissection in the Absence of Aortic Dilatation

INTRODUCTION R1 R2 Acute aortic dissection is a cardiovascular emergency, which is associated with high mortality R3 and morbidity.1-4 Increasing aortic diameter is thought to be an important risk factor of acute R4 aortic dissection and rupture,5-8 and prophylactic descending thoracic aortic repair is generally R5 recommended for an aortic diameter larger than 5.5 or 6.0 cm.4-9 However, a recent IRAD R6 study revealed that the majority of patients with acute type B aortic dissection (ABAD) appear 3 R7 to develop dissection in aortas smaller than this threshold.10 Closer evaluation of this group R8 suggested that a considerable number of patients present ABAD in the absence of aortic dilatation R9 (figure 1). It is currently unclear how many patients develop ABAD without aortic enlargement, R10 who these patients are, and why they develop aortic dissection. Because these patients do not R11 have any indications for prophylactic descending thoracic aortic repair prior to dissection, it is R12 important to characterize this group. The purpose of this study was to investigate ABAD patients R13 with a descending thoracic aortic diameter smaller than 3.5 cm. R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 Figure 1. Acute type B aortic dissection in the absence of aortic dilatation R37 R38 R39

33 Chapter 3

R1 METHODS R2 R3 Patient selection and data collection R4 IRAD is an ongoing multi-national multi-center registry that enrolls patients with acute aortic R5 dissection at 24 large referral centers, the inception and structure of IRAD has been described R6 previously.11 All patients presenting with ABAD that were enrolled in the IRAD registry between R7 1996 and September 2009 were selected for analysis. Only patients of whom the descending R8 aortic diameter at presentation was available were included for this evaluation. In total, we R9 identified 613 patients with ABAD which were included for analysis. Patients were categorized R10 according to an aortic diameter < 3.5 cm (group 1), and ≥ 3.5 cm (group 2), and the two groups R11 were compared regarding demographics, clinical presentation, management and outcomes. R12 The maximum descending aortic diameters were measured by computed tomography (CTA), R13 transesophageal echocardiography (TEE), and/or magnetic resonance imaging (MRI) at the time R14 of presentation. If patients underwent multiple imaging studies, the study that reflected the R15 largest aortic diameter was selected for analysis. The maximum aortic diameter was measured at R16 cross-sectional images perpendicular to the long axis of the descending aorta. In IRAD, all aortic R17 diameter measurements were obtained after aortic dissection had occurred. R18 R19 Statistical analysis R20 Data analysis was performed with the use of SPSS statistical analysis software (SPSS Inc., Chicago, R21 Ill.). The Chi-square test was used for comparing categorical variables between patients with an R22 aortic diameter <3.5 cm and ≥3.5 cm; the Student’s t-test was used for comparing continuous R23 variables between both diameter groups. Summary statistics are presented as frequencies and R24 percentages for categorical variables and mean ±SD for continuous variables. In all cases, missing R25 data were not defaulted to negative, and denominators reflect only cases reported. A P value < R26 .05 was considered significant. The authors had full access to and took full responsibility for the R27 integrity of the data. All authors have read and agree to the manuscript as written. R28 R29 R30 RESUlTS R31 R32 Aortic Measurements R33 The mean aortic diameter at the level of ABAD was 4.40 ±1.35 cm (median 4.1 cm, range R34 2.1 to 13.0 cm). Acute type B aortic dissection in the absence of aortic dilatation, defined as a R35 descending aortic diameter smaller than 3.5 cm, was present in 21.2% (n=130) of patients in R36 IRAD. The mean aortic diameter in group 1 was 2.97 ±0.32 cm, compared with 4.79 ±1.26 R37 cm in group 2 (P<.001). In females, the mean aortic diameter in the overall group was 4.30 cm R38 compared with 4.46 cm in males (P=.199). R39

34 Acute Type B Aortic Dissection in the Absence of Aortic Dilatation

Demographics and Patient History R1 The mean age of ABAD patients in IRAD was 63.9 ±14.0 years, and 67.7% were males. The mean R2 age of patients without aortic dilatation was 60.5 years, compared with 64.0 years in patients R3 with an aortic diameter larger than 3.5 cm (P=.015). Among patients in group 1, 50.8% were R4 females, compared with 28.6% among patients in group 2 (P<.001). After stratification for aortic R5 diameter, ABAD patients were predominantly males in all aortic diameter categories, except for R6 those in the aortic diameter categories below 3.5 cm (figure 2), in which the distribution of males 3 R7 and females was approximately equal. Biometric characteristics such as height and weight did not R8 differ significantly between patients with and without aortic enlargement (table 1). R9 Patients without aortic dilatation were more likely to report a history of diabetes mellitus (10.9% R10 vs. 5.9%, P=.050), while pre-existing hypertension (73.6% vs. 81.5%, P=.050) was less common R11 in this cohort. Marfan’s syndrome and bicuspid aortic valve were present in 4.3% and 1.9% of R12 all ABAD patients, respectively, and prevalences did not differ between both diameter groups. R13 Prior coronary artery bypass grafting (CABG) and cardiac catheterization were more frequently R14 represented in patients without aortic dilatation. In this group, an iatrogenic cause of dissection R15 was also more present (5.6% vs. 1.1%, P=.002) (table 1). R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 Figure 2. Distribution of gender in different aortic diameter groups R35 ABAD patients in IRAD were predominantly males in all aortic diameter categories, except for those in the aortic diameter categories below 3.5 cm, in which the distribution of males and females was approximately R36 equal. R37 R38 R39

35 Chapter 3

R1 Table 1. Demographics and clinical history R2 Aortic diameter Aortic diameter R3 < 3.5 cm > 3.5 cm R4 n % n % P value 130 (21.2) 483 (78.8) --- R5 Demographics R6 Age (y) 60.5 ± 14.8 64.0 ± 13.3 .015 R7 Female 66 (50.8) 138 (28.6) <.001 R8 Biometrics Height (m) 1.68 (±14.6) 1.69 (±12.1) .758 R9 Weight (kg) 79.1 (±21.5) 87.1 (±20.7) .188 R10 BMI (m/kg2) 28.1 (±6.7) 29.5 (±5.7) .486 R11 History R12 Marfan’s syndrome 6 (4.6) 20 (4.2) .840 Bicuspid Aortic Valve 2 (1.9) 7 (1.9) .966 R13 Hypertension 95 (73.6) 391 (81.5) .050 R14 Atherosclerosis 42 (32.6) 165 (34.8) .633 R15 Diabetes 14 (10.9) 28 (5.9) .050 R16 Prior aortic repair 11 (8.9) 73 (15.5) .058 R17 Prior CABG 12 (9.7) 16 (3.4) .004 Prior catheterization 18 (17.0) 24 (6.7) .001 R18 Iatrogenic dissection 7 (5.6) 5 (1.1) .002 R19 CABG coronary artery bypass grafting. Data regarding height were available 8.2% (n=50), weight for 11.1% R20 (n=68), BMI (body mass index) for 8.0% (n=49) and BSA (body surface area) for 8.0% of patients (n=49). R21 R22 Presentation and Diagnostic findings R23 ABAD patients without aortic dilatation presented less frequently with chest pain (64.0% vs. R24 73.4%, P=.038) or abdominal pain (31.1% vs. 42.1%, P=.028) than patients with a larger aortic R25 diameter. Syncope at presentation was more common in patients without aortic enlargement R26 (7.2% vs. 2.3%, P=.007). Complicated ABAD, defined as presence of malperfusion syndromes, R27 aortic rupture, recurrent or refractory pain and/or refractory hypertension, was present in 31.5% R28 of patients without aortic dilatation, compared with 40.6% of patients with an aortic diameter R29 ≥ 3.5 cm (P=.060) (table 2). R30 A completely normal chest X-ray was seen in 43.5% in group 1 compared with 20.8% in group 2 R31 (P<.001, table 3). Additional imaging showed a patent false lumen in 56.7% of patients without R32 aortic dilatation and in 45.5% of the patients with an aortic diameter > 3.5 cm (P=.042). R33 R34 R35 R36 R37 R38 R39

36 Acute Type B Aortic Dissection in the Absence of Aortic Dilatation

Table 2. Symptoms and signs of aortic dissection R1 Aortic diameter Aortic diameter R2 < 3.5 cm > 3.5 cm R3 n % n % P value R4 Presentation Chest Pain 80 (64.0) 348 (73.4) .038 R5 Abdominal pain 38 (31.1) 196 (42.1) .028 R6 Shock 2 (1.6) 5 (1.1) .628 3 R7 Syncope 9 (7.2) 11 (2.3) .007 R8 Complications 41 (31.5) 196 (40.6) .060 R9 Limb ischemia 9 (7.6) 36 (8.1) .869 Any pulse deficit 22 (18.8) 66 (15.5) .389 R10 Visceral ischemia 8 (6.7) 16 (3.6) .132 R11 Acute renal failure 14 (11.9) 52 (11.6) .938 R12 Spinal cord ischemia 1 (0.9) 11 (2.5) .290 R13 Periaortic hematoma 11 (9.4) 79 (18.2) .023 Recurrent pain 11 (44.0) 51 (33.8) .322 R14 Refractory pain 1 (4.0) 20 (13.2) .187 R15 Refractory hypertension 1 (4.0) 10 (6.6) .616 R16 R17 Table 3. Diagnostic findings of aortic dissection R18 Aortic diameter Aortic diameter R19 < 3.5 cm > 3.5 cm R20 n % n % P value R21 Chest X-rays findings Normal 50 (43.5) 90 (20.8) <.001 R22 Widened mediastinum 39 (34.2) 210 (49.4) .004 R23 Abnormal aortic contour 33 (29.5) 211 (50.7) <.001 R24 Additional imaging findings R25 Intramural hematoma 27 (20.9) 89 (18.6) .547 R26 Patent false lumen 59 (56.7) 171 (45.5) .042 Partial false lumen thrombosis 31 (29.8) 144 (38.3) .111 R27 False lumen thrombosis 14 (13.5) 61 (16.2) .492 R28 Mean descending aorta diameter 2.97 (±0.32) 4.79 (±1.26) <.001 R29 Entry tear located at LSA 76 (59.4) 259 (55.1) .388 R30 Entry tear located at descending aorta 22 (17.2) 113 (24.0) .100 R31 LSA left subclavian artery R32 R33 Management and Outcomes R34 Medical management alone was offered to 69.7% of all patients. Endovascular methods, either R35 endovascular stenting or fenestration, were more frequently offered to ABAD patients without R36 aortic dilatation (13.8% vs. 4.2%, P<.001), while surgical management was more common R37 among patients with an aortic diameter > 3.5 cm (table 4). The overall in-hospital mortality was R38 R39

37 Chapter 3

R1 9.6% (n=53), and the mortality did not differ significantly between both diameter groups (7.6% R2 vs. 10.1%, P=.398). The mortality rate after endovascular management of ABAD patients with a R3 normal aortic diameter was 23.5% (4 out of 18), while no patients with an aortic diameter ≥ 3.5 R4 cm died after endovascular management (P=.040). R5 R6 Table 4. In-hospital management and mortality R7 Aortic diameter Aortic diameter R8 < 3.5 cm > 3.5 cm R9 n (%) n (%) P value Definitive Management R10 Surgery 21 (16.2) 126 (26.2) .017 R11 Medical 91 (70.0) 335 (69.6) .938 R12 Endovascular 18 (13.8) 20 (4.2) <.001 R13 In-hospital Mortality 9 (7.6) 44 (10.1) .398 Surgery 1 (5.0) 19 (16.1) .306 R14 Medical 4 (4.9) 25 (8.4) .286 R15 Endovascular 4 (23.5) 0 (0.0) .040 R16 R17 R18 DISCUSSION R19 R20 Increasing aortic diameter is thought to be an important risk factor for acute aortic dissection R21 and rupture,5-8 however, about one fifth of acute type B aortic dissections occur in patients R22 without aortic enlargement. In this analysis, aortic diameter measurements were obtained after R23 aortic dissection had occurred, and therefore the original diameter may have been even smaller. R24 We observed remarkable differences between the two patient groups; most notably that women R25 accounted for half of the ABAD patients with a normal diameter, while the majority of the type R26 B dissections in enlarged aortas consisted of male patients. Male gender is a well known risk R27 factor for the development of various cardiovascular diseases, including acute aortic dissection, R28 and approximately 65% to 75% of admitted patients with ABAD are males.12-16 R29 It remains a mystery why patients sustaining ABAD in the absence of aortic dilatation are more R30 frequently females. Women are typically smaller in size than men, and the average female aorta R31 may be slightly smaller than the average male aorta. Consequently, a logical hypothesis for the R32 increased proportion of females among ABAD patients with a normal aortic diameter could be R33 that the female aorta may dissect at a smaller diameter, because a 3 cm aortic diameter may be R34 considerably increased compared with the normal aortic measurements in females. However, in R35 the overall IRAD cohort the mean aortic diameter of ABAD patients did not differ substantially R36 between males and females (4.5 cm vs. 4.3 cm). In addition, patients’ height did not differ R37 between those presenting with and without aortic dilatation. R38 R39

38 Acute Type B Aortic Dissection in the Absence of Aortic Dilatation

The increased proportion of females in the group without aortic enlargement may have other R1 explanations beyond anatomical reasons. The natural history of thoracic aortic disease has not R2 been well defined yet,17, 18 and in particular the influence of gender remains currently controversial. R3 In contrast to abdominal aortic aneurysms (AAA), for which men are affected predominantly, the R4 prevalence of thoracic aortic aneurysm (TAA) is thought to be more equal in women and men.19 R5 Women however typically present with aortic aneurysms and dissections at a greater age than R6 men.12, 19, 20 Moreover, female gender appears to be a risk factor for aortic aneurysm rupture.19-21 3 R7 Another remarkable finding was that ABAD patients without aortic dilatation were on average R8 3.5 years younger, which is especially interesting since women generally present with ABAD at a R9 higher age.12 It is unclear why ABAD without aortic dilatation occurred at a younger age, but this R10 may suggest a genetic or familial predisposition in some of these patients. Coady and colleagues R11 have investigated familial TAA, which accounted for at least 20% of patients with TAA, after R12 excluding Marfan patients.22, 23 They compared these nonsyndromic familial TAA patients, with R13 sporadic TAA patients and Marfan’s syndrome, and found that patients with familial TAA were R14 about 7 years younger than patients with sporadic TAA. Interestingly, the aortic growth rate was R15 highest for the familial TAA group in their analysis.22, 23 It is possible that a similar unknown R16 familial or genetic predisposition exists in some young ABAD patients without aortic dilatation. R17 Marfan’s syndrome is an important risk factor for aortic dissection especially in young patients,5, R18 7 and thresholds for prophylactic aortic replacement are typically lower for this specific patient R19 group.5-7, 24-26 In ABAD patients without aortic dilatation, we did not find an increased prevalence R20 of Marfans, perhaps because Marfan’s syndrome patients typically suffer from aortic root R21 enlargement and dissection, instead of descending thoracic aortic disease.24-26 Other inherited R22 disorders, including Ehlers-Danlos syndrome type IV, Turner’s syndrome, and other less well R23 known connective tissue disorders are known to affect the integrity of the arterial wall, which R24 could result in aortic dissection.4, 27-31 IRAD does not contain data regarding inherited disorders R25 other than Marfan’s syndrome, the most common connective tissue disease, and it is possible R26 that the normal diameter group had an increased prevalence of known, or currently unknown, R27 connective tissue disorders. The genes that predispose to such connective tissue disorders and R28 familial forms of thoracic aortic disease have only recently begun to be identified. Research R29 initiatives such as the National Registry of Genetically Triggered Thoracic Aortic Aneurysms and R30 Cardiovascular Conditions (GenTAC),32 have been established to further investigate genetically R31 triggered thoracic aortic disease, and these registries may provide more insights about patients R32 who develop ABAD in the absence of aortic dilatation. R33 Furthermore, iatrogenic dissection was more common among patients that suffered ABAD R34 without aortic dilatation. It remains unclear why iatrogenic dissection was more common among R35 this group, although a history of CABG and catheterization was more frequently present among R36 ABAD patients without aortic dilatation. If iatrogenic aortic dissection occurs, it often arises R37 during cardiac surgery or catheterization and the incidence of iatrogenic type A dissection during R38 cardiac surgery or catherization is approximately 0.20% and 0.024%, respectively.33, 34 R39

39 Chapter 3

R1 Although investigating the management and outcomes was not the purpose of this study, we R2 observed that endovascular stenting and fenestration were more frequently offered to patients R3 with a normal aortic diameter, while those with aortic dilatation more frequently underwent R4 surgery, perhaps because large dissected aortas more frequently have inadequate aortic landing R5 zones for stenting. The in-hospital mortality after endovascular management appeared to be R6 higher for the patients with normal aortic diameters than for those with aortic dilatation, R7 however, the number of patients treated with endovascular management was too small to draw R8 strong conclusions. R9 The IRAD database contains the largest series of patients with ABAD, which provided a unique R10 opportunity to clinically characterize this subset of patients without aortic dilatation. As with all R11 observational studies, this investigation has limitations that must be kept in mind when the data R12 are interpreted. The aortic size measurements were obtained after acute dissection and thus the R13 actual aortic diameters prior to the event may be even smaller. IRAD does not have a disease- R14 free control group, and therefore the data cannot be used to investigate which individuals with a R15 normal aortic diameter are at risk for developing ABAD. R16 R17 R18 CONClUSION R19 R20 About one fifth of patients with ABAD do not have any aortic dilatation prior to aortic dissection. R21 This report is an initial attempt to clinically characterize this cohort, which revealed that these R22 patients are younger and more frequently women. Further research is needed to explore the R23 molecular and genetic predisposition of these patients. R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

40 Acute Type B Aortic Dissection in the Absence of Aortic Dilatation

REfERENCES R1 R2 (1) Anagnostopoulos CE, Prabhakar MJ, Kittle CF. Aortic dissections and dissecting aneurysms. Am R3 J Cardiol 1972;30:263-73. R4 (2) Wheat MW, Jr. Acute dissecting aneurysms of the aorta: diagnosis and treatment--1979. Am R5 Heart J 1980;99:373-87. R6 (3) Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection: 3 R7 increasing prevalence and improved outcomes reported in a nationwide population-based study R8 of more than 14,000 cases from 1987 to 2002. Circulation 2006;114:2611-8. R9 (4) Golledge J, Eagle KA. Acute aortic dissection. Lancet 2008;372:55-66. R10 (5) Elefteriades JA. Natural history of thoracic aortic aneurysms: indications for surgery, and surgical R11 versus nonsurgical risks. Ann Thorac Surg 2002;74:S1877-S1880. R12 (6) Coady MA, Rizzo JA, Hammond GL, Kopf GS, Elefteriades JA. Surgical intervention criteria R13 for thoracic aortic aneurysms: a study of growth rates and complications. Ann Thorac Surg R14 1999;67:1922-6. R15 (7) Davies RR, Goldstein LJ, Coady MA, Tittle SL, Rizzo JA, Kopf GS, et al. Yearly rupture or R16 dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg R17 2002;73:17-27. R18 (8) Lobato AC, Puech-Leao P. Predictive factors for rupture of thoracoabdominal aortic aneurysm. J R19 Vasc Surg 1998;27:446-53. R20 (9) Svensson LG, Kouchoukos NT, Miller DC, Bavaria JE, Coselli JS, Curi MA, et al. Expert R21 consensus document on the treatment of descending thoracic aortic disease using endovascular R22 stent-grafts. Ann Thorac Surg 2008;85:S1-41. R23 (10) Trimarchi S, Jonker FH, Hutchison S, Isselbacher EM, Pape LA, Patel HJ, et al. Descending R24 aortic diameter of 5.5 cm or greater is not an accurate predictor of acute type B aortic dissection. R25 J Thorac Cardiovasc Surg 2011;142:e101-e107. R26 (11) Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, et al. The R27 International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. R28 JAMA 2000;283:897-903. R29 (12) Nienaber CA, Fattori R, Mehta RH, Richartz BM, Evangelista A, Petzsch M, et al. Gender- R30 related differences in acute aortic dissection. Circulation 2004;109:3014-21. R31 (13) Neya K, Omoto R, Kyo S, Kimura S, Yokote Y, Takamoto S, et al. Outcome of Stanford type B R32 acute aortic dissection. Circulation 1992;86:II1-II7. R33 (14) Elefteriades JA, Hartleroad J, Gusberg RJ, Salazar AM, Black HR, Kopf GS, et al. Long-term R34 experience with descending aortic dissection: the complication-specific approach. Ann Thorac R35 Surg 1992;53:11-20. R36 (15) Umana JP, Lai DT, Mitchell RS, Moore KA, Rodriguez F, Robbins RC, et al. Is medical therapy R37 still the optimal treatment strategy for patients with acute type B aortic dissections? J Thorac R38 Cardiovasc Surg 2002;124:896-910. R39

41 Chapter 3

R1 (16) Estrera AL, Miller CC, Goodrick J, Porat EE, Achouh PE, Dhareshwar J, et al. Update on R2 outcomes of acute type B aortic dissection. Ann Thorac Surg 2007;83:S842-S845. R3 (17) Ince H, Nienaber CA. Etiology, pathogenesis and management of thoracic aortic aneurysm. Nat R4 Clin Pract Cardiovasc Med 2007;4:418-27. R5 (18) Isselbacher EM. Thoracic and abdominal aortic aneurysms. Circulation 2005;111:816-28. R6 (19) Clouse WD, Hallett JW, Jr., Schaff HV, Gayari MM, Ilstrup DM, Melton LJ, III. Improved R7 prognosis of thoracic aortic aneurysms: a population-based study. JAMA 1998;280:1926-9. R8 (20) Mureebe L, Egorova N, McKinsey JF, Kent KC. Gender trends in the repair of ruptured R9 abdominal aortic aneurysms and outcomes. J Vasc Surg 2010;51:9S-13S. R10 (21) Brown LC, Powell JT. Risk factors for aneurysm rupture in patients kept under ultrasound R11 surveillance. UK Small Aneurysm Trial Participants. Ann Surg 1999;230:289-96. R12 (22) Coady MA, Davies RR, Roberts M, Goldstein LJ, Rogalski MJ, Rizzo JA, et al. Familial patterns R13 of thoracic aortic aneurysms. Arch Surg 1999;134:361-7. R14 (23) Albornoz G, Coady MA, Roberts M, Davies RR, Tranquilli M, Rizzo JA, et al. Familial thoracic R15 aortic aneurysms and dissections--incidence, modes of inheritance, and phenotypic patterns. R16 Ann Thorac Surg 2006;82:1400-5. R17 (24) Treasure T. Cardiovascular surgery for Marfan syndrome. Heart 2000 Dec;84(6):674-8. R18 (25) Dean JC. Management of Marfan syndrome. Heart 2002;88:97-103. R19 (26) Gott VL, Greene PS, Alejo DE, Cameron DE, Naftel DC, Miller DC, et al. Replacement of the R20 aortic root in patients with Marfan’s syndrome. N Engl J Med 1999;340:1307-13. R21 (27) Loeys BL, Schwarze U, Holm T, Callewaert BL, Thomas GH, Pannu H, et al. Aneurysm R22 syndromes caused by mutations in the TGF-beta receptor. N Engl J Med 2006;355:788-98. R23 (28) Pepin M, Schwarze U, Superti-Furga A, Byers PH. Clinical and genetic features of Ehlers-Danlos R24 syndrome type IV, the vascular type. N Engl J Med 2000;342:673-80. R25 (29) Elefteriades JA. Perspectives on diseases of the thoracic aorta. Adv Cardiol 2004;41:75-86. R26 (30) Groenink M, Rozendaal L, Naeff MS, Hennekam RC, Hart AA, Van Der Wall EE, et al. Marfan R27 syndrome in children and adolescents: predictive and prognostic value of aortic root growth for R28 screening for aortic complications. Heart 1998;80:163-9. R29 (31) Mukherjee D, Eagle KA. Aortic dissection--an update. Curr Probl Cardiol 2005;30:287-325. R30 (32) Eagle KA. Rationale and design of the National Registry of Genetically Triggered Thoracic Aortic R31 Aneurysms and Cardiovascular Conditions (GenTAC). Am Heart J 2009;157:319-26. R32 (33) Januzzi JL, Sabatine MS, Eagle KA, Evangelista A, Bruckman D, Fattori R, et al. Iatrogenic aortic R33 dissection. Am J Cardiol 2002;89:623-6. R34 (34) Jonker FH, Schlosser FJ, Indes JE, Sumpio BE, Botta DM, Moll FL, et al. Management of type R35 A aortic dissections: a meta-analysis of the literature. Ann Thorac Surg 2010;89:2061-6. R36 R37 R38 R39

42 Chapter 4

Role of Age in Acute Type A Aortic Dissection Outcome: Report from The International Registry of Acute Aortic Dissection (IRAD)

Journal Thoracic Cardiovascular Surgery 2010 Oct;140(4):784-9.

Santi Trimarchi, MD,1 Kim A. Eagle, MD,2 Christoph A. Nienaber, MD,3 Vincenzo Rampoldi, MD,1 Frederik H. W. Jonker, MD,4 Carlo De Vincentiis, MD,1 Alessandro Frigiola, MD,1 Lorenzo Menicanti, MD,1 Thomas Tsai, MD,2 Jim Froehlich, MD,2 Arturo Evangelista, MD,5 Daniel Montgomery, MD,2 Eduardo Bossone, MD,6 Jeanna V. Cooper, MS,2 Jin Li, MS,2 Michael G. Deeb, MD,2 Gabriel Meinhardt, MD,7 Thoralf M. Sundt, MD,8 and Eric M. Isselbacher, MD,9

On behalf of the International Registry of Acute Aortic Dissection (IRAD) investigators

1 Policlinico San Donato IRCCS, San Donato Milanese, Italy 2 University of Michigan, Ann Arbor, Mich, USA 3 University of Rostock, Rostock, Germany 4 Yale University, New Haven, Conn, USA 5 Hospital General Universitari Vall d’Hebron, Barcelona, Spain 6 National Research Council, Lecce, Italy 7 Robert-Bosch Krankenhaus, Stuttgart, Germany 8 Mayo Clinic, Rochester, Min, USA 9 Massachusetts General Hospital, Boston, Mass, USA Chapter 4

R1 AbSTRACT R2 R3 Objective R4 The increasing life expectancy of the population will likely be accompanied by a rise in the R5 incidence of acute type A aortic dissection (AAD). However, because of an increased risk of R6 cardiac surgery in an elderly population, it is important to define when, if at all, the risks of aortic R7 repair outweigh the risk of death from unoperated AAD. R8 R9 Methods R10 We analyzed 936 patients with AAD enrolled in the International Registry of Acute Aortic R11 Dissection (IRAD) from 1996-2004. AAD patients were categorized according to patient age R12 by decade, and surgical vs. medical management, and outcomes of both management types were R13 investigated in the different age groups. R14 R15 Results R16 The rate of surgical aortic repair decreased progressively with age, whereas surgical mortality R17 significantly increased with age. Age ≥ 70 was an independent predictor for mortality (38.2 vs. R18 26.0%; p<0.0001, OR 1.73). The in-hospital mortality rate was significantly lower after surgical R19 management compared with medical management until the age of 80. For patients aged 80 to R20 90 years, the in-hospital mortality appeared to be lower after surgical management(37.9% vs. R21 55.2%, p=0.188), however this failed to reach clinical significance due to the limited patient R22 number in this age group. R23 R24 Conclusions R25 Although the surgical mortality significantly increased with increasing age, surgical management R26 was still associated with significantly lower in-hospital mortality rates compared with medical R27 management until the age of 80 years. Surgery may decrease the in-hospital mortality rate for R28 octogenarians with AAD and might be considered in all patients with AAD regardless of age. R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

44 Role of age in acute type A aortic dissection outcome

INTRODUCTION R1 R2 The increasing life expectancy of the population will be accompanied by a rise in the incidence R3 of cardiovascular disease including acute type A aortic dissection (AAD) (1). R4 The mortality rate of untreated AAD increases 1% to 2% every hour after first presentation, and R5 immediate surgical intervention is indicated after diagnosis of AAD. Currently, the in-hospital R6 mortality rate of AAD is 15% to 30% (2-6), and this outcome appears generally unimproved R7 over time (4). Increasing age has shown to be a strong independent predictor of in-hospital R8 mortality of cardiovascular interventions (7-10), including surgical repair of AAD (2;11;12). For R9 very elderly patients with AAD, medical management may be an alternative treatment option, 4 R10 and therefore it is important to define at what age, if at all, the risks of aortic repair outweigh the R11 risk of death from medically managed AAD. R12 The International Registry of Acute Aortic Dissection (IRAD) provides an opportunity to R13 investigate the role of age in AAD outcomes, and to determine if the management type of AAD R14 should depend on the patient’s age. We utilized the registry to evaluate outcome of patients with R15 AAD in after surgical and medical management in different age groups, and to define individual R16 treatment strategies for elderly AAD patients. R17 R18 R19 METHODS R20 R21 Patient selection R22 We analyzed all patients with acute type A aortic dissection (AAD) enrolled in the International R23 Registry of Acute Aortic Dissection (IRAD) from 1996 to 2004. IRAD is an ongoing international R24 multi-centre registry started in 1996 that includes consecutive patients with acute aortic R25 dissection at 24 large referral centres (IRAD centres; Appendix I), which rationale has previously R26 been described (13). Acute type A dissection was defined as any dissection that involved the R27 ascending aorta that was presented within 14 days after onset of symptoms. In total, 936 patients R28 with AAD were identified and included for analysis. All patients were categorized according to R29 patient age by decade, and surgical vs. medical management, and outcomes of both management R30 types were investigated in the different age groups. R31 R32 Data extraction R33 Data were collected on a standard questionnaire form developed by IRAD investigators. Data R34 collection included variables regarding demographics, history, clinical presentation, physical R35 findings, imaging study results, medical and surgical management, in-hospital clinical events, R36 length of stay, and hospital mortality. Completed data forms were submitted to the IRAD R37 coordinating centre at the University of Michigan. R38 R39

45 Chapter 4

R1 Statistical analysis R2 Summary statistics were presented as frequencies and percentages, mean + standard deviation, or R3 as a median and interquartile range. Missing data were not defaulted to negative and denominators R4 reflect only actual reported cases. Nominal variables were compared between patients < 70 years R5 and ≥ 70 years, and between medical and surgical groups, using the Chi-square test or two-sided R6 Fisher exact test. Iterative logistic modelling was performed to investigate independent predictors R7 of in-hospital mortality using likelihood ratio tests. Initial modelling implemented elements R8 marginally suggestive of an unadjusted association to in-hospital mortality (p<0.20). Variables R9 were reviewed for clinical significance prior to testing. Diagnostic routines (Hosmer-Lemeshow R10 test for lack of fit, change in deviance and residuals, and leverage indicators) were used for the R11 final model selection. AP value < 0.05 was considered significant. SAS 8.2 (SAS Institute, Cary, R12 NC) and SPSS 11.5 (SPSS Inc., Chicago, IL) were utilized for the analyses. R13 R14 R15 RESUlTS R16 R17 Demographics, patient history and presentation R18 The incidence of AAD was higher during th7 and 8th decades (figure 1). Out of all patients with R19 AAD, 32.2% (n=301) were 70 years of age and older. Men constituted 46.8% of the elderly, and R20 53.2% of the younger cohort (p<0.0001). A history of hypertension, atherosclerosis, prior aortic R21 aneurysm, diabetes mellitus, and prior cardiac surgery were more frequently present among R22 patients of 70 and older, while Marfan syndrome was only seen in the younger cohort (table 1). R23 Patients with AAD that were younger than 70 years presented more frequently with an abrupt R24 onset of pain and pulse deficit compared to patients 70 years of age and older; hypotension was R25 more often seen at presentation among the older patient cohort (table 1). R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

46 Role of age in acute type A aortic dissection outcome

R1 R2 R3 R4 R5 R6 R7 R8 R9 4 R10 R11 R12 R13 R14 R15 Figure 1. Age of Patients with AAD in IRAD R16 Age groups are labelled according to the upper age limit; e.g. age group 30 includes patients 20 years of age to 30 years. R17 R18 R19 Table 1. Demographics, Patient History, Presentation and Signs R20 Variable Patients age<70 Patients age ≥70 P-value R21 635 (67.8%) 301 (32.2%) R22 Demographics R23 Gender (male) 166 (53.2) 146 (46.8) <0.0001 Etiology and History R24 Marfan syndrome 45 (7.3) 0 (0) <0.0001 R25 Hypertension 410 (66.7) 219 (75.8) 0.0055 R26 Atherosclerosis 119 (19.5) 112 (38.9) <0.0001 R27 Prior aortic aneurysm 62 (10.2) 48 (16.7) 0.0054 R28 Prior aortic dissection 23 (3.8) 7 (2.4) 0.30 Diabetes 15 (2.5) 21 (7.4) 0.0006 R29 Prior cardiac surgery 77 (13.0) 66 (24.2) <0.0001 R30 CABG 30 (5.0) 27 (9.7) 0.0084 R31 Cardiac catheterization 37 (9.3) 38 (20.2) 0.0002 R32 Clinical Presentation and signs Hypotension or shock 157 (26.2) 85 (31.6) 0.10 R33 Abrupt onset of pain 520 (87.3) 217 (80.4) 0.01 R34 Hypotension 86 (14.5) 54 (20.3) 0.03 R35 Coma/altered consciousness 82 (13.9) 33 (11.7) 0.38 R36 Spinal cord ischemia 16 (2.7) 2 (0.7) 0.07 Any pulse deficit 176 (32.3) 63 (23.9) 0.01 R37 R38 CABG: coronary artery bypass grafting * Data regarding prior cardiac catheterization were missing in 38%. R39

47 Chapter 4

R1 In-hospital management R2 Out of all patients with AAD, 82.9% were managed surgically, 16.6% were managed medically, R3 and 0.5% underwent percutaneous stenting or fenestration. Surgery and medical therapy R4 were significantly more adopted in younger and older patients than 70, respectively (Surgery R5 management in pts age < 70 vs ≥ 70: 88.5% vs 71.1%, p< 0.0001; Medical management in pts R6 age < 70 vs ≥ 70: 10.9% vs 28.6%, P < 0.0001) (Table 2). R7 In IRAD AAD patients, the reasons listed for medical treatment only were advanced age, R8 intramural hematoma, severe co-morbidities, or refusal surgery by patients, families, or the R9 care team. The rate of surgical aortic repair decreased progressively with age, whereas the rate R10 of medical management increased with age (table 3, figure 2). The utilization of surgical and R11 medical management was equal for octogenarians, and medical treatment was more frequently R12 offered to the 5 AAD patients above 90 years of age (figure 2). R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 Figure 2. Management of AAD in different age groups R34 R35 R36 R37 R38 R39

48 Role of age in acute type A aortic dissection outcome

Table 2. In-hospital management and Mortality of AAD R1 Patients age <70 Patients age ≥70 P-value R2 635 (67.8%) 301 (32.2%) R3 Definitive management R4 Surgery 562 (88.5) 214 (71.1) <0.0001 Medical Treatment 69 (10.9) 86 (28.6) <0.0001 R5 Percutaneous 4 (0.6) 1 (0.3) NA R6 Initial medicine R7 Beta blocker 299 (51.7) 128 (47.4) 0.24 R8 Nitroprusside 182 (31.9) 67 (25.8) 0.08 R9 CA channel blocker 82 (14.6) 35 (13.8) 0.77 Mortality 165 (26.0) 115 (38.2) 0.0001 4 R10 Surgery mortality 119 (21.2) 66 (30.8) 0.005 R11 Medical mortality 44 (63.8) 48 (55.8) 0.316 R12 Percutaneous includes percutaneous stenting and fenestration, CA: calcium. R13 * Data regarding chronic medicine were missing in 30%. R14 R15 In-hospital outcomes R16 The overall in-hospital mortality was 23.8% among patients treated surgically vs. 59.3% among R17 patients managed medically (p<0.0001). The in-hospital mortality rate for patients managed R18 surgically increased with increasing age, while the in-hospital mortality rate of patients treated R19 medically remained roughly stable (table 3, figure 3). The in-hospital mortality rate was R20 significantly lower after surgical management compared with medical management until the age R21 of 80 (table 3). For patients aged 80 to 90 years, surgical management showed a trend to decrease R22 the in-hospital mortality rate compared with medical management, however this failed to reach R23 clinical significance (37.9% vs. 55.2%, p=0.188). The point estimates for in-hospital mortality R24 rates of both management types for patients 90 years of age and older were not reliable due to the R25 low number of patients in both groups (table 3). R26 R27 Table 3. In-hospital mortality after medical and surgical management for different age groups. R28 Age (yrs) Management Mortality rate P value R29 Overall Medical Surgical Overall Medical Surgical R30 0 < 10 1 (100.0) 1 (100.0) 0 (0.0) 0 (0.0) - - NA R31 10 < 20 5 (100.0) 1 (20.0) 4 (80.0) 2 (40.0) 1 (100.0) 1 (25.0) NA 20 < 30 17 (100.0) 1 (5.9) 16 (94.1) 5 (29.4) 1 (100.0) 4 (25.0) NA R32 30 < 40 44 (100.0) 0 (0.0) 44 (100.0) 9 (20.5) - 9 (20.5) NA R33 40 < 50 124 (98.4) 12 (9.5) 112 (88.9 ) 27 (21.8) 8 (66.7) 20 (17.5) <0.001 R34 50 < 60 192 (99.5) 18 (9.3) 174 (90.2) 44 (22.9) 11 (61.1) 33 (18.9) <0.001 R35 60 < 70 239 (99.6) 34 (14.2) 205 (85.4) 72 (30.1) 21 (61.8) 52 (25.2) <0.001 R36 70 < 80 237 (99.6) 53 (22.3) 184 (77.3) 83 (35.0) 29 (54.7) 55 (29.7) 0.001 80 < 90 58 (100.0) 29 (50.0) 29 (50.0) 27 (46.6) 16 (55.2) 11 (37.9) 0.188 R37 90 < 100 5 (100.0) 4 (80.0) 1 (20.0) 4 (80.0) 3 (75.0) 1 (100.0) NA R38 R39

49 Chapter 4

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 Figure 3. In-hospital mortality of AAD categorized by age and management type R19 R20 Predictors of in-hospital mortality R21 Age ≥ 70 was an independent predictor for in-hospital mortality among patients with AAD R22 (38.2 vs. 26.0%; p<0.0001, OR 1.73, table 4). Other significant independent predictors of R23 in-hospital mortality for the cohort included coma and/or cerebrovascular accident (OR 2.93), R24 preoperative acute renal failure (OR 2.46), hypotension/shock at presentation (OR 3.21), R25 abrupt onset of symptoms (OR 2.10), ischemic peripheral neuropathy (OR 3.34), prior cardiac R26 surgery (OR 1.84), and a history of hypertension was associated with lower mortality (OR 0.55). R27 R28 Table 4. Independent predictors of in-hospital mortality among patients with AAD R29 Variables in the model OR 95% CI P-value R30 Age ≥ 70 1.728 1.060 2.816 0.0281 R31 History of hypertension 0.556 0.339 0.912 0.0201 Prior cardiac surgery 1.841 1.013 3.347 0.0452 R32 Abrupt onset 2.107 1.004 4.425 0.0489 R33 Hypotension + shock 3.211 1.994 5.170 <0.0001 R34 Coma and/or CVA 2.937 1.410 6.119 0.0040 R35 Preoperative ARF 2.468 1.133 5.379 0.0230 Ischemic periph neuropathy 3.349 1.121 10.003 0.0304 R36 Pleural effusion on chest X-rays 1.706 0.950 3.063 0.0736 R37 R38 OR: odds ratio, CI: confidence interval,CVA : cerebrovascular accident, ARF: acute renal failure, periph peripheral. R39

50 Role of age in acute type A aortic dissection outcome

Proportion of preoperative risk factors among in-hospital deaths < 70 years and ≥ 70 years R1 Among preoperative risk factors that usually are associated with higher mortality, such as R2 shock (systolic blood pressure < 80 mmHg), coma and/or cerebrovascular accident, cardiac R3 tamponade, limb ischemia, acute renal failure, visceral ischemia, myocardial ischemia and/or R4 infarction, history of aortic valve replacement and prior cardiac surgery, only the last variable was R5 significantly more frequently present among non-survivors of 70 years and older, compared with R6 non-survivors younger than 70 (p=0.0005, table 5). R7 R8 Table 5. Proportion of preoperative risk factors among in-hospital deaths < 70 years and ≥ 70 R9 years. 4 R10 Variable Patients age < 70 Patients age ≥ 70 P-value R11 Shock 47 (30.3) 29 (27.6) 0.64 R12 Coma and/or CVA 26 (18.3) 10 (10.9) 0.12 R13 Preoperative ARF 21 (13.7) 11 (10.9) 0.51 Preoperative visceral ischemia 7 (4.6) 8 (7.8) 0.28 R14 Preoperative limb ischemia 30 (19.5) 13 (13.1) 0.19 R15 Preoperative cardiac tamponade 16 (10.1) 9 (8.1) 0.57 R16 ECG with ischemia signs* 15 (11.0) 9 (8.9) 0.61 R17 History of AVR 11 (7.1) 5 (4.8) 0.43 R18 Prior cardiac surgery 22 (14.5) 34 (32.7) 0.0005 R19 Shock: systolic blood pressure < 80mmHg. CVA: cerebrovascular accident, ARF: acute renal failure, AVR: R20 aortic valve repair, ECG: electrocardiogram, * Ischemia signs include myocardial infarction and/or new Q-wave or ST elevation. R21 R22 R23 DISCUSSION R24 R25 With the increasing life expectancy of the population an increased incidence of acute aortic R26 diseases is expected. IRAD confirmed that AAD is more common during 7th and 8th decades, R27 although the condition is not uncommon during 5th and 6th decades. Previous studies, as well R28 reports of IRAD, have shown that increasing age is associated with increased short and long-term R29 mortality rates after cardiovascular interventions (7-10), including AAD surgery (11;12;14). In R30 this report, we confirmed that increasing age is an independent predictor of in-hospital mortality, R31 and we further investigated outcomes of AAD patients after surgical and medical management R32 in different age groups. R33 Ascending aortic repair is the standard of care for managing AAD, although this is still associated R34 with important mortality rates. Whether or not to operate on very elderly patients with AAD R35 has been debated in the literature (15-21). Some have suggested that surgical repair of AAD can R36 be performed in elderly patients with acceptable results (15;16;18-20), while others advocate R37 for a less aggressive approach, or denial of surgical treatment for patients above 80 years, where R38 R39

51 Chapter 4

R1 some reports cite surgical mortality rates of up to 83% (17;21). When elderly AAD patients R2 treated surgically do survive the hospitalization, their long-term survival is often satisfactory R3 (16;18-21). Most of these studies do not compare the surgical outcome of AAD treated with the R4 alternative treatment, medical management. In the present study, we observed that, although R5 the surgical mortality increased with increasing age, surgical management was still associated R6 with significantly lower in-hospital mortality rates compared with medical management for R7 AAD patients until the age of 80 years, and, in addition, it appeared to decrease the in-hospital R8 mortality rate for octogenarians with AAD (38% vs. 55%). R9 In IRAD AAD patients, other important preoperative risk factors for in-hospital mortality, R10 besides an increasing age, were an abrupt onset of symptoms, hypotension and shock, coma and/ R11 or CVA, preoperative acute renal failure, and prior cardiac surgery, which have been reported R12 previously (22). Surprisingly, among non-survivors ≥ 70 years, of the most common preoperative R13 risk factors for mortality, only prior cardiac surgery was more frequent compared with non- R14 survivors < 70 years. This risk factor was more frequently present preoperatively among the R15 elderly AAD patients as well. R16 In recent years, improving cardiopulmonary bypass technology allows for safer procedures with a R17 reduced morbidity and mortality in older patients (23;24). Although we observed that advanced R18 age is not a reason per se to refuse surgery, institutions or individual physicians may approach R19 this issue differently. The decision whether or not to offer surgical repair to a very elderly AAD R20 patient should be based on individual patient characteristics, and the expertise of the operator R21 or institution. R22 A surprising finding was that, in contrast to the surgical mortality, medical mortality did not R23 increase with increasing age. Perhaps medical management was only offered to younger AAD R24 patients if more significant co morbidities or complications were present that prohibited surgery. R25 Additionally, younger patients that were not managed surgically may have expired before surgical R26 repair could have been offered. IRAD is an observational registry and therefore a selection bias R27 may have been present. R28 For surviving patients, it is important to consider subsequent survival and quality of life. Generally, R29 advanced age is associated with considerable neurologic complications after management of AAD R30 (11;15). Elderly patients may become completely bedridden and dependent on the care of family R31 or a nursing home after successful surgical repair of AAD. Recently Hata et al. underlined the R32 impact of postoperative complications in octogenarians with AAD, such as stroke, depression, R33 dementia and/or bedridden status, on their families (15). Several families who had to take R34 responsibility of the patients that survived the operation, voiced their complaints and even R35 refused to pay for the treatment. Further study of this aspect of care and outcomes is sorely R36 needed. R37 R38 R39

52 Role of age in acute type A aortic dissection outcome

The findings of the present study should be viewed in the light of its limitations. Data were R1 collected retrospectively and subject to incomplete or missing reporting of events. Most IRAD R2 centres were tertiary referral sites that have significant expertise and experience in the surgical R3 treatment of patients with acute aortic dissection, thus limiting the applicability to centres that R4 lack such capability (25). In addition, because the treatment allocation was not random, many R5 factors besides those captured in our study may have contributed to the choice of treatment R6 modality, and a selection bias may have been present. Finally, long-term outcomes were not R7 addressed in this study. R8 R9 CONClUSION 4 R10 R11 Although the surgical mortality significantly increased with increasing age, surgical management R12 was still associated with significantly lower in-hospital mortality rates compared with medical R13 management until the age of 80 years, and tended to decrease the in-hospital mortality rate for R14 octogenarians with AAD. Therefore, surgery is still recommended for AAD patients between 70 R15 and 80 years, and appears to be beneficial for octogenarians with AAD. R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

53 Chapter 4

R1 APPENDIX 1 R2 R3 IRAD co-principal investigators R4 Kim A. Eagle, MD, University of Michigan, Ann Arbor, Michigan; Eric M. Isselbacher, MD, R5 Massachusetts General Hospital, Boston, Massachusetts; Christoph A. Nienaber, MD, University R6 of Rostock, Rostock, Germany. R7 R8 IRAD co-investigators R9 Eduardo Bossone, MD, National Research Council, Lecce, Italy; Alan Braverman, MD, R10 Washington University School of Medicine, St. Louis, Missouri; Stefanos Demertzis, MD, R11 Cardiocentro Ticino, Lugano, Switzerland; Giuseppe DiBenedetto, MD, San Giovanni e Ruggi, R12 Salerno, Italy; Mark Ehrlich, MD, University of Vienna, Vienna, Austria; Arturo Evangelista, R13 MD, Hospital General Universitari Vall d’Hebron, Barcelona, Spain; Rossella Fattori, MD, R14 University Hospital S. Orsola, Bologna, Italy; James Froehlich, MD, and Thomas Tsai, MD, R15 University of Michigan, Ann Arbor, Michigan; Dan Gilon, MD, Hadassah University Hospital, R16 Jerusalem, Israel; Alan Hirsch, MD, and Kevin Harris, MD, Minneapolis Heart Institute, R17 Minneapolis, Minnesota; G. Chad Hughes, MD, Duke University, Durham, North Carolina; R18 Stuart Hutchison, MD, St. Michael’s Hospital, Toronto, Ontario, Canada; James L. Januzzi, R19 MD, Massachusetts General Hospital, Boston, Massachusetts; Alfredo Llovet, MD, Hospital R20 Universitario 12 de Octubre, Madrid, Spain; Truls Myrmel, MD, Tromsø University Hospital, R21 Tromsø, Norway; Peter Oberwalder, MD, Medical School Graz, Graz, Austria; Patrick O’Gara, R22 MD, and Joshua Beckman, MD, Brigham and Women’s Hospital, Boston, Massachusetts; Jae K. R23 Oh, MD, Mayo Clinic, Rochester, Minnesota; Linda A. Pape, MD, University of Massachusetts R24 Hospital, Worcester, Massachusetts; Reed Pyeritz, MD, University of Pennsylvania School of R25 Medicine, Philadelphia, Pennsylvania; Udo Sechtem, MD, and Gabriel Meinhardt, MD, Robert- R26 Bosch Krankenhaus, Stuttgart, Germany; P. Gabriel Steg, MD, Hôpital Bichat, Paris, France; R27 Toru Suzuki, MD, University of Tokyo, Tokyo, Japan; Santi Trimarchi, MD, IRCCS Policlinico R28 San Donato, San Donato, Italy. R29 R30 Data management and biostatistical support R31 Jeanna V. Cooper, MS, Dean E. Smith, PhD, and Elise Woznicki, University of Michigan, Ann R32 Arbor, Michigan. R33 R34 R35 R36 R37 R38 R39

54 Role of age in acute type A aortic dissection outcome

REfERENCES R1 R2 (1) Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection: R3 increasing prevalence and improved outcomes reported in a nationwide population-based study R4 of more than 14,000 cases from 1987 to 2002. Circulation 2006 Dec 12;114(24):2611-8. R5 (2) Ehrlich MP, Ergin MA, McCullough JN, Lansman SL, Galla JD, Bodian CA, et al. Results of R6 immediate surgical treatment of all acute type A dissections. Circulation 2000 Nov 7;102(19 R7 Suppl 3):III248-III252. R8 (3) Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Mehta RH, et al. Contemporary R9 results of surgery in acute type A aortic dissection: The International Registry of Acute Aortic 4 R10 Dissection experience. J Thorac Cardiovasc Surg 2005 Jan;129(1):112-22. R11 (4) Narayan P, Rogers CA, Davies I, Angelini GD, Bryan AJ. Type A aortic dissection: has surgical R12 outcome improved with time? J Thorac Cardiovasc Surg 2008 Nov;136(5):1172-7. R13 (5) Bachet J, Goudot B, Dreyfus GD, Brodaty D, Dubois C, Delentdecker P, et al. Surgery for R14 acute type A aortic dissection: the Hopital Foch experience (1977-1998). Ann Thorac Surg 1999 R15 Jun;67(6):2006-9. R16 (6) David TE, Armstrong S, Ivanov J, Barnard S. Surgery for acute type A aortic dissection. Ann R17 Thorac Surg 1999 Jun;67(6):1999-2001. R18 (7) Rich MW, Bosner MS, Chung MK, Shen J, McKenzie JP. Is age an independent predictor of early R19 and late mortality in patients with acute myocardial infarction? Am J Med 1992 Jan;92(1):7-13. R20 (8) Maggioni AP, Maseri A, Fresco C, Franzosi MG, Mauri F, Santoro E, et al. Age-related increase R21 in mortality among patients with first myocardial infarctions treated with thrombolysis. The R22 Investigators of the Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico R23 (GISSI-2). N Engl J Med 1993 Nov 11;329(20):1442-8. R24 (9) Tsai TP, Matloff JM, Chaux A, Kass RM, Lee ME, Czer LS, et al. Combined valve and coronary R25 artery bypass procedures in septuagenarians and octogenarians: results in 120 patients. Ann R26 Thorac Surg 1986 Dec;42(6):681-4. R27 (10) Blakeman BM, Pifarre R, Sullivan HJ, Montoya A, Bakhos M, Grieco JG, et al. Aortic valve R28 replacement in patients 75 years old and older. Ann Thorac Surg 1987 Dec;44(6):637-9. R29 (11) Mehta RH, O’Gara PT, Bossone E, Nienaber CA, Myrmel T, Cooper JV, et al. Acute type A R30 aortic dissection in the elderly: clinical characteristics, management, and outcomes in the current R31 era. J Am Coll Cardiol 2002 Aug 21;40(4):685-92. R32 (12) Tan ME, Morshuis WJ, Dossche KM, Kelder JC, Waanders FG, Schepens MA. Long-term R33 results after 27 years of surgical treatment of acute type a aortic dissection. Ann Thorac Surg R34 2005 Aug;80(2):523-9. R35 (13) Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, et al. The R36 International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. R37 JAMA 2000 Feb 16;283(7):897-903. R38 R39

55 Chapter 4

R1 (14) Tsai TT, Evangelista A, Nienaber CA, Trimarchi S, Sechtem U, Fattori R, et al. Long-term R2 survival in patients presenting with type A acute aortic dissection: insights from the International R3 Registry of Acute Aortic Dissection (IRAD). Circulation 2006 Jul 4;114(1 Suppl):I350-I356. R4 (15) Hata M, Sezai A, Niino T, Yoda M, Unosawa S, Furukawa N, et al. Should emergency surgical R5 intervention be performed for an octogenarian with type A acute aortic dissection? J Thorac R6 Cardiovasc Surg 2008 May;135(5):1042-6. R7 (16) Kawahito K, Adachi H, Yamaguchi A, Ino T. Early and late surgical outcomes of acute type A R8 aortic dissection in patients aged 75 years and older. Ann Thorac Surg 2000 Nov;70(5):1455-9. R9 (17) Neri E, Toscano T, Massetti M, Capannini G, Carone E, Tucci E, et al. Operation for acute type A R10 aortic dissection in octogenarians: is it justified? J Thorac Cardiovasc Surg 2001 Feb;121(2):259- R11 67. R12 (18) Caus T, Frapier JM, Giorgi R, Aymard T, Riberi A, Albat B, et al. Clinical outcome after R13 repair of acute type A dissection in patients over 70 years-old. Eur J Cardiothorac Surg 2002 R14 Aug;22(2):211-7. R15 (19) Chiappini B, Tan ME, Morshuis W, Kelder H, Dossche K, Schepens M. Surgery for acute type A R16 aortic dissection: is advanced age a contraindication? Ann Thorac Surg 2004 Aug;78(2):585-90. R17 (20) Santini F, Montalbano G, Messina A, D’Onofrio A, Casali G, Viscardi F, et al. Survival and R18 quality of life after repair of acute type A aortic dissection in patients aged 75 years and older R19 justify intervention. Eur J Cardiothorac Surg 2006 Mar;29(3):386-91. R20 (21) Piccardo A, Regesta T, Pansini S, Concistre G, Dell’Aquila A, Scarano F, et al. Should R21 octogenarians be denied access to surgery for acute type A aortic dissection? J Cardiovasc Surg R22 (Torino) 2009 Apr;50(2):205-12. R23 (22) Rampoldi V, Trimarchi S, Eagle KA, Nienaber CA, Oh JK, Bossone E, et al. Simple risk models R24 to predict surgical mortality in acute type A aortic dissection: the International Registry of Acute R25 Aortic Dissection score. Ann Thorac Surg 2007 Jan;83(1):55-61. R26 (23) Bacchetta MD, Ko W, Girardi LN, Mack CA, Krieger KH, Isom OW, et al. Outcomes of cardiac R27 surgery in nonagenarians: a 10-year experience. Ann Thorac Surg 2003 Apr;75(4):1215-20. R28 (24) Ullery BW, Peterson JC, Milla F, Wells MT, Briggs W, Girardi LN, et al. Cardiac surgery in select R29 nonagenarians: should we or shouldn’t we? Ann Thorac Surg 2008 Mar;85(3):854-60. R30 (25) Birkmeyer JD, Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. volume R31 and operative mortality in the United States. N Engl J Med 2003 Nov 27;349(22):2117-27. R32 R33 R34 R35 R36 R37 R38 R39

56 Chapter 5

Age-related Decision Making in Complicated Acute Type B Aortic Dissection

(Submitted)

Frederik H.W. Jonker, MD, PhD,1 Santi Trimarchi, MD,2 Bart E. Muhs, MD, PhD,1 Vincenzo Rampoldi, MD,2 Daniel G Montgomery, MS,3 James B Froehlich, MD,3 Mark D Peterson, MD,4 Kristian Bartnes, MD,5 Venu Gourineni, MD, 3 Vijay S. Ramanath, MD, 3 Alan C Braverman, MD,6 Christoph A Nienaber, MD,7 Eric M Isselbacher, MD,8 Kim A. Eagle, MD, 3

On behalf of the International Registry of Acute Aortic Dissection (IRAD) investigators

1 Yale University, New Haven, Conn, USA 2 Policlinico San Donato IRCCS, San Donato Milanese, Italy 3 University of Michigan, Ann Arbor, Mich, USA 4 St. Michael’s Hospital, University of Toronto, Canada 5 Tromsø University Hospital, Tromsø, Norway 6 Washington University School of Medicine, St. Louis, Miss, USA 7 University of Rostock, Rostock, Germany 8 Massachusetts General Hospital, Boston, Mass, USA Chapter 5

R1 AbSTRACT R2 R3 Introduction R4 Complicated acute type B aortic dissection (cABAD) generally requires urgent intervention. R5 Advanced age is a risk factor for mortality after thoracic aortic intervention, including surgery R6 for aortic dissection. The purpose of this study was to investigate the exact impact of increasing R7 age on the management and outcomes of cABAD. R8 R9 Methods R10 We analyzed the outcomes of 464 patients with cABAD (44.0%) among the 1054 type B patients R11 enrolled in the International Registry of Acute Aortic Dissection (IRAD) between 1996 and R12 2010. All patients with cABAD were categorized according to age by decade and management R13 type (medical, surgical or endovascular treatment), and outcomes were subsequently investigated R14 in the different age groups. R15 R16 Results R17 The mean age for patients treated with surgery was 60.5 years, compared with 62.1 years in the R18 endovascular group, and 66.8 years in the medical management group (p<0.001). The utilization R19 of surgery progressively decreased with patient age, while the rate of medical management R20 significantly increased with age (p<0.001). The in-hospital mortality rates for complicated R21 patients ≤ 70 years vs. > 70 years were 7.7% vs. 35.7% for endovascular treatment (p=0.012), R22 13.5% vs. 32.8% for surgical treatment (p=0.001), and 14.6% vs. 26.6% for medical treatment R23 (p=0.036). R24 Predictors of mortality in multivariate regression analysis were: age > 70 years (OR 6.6, 95% CI R25 1.9-22.8), hypotension/shock (OR 10.0, 95% CI 2.8-36.3), lack of chest/back pain (OR 4.4, R26 95% CI 1.1-18.1), and pre-operative limb ischemia (OR 5.4, 95% CI 1.4-20.9). R27 R28 Conclusions R29 The rate of surgical repair progressively decreased with age, while the rate of medical management R30 significantly increased with age, which may serve as an alternative to invasive treatment in very R31 elderly patients. The use of endovascular methods for cABAD has increased over the years, R32 however, advanced age has a dramatic impact on the mortality of cABAD, irrespective of the R33 management type. R34 R35 R36 R37 R38 R39

58 Age-related Decision Making in Complicated Acute Type B Aortic Dissection

INTRODUCTION R1 R2 Acute type B aortic dissection (ABAD) is a cardiovascular emergency with considerable mortality R3 and morbidity.1-4 In the absence of complications, ABAD is typically managed with medical R4 treatment only, which is associated with low in-hospital mortality rates, between 0% and 8%.3-5 R5 Patients with complicated acute type B aortic dissection (cABAD), such as aortic rupture, R6 malperfusion of the lower limbs, acute renal failure, or visceral ischemia, have much poorer R7 outcomes, and mortality rates usually ranging between 10% to 30%.3-8 Invasive treatment, either R8 surgery or endovascular, is indicated in patients with cABAD.3-8 R9 Elderly patients typically have more pre-existing co-morbidities, and advanced age is a well R10 known risk factor for mortality after thoracic aortic intervention, including surgery for acute R11 type A and B aortic dissections.5, 9, 10 Due to the relatively low incidence of ABAD, the exact R12 effects of age on the management and outcomes of cABAD are unclear. The International 5 R13 Registry of Acute Aortic Dissection (IRAD) represents the largest cohort of patients with acute R14 aortic dissection, and therefore offers an opportunity to study this cardiovascular emergency. R15 In this study, we used the IRAD Registry to investigate the outcomes of cABAD after medical, R16 surgical and endovascular treatment in different age groups, in order to gain insight regarding R17 effectiveness of current treatment strategies for patients with cABAD. R18 R19 R20 METHODS R21 R22 Patient selection R23 IRAD is an ongoing multi-center registry that includes patients with acute aortic dissection at 24 R24 large referral centers (IRAD centers; Appendix I), which rationale and methods have previously R25 been described.11 For the present study, we included all patients with cABAD enrolled in the IRAD R26 between January 1996 and February 2010. We defined cABAD as one or more of the following R27 complications: shock, periaortic hematoma, spinal cord ischemia, pre-operative mesenteric R28 ischemia/infarction, acute renal failure, limb ischemia, recurrent hypertension, recurrent and/or R29 refractory pain. Patients with uncomplicated aortic dissection or traumatic aortic dissection were R30 excluded, as well as those patients for whom data regarding age was unavailable. R31 All patients with cABAD were categorized according to patient age by decade, and management R32 type (medical, surgical or endovascular treatment). The management and outcomes were R33 subsequently investigated in the different age groups. The study was approved by the institutional R34 review committee at all participating IRAD institutions. R35 R36 R37 R38 R39

59 Chapter 5

R1 Data collection and Analysis R2 Data were collected using a standardized data form of 290 clinical variables including patient R3 demographics, history, clinical presentation, physical findings, imaging studies, management, R4 and in-hospital mortality. Completed data forms were forwarded to the coordinating center at R5 the University of Michigan. Data forms were reviewed for internal validity and completeness of R6 data, and were then entered into an Access database. R7 Summary statistics were presented as frequencies and percentages, mean + standard deviation, or R8 as a median and interquartile range. Missing data were not defaulted to negative and denominators R9 reflect only actual reported cases. Nominal variables were compared between patients ≤70 years R10 and >70 years, and between medical, surgical and endovascular groups, using the Chi-square R11 test or two-sided Fisher exact test. The mean age of patients treated with medical, surgical and R12 endovascular treatment was compared using the ANOVA test. R13 Multivariate logistic regression analysis was performed to investigate independent predictors R14 of in-hospital mortality among patients with cABAD. Variables suggestive of an unadjusted R15 association to in-hospital mortality (p<0.20) were integrated in a multivariable regression model R16 to calculate independent effects on mortality. SPSS 17.0 was utilized for the analyses, and a R17 P value < 0.05 was considered significant. R18 R19 R20 RESUlTS R21 R22 Demographics and Medical History R23 cABAD was present in 44.0% (n=464) of the 1054 type B patients in IRAD. The mean age of R24 complicated ABAD patients was 63.3 ±14.2 years (figure 1), 35.8% (n=166) of patients were R25 older than 70 years and 64.2% (n=298) were ≤ 70 years. The mean age of patients undergoing R26 surgery was 60.5 years compared with 62.1 years for patients treated with endovascular R27 management and 66.8 years for patients with medical management (p<0.001). The younger R28 cohort consisted of more male patients (72.1% vs. 60.2%, p=0.001). The elderly patients with R29 cABAD had more frequently pre-existing hypertension, diabetes mellitus, atherosclerosis, and R30 prior aortic aneurysm, while a history of Marfan’s Syndrome was only present in the younger R31 patients (table 1). R32 R33 R34 R35 R36 R37 R38 R39

60 Age-related Decision Making in Complicated Acute Type B Aortic Dissection

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 5 R13 R14 R15 Figure 1. Age of patients with cABAD in IRAD R16 The mean age of cABAD patients was 63.3 ±14.2 years (figure 1), 35.8% (n=166) of patients were older than 70 years and 64.2% (n=298) were ≤ 70 years. R17 R18 Table 1. Demographics and Patient History R19 Variable Age ≤70 (n=298) Age >70 (n=166) P value R20 N (%) N (%) R21 Demographics R22 Male gender 215 (72.1) 100 (60.2) 0.001 Etiology and History R23 Hypertension 231 (80.2) 144 (88.9) 0.011 R24 Diabetes 16 (5.8) 24 (15.3) 0.001 R25 Atherosclerosis 60 (21.7) 96 (60.4) <0.001 R26 Marfan’s syndrome 21 (7.2) 0 (0) <0.001 R27 Prior aortic aneurysm 41 (14.4) 47 (29.6) <0.001 Prior cardiac surgery 52 (18.7) 38 (26.0) 0.080 R28 Cardiac catheterization 22 (9.6) 18 (14.8) 0.148 R29 Iatrogenic dissection 5 (1.8) 4 (2.5) 0.644 R30 R31 Presentation and Imaging findings R32 Patients ≤ 70 years presented more frequently with visceral ischemia (15.7% vs. 6.3%, p=0.002), R33 limb ischemia (25.7% vs. 5.7%, p<0.001) and pulse deficits (26.9% vs. 13.4%, p<0.001). R34 Recurrent hypertension was more common in the younger patients as well (11.3% vs. 2.8%, R35 p=0.001). Pleural effusion (28.9% vs. 18.7%, p=0.025) and periaortic hematoma (41.0% vs. R36 22.5%, p<0.001) were more often observed on imaging studies among elderly patients, while R37 abdominal vessel involvement was less frequently seen (23.5% vs. 48.1%, p<0.001). The mean R38 R39

61 Chapter 5

R1 diameter of the descending thoracic aorta was larger in patients above 70 years (5.1 cm vs. 4.4 R2 cm, p=0.016). R3 R4 Table 2. Presentation, Signs and Imaging Results R5 Variable Age ≤70 (n=298) Age >70 (n=166) P value R6 N (%) N (%) R7 Clinical Presentation and Signs Abrupt onset of pain 238 (86.9) 144 (92.3) 0.085 R8 Lack of chest/back pain 33 (11.8) 18 (11.0) 0.803 R9 Shock at presentation 7 (2.5) 5 (3.1) 0.771 R10 Spinal cord ischemia 5 (1.8) 2 (1.3) 0.612 R11 Limb ischemia 71 (25.7) 9 (5.7) <0.001 Visceral ischemia 44 (15.7) 10 (6.3) 0.002 R12 Acute renal failure 83 (30.3) 45 (28.3) 0.512 R13 Any pulse deficit 68 (26.9) 18 (13.4) 0.001 R14 Recurrent pain 55 (23.1) 27 (18.8) 0.223 R15 Refractory pain 25 (11.0) 13 (9.0) 0.548 Recurrent hypertension 28 (11.3) 4 (2.8) 0.001 R16 Diagnostic imaging results R17 Chest X-ray R18 Pleural effusion 45 (18.7) 43 (28.9) 0.025 R19 Mediastinal widening 116 (49.2) 83 (56.5) 0.164 R20 Additional imaging findings Diameter descending aorta (cm) 4.4 (±1.2) 5.1 (±2.8) 0.016 R21 Peri-aortic hematoma 64 (22.5) 68 (41.0) <0.001 R22 Abdominal vessel involvement 136 (48.1) 38 (23.5) <0.001 R23 R24 Management R25 Invasive treatment was offered to 57.5% of the complicated cohort, and consisted of surgery R26 in 46.1% and endovascular treatment in 11.4% of all patients. The remaining 42.5% of the R27 complicated cohort received medical treatment alone (table 3). The utilization of endovascular R28 management for cABAD in IRAD increased over the years, while medical and surgical R29 management were less frequently offered to more recent patients with cABAD (figure 2). R30 The number of patients undergoing surgery decreased with increasing age, while the proportion R31 of patients that received medical treatment alone increased with increasing age (table 4, figure R32 3). Surgery was performed in 52.3% of all patients ≤ 70 years, and in 34.9% of patients older R33 than 70 years (p<0.001). Among the patients ≤ 70 years, medical treatment was offered to 34.6% R34 of patients, while 56.6% of the elderly cohort was treated medically (p<0.001). The number R35 of patients undergoing endovascular management slightly decreased with increasing age of the R36 cABAD patients (figure 3). R37 R38 R39

62 Age-related Decision Making in Complicated Acute Type B Aortic Dissection

Table 3. In-hospital management and mortality R1 Overall Age ≤70 (n=298) Age >70 (n=166) P value R2 N (%) N (%) R3 Definitive management R4 Medical 197 (42.5) 103 (34.6) 94 (56.6) <0.001 Surgery 214 (46.1) 156 (52.3) 58 (34.9) <0.001 R5 Endovascular 53 (11.4) 39 (13.1) 14 (8.4) 0.131 R6 In-hospital mortality 88 (19.0) 39 (13.1) 49 (29.5) <0.001 R7 Medical 40 (20.3) 15 (14.6) 25 (26.6) 0.036 R8 Surgery 40 (18.7) 21 (13.5) 19 (32.8) 0.001 R9 Endovascular 8 (15.1) 3 (7.7) 5 (35.7) 0.012 R10 R11 R12 5 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 Figure 2. Management of cABAD in different age groups R27 The mean age for patients treated with surgery was 60.5 years, compared with 62.1 years in the endovascular group, and 66.8 years in the medical management group (p<0.001). R28 R29 Table 4. Independent predictors of in-hospital mortality R30 Variable Odds ratio 95%CI P Value R31 Age >70 6.62 1.92 – 22.82 0.003 R32 Female gender 0.84 0.22 – 3.24 0.798 R33 Branch vessel involvement 1.09 0.32 – 3.77 0.888 Lack of chest/back pain 4.37 1.06 – 18.07 0.042 R34 Refractory pain/hypertension 1.79 0.54 – 5.98 0.343 R35 Hypotension / shock 10.01 2.75 – 36.31 <0.001 R36 Visceral ischemia pre-op 2.41 0.54 – 10.81 0.252 R37 Limb ischemia pre-op 5.40 1.40 – 20.88 0.014 R38 Acute renal failure pre-op 0.47 0.10 – 2.12 0.324 R39

63 Chapter 5

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 Figure 3. In-hospital mortality of cABAD in different age groups The overall in-hospital mortality rate after endovascular treatment was 15.1%, and the mortality rates after R16 surgical and medical treatment were 18.7% and 20.3%, respectively (p=0.675) R17 R18 Outcomes R19 The overall in-hospital mortality of cABAD in IRAD was 19.0%. The mortality rate after R20 endovascular treatment of cABAD was 15.1%, and the mortality rates after surgical and medical R21 treatment were 18.7% and 20.3% (table 3), respectively (p=0.675). The overall in-hospital R22 mortality rate was 13.1% among patients ≤ 70 years, compared with 29.5% among patients R23 older than 70 years (p<0.001). For patients ≤ 70 years, there was a trend towards lower mortality R24 after endovascular treatment (figure 4), while for patients older than 70, the mortality was slightly R25 lower after medical management (table 3). The in-hospital mortality significantly increased with R26 increasing age, irrespective of the management type (table 3). In particular after the seventh R27 decade, the mortality considerably increased, approaching 50% for patients > 80 years (figure 4). R28 R29 Predictors of in-hospital mortality R30 Multivariate logistic regression analysis showed that age > 70 years was a strong predictor of R31 mortality among patients with cABAD (OR 6.6, 95% CI 1.9-22.8) (table 4). Other independent R32 predictors of mortality were hypotension/shock at admission (OR 10.0, 95% CI 2.8-36.3), lack R33 of chest/back pain at presentation (OR 4.4, 95% CI 1.1-18.1), and pre-operative limb ischemia R34 (OR 5.4, 95% CI 1.4-20.9). R35 R36 R37 R38 R39

64 Age-related Decision Making in Complicated Acute Type B Aortic Dissection

DISCUSSION R1 R2 Our analysis shows that increasing patient age has a dramatic impact on the management and R3 outcomes of cABAD. The rate of surgical repair progressively decreased with age, while the rate of R4 medical management significantly increased with age. There was a non significant trend towards R5 lower mortality after endovascular treatment, but patients older than 70 years had a significantly R6 increased mortality rate, irrespective of the management type. R7 Although patients with uncomplicated type B dissection have a favorable prognosis, outcomes of R8 cABAD are generally poor, and improvements in current treatment strategies for these patients R9 are needed. The proportion of patients with cABAD was 44% in IRAD, similar to previous R10 studies in which 30% to 47% of all ABAD patients had complications at presentation.3, 5, 6, 8, 12, 13 R11 Invasive treatment, either surgical or endovascular management, is generally indicated in patients R12 with cABAD,3-6, 8 in order to prevent further decline of the patient’s condition. 5 R13 Acute aortic syndromes are typically associated with significant morbidity and mortality, and R14 the risks of surgery are dramatically increased in elderly patients. Advanced age is an important R15 predictor of death after intervention for ruptured abdominal aortic aneurysms,14-17 ruptured R16 thoracic aortic aneurysms,8, 18, 19 traumatic thoracic aortic injuries,20 and acute type A and B R17 aortic dissections.5, 9, 10, 19, 21 This IRAD report confirms that age > 70 years is a strong and R18 independent risk factor for mortality (OR 6.6) among patients with cABAD. Mortality rates R19 increased significantly with increasing age, irrespective of surgical, endovascular or medical R20 treatment. Elderly patients more frequently have extensive pre-existing co-morbidities, such as R21 hypertension, atherosclerosis, and diabetes (table 1), which results in an increased risk of an R22 adverse outcome. In the current study, elderly patients had higher rates of aortic aneurysms, R23 a measure of advanced atherosclerosis, associated with increased cardiovascular mortality.22, 23 R24 Moreover, periaortic hematoma, was more common in elderly patients as well, and this sign R25 of aortic rupture is a strong predictor of mortality.24 Another factor that may have contributed R26 to adverse outcomes in the elderly patients, is the fact that this cohort consisted of relatively R27 more female patients, and women with cardiovascular disease have been shown to have poorer R28 outcomes compared with men.25-29 R29 Because of the increased risks of thoracic aortic surgery in elderly patients, physicians offered less R30 frequently surgical treatment to these cABAD patients. It is also plausible that some patients with R31 advanced age refused invasive therapy. Endovascular treatment, which was not yet available in the R32 early years of the IRAD, was offered in similar rates in the different age categories. On average, R33 the descending aortic diameters were larger in patients > 70 years, which may correlate with a R34 higher frequency of unsuitable landing zones for an endograft in these elderly patients. Generally, R35 the utilization of endovascular management of thoracic aortic emergencies is becoming more R36 widespread due to its less invasiveness compared with surgery, which is especially desirable in R37 elderly patients with extensive co-morbidities. R38 R39

65 Chapter 5

R1 In IRAD, the overall in-hospital mortality of cABAD was 19%, which is comparable to the results R2 of other evaluations which typically report mortality rates between 10% and 30%.3, 6, 8, 12, 13 We R3 observed a non significant trend towards lower mortality after endovascular treatment (15.1%), R4 compared with surgery (18.7%), or medical treatment (20.3%), similar to recent reports.6, 30-35 In R5 the present evaluation, the largest cABAD cohort ever studied, we did not observe a statistically R6 significant improvement in survival after endovascular management. This may be the result of the R7 relatively small number of endovascular repairs in IRAD, non-superiority of this approach and/ R8 or the fact that IRAD is a non-randomized observational database, in which potential benefit of R9 treatment is hidden by selection bias. Patients with cABAD that were treated with endovascular R10 methods were significantly older than those treated with surgery, and endovascular management R11 may have served as a last option in some patients in critical condition, which were thought to be R12 unfit for open surgery. R13 Although invasive treatment is typically recommended for cABAD, a considerable number R14 of the cABAD patients in IRAD received medical treatment only, in particular in the initial R15 IRAD experience (figure 2). Surprisingly, the outcomes of medical management of cABAD R16 were acceptable, and the mortality rate after medical treatment was slightly lower compared R17 with mortality rates after surgical or endovascular treatment among patients older than 70 R18 years, although the study sample was limited. Therefore, medical treatment may be a reasonable R19 alternative in the management of cABAD among patients above 70 years. However, due to R20 the observational characteristics of IRAD, the patients treated with medical management may R21 have had less catastrophic complications, such as recurrent pain and/or hypertension, than those R22 who presented with aortic rupture or ischemic complications which were managed with invasive R23 treatment. R24 R25 limitations R26 The findings of the present study should be viewed in the light of its limitations. Patients were R27 not randomized to a predetermined management strategy, and therefore a selection bias could R28 have been present. In the absence of large randomized trials, there remains uncertainty as to the R29 optimal strategy to manage patients with cABAD. R30 R31 R32 CONClUSIONS R33 R34 The rate of surgical repair progressively decreased with age, while the rate of medical management R35 significantly increased with age, which may serve as an alternative to invasive treatment in very R36 elderly patients. The use of endovascular methods for cABAD has increased over the years, R37 however, advanced age has a dramatic impact on the mortality of cABAD, irrespective of the R38 management type. R39

66 Age-related Decision Making in Complicated Acute Type B Aortic Dissection

REfERENCES R1 R2 (1) Anagnostopoulos CE, Prabhakar MJ, Kittle CF. Aortic dissections and dissecting aneurysms. Am R3 J Cardiol 1972;30:263-273. R4 (2) Wheat MW, Jr. Acute dissecting aneurysms of the aorta: diagnosis and treatment--1979. Am R5 Heart J 1980;99:373-387. R6 (3) Elefteriades JA, Lovoulos CJ, Coady MA, Tellides G, Kopf GS, Rizzo JA. Management of R7 descending aortic dissection. Ann Thorac Surg 1999;67:2002-2005. R8 (4) Golledge J, Eagle KA. Acute aortic dissection. Lancet 2008;372:55-66. R9 (5) Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Bossone E, Tolva V, Deeb MG, R10 Upchurch GR, Jr., Cooper JV, Fang J, Isselbacher EM, Sundt TM, III, Eagle KA. Role and results R11 of surgery in acute type B aortic dissection: insights from the International Registry of Acute R12 Aortic Dissection (IRAD). Circulation 2006;114:I357-I364. 5 R13 (6) Fattori R, Tsai TT, Myrmel T, Evangelista A, Cooper JV, Trimarchi S, Li J, Lovato L, Kische S, R14 Eagle KA, Isselbacher EM, Nienaber CA. Complicated acute type B dissection: is surgery still R15 the best option?: a report from the International Registry of Acute Aortic Dissection. JACC R16 Cardiovasc Interv 2008;1:395-402. R17 (7) Fann JI, Sarris GE, Mitchell RS, Shumway NE, Stinson EB, Oyer PE, Miller DC. Treatment R18 of patients with aortic dissection presenting with peripheral vascular complications. Ann Surg R19 1990;212:705-713. R20 (8) Eggebrecht H, Lonn L, Herold U, Breuckmann F, Leyh R, Jakob HG, Erbel R. Endovascular R21 stent-graft placement for complications of acute type B aortic dissection. Curr Opin Cardiol R22 2005;20:477-483. R23 (9) Trimarchi S, Eagle KA, Nienaber CA, Rampoldi V, Jonker FH, De VC, Frigiola A, Menicanti R24 L, Tsai T, Froehlich J, Evangelista A, Montgomery D, Bossone E, Cooper JV, Li J, Deeb G, R25 Meinhardt G, Sundt TM, Isselbacher EM. Role of age in acute type A aortic dissection outcome: R26 Report from the International Registry of Acute Aortic Dissection (IRAD). J Thorac Cardiovasc R27 Surg 2010 [Epub ahead of print]. R28 (10) Ehrlich MP, Ergin MA, McCullough JN, Lansman SL, Galla JD, Bodian CA, Apaydin A, R29 Griepp RB. Results of immediate surgical treatment of all acute type A dissections. Circulation R30 2000;102:III248-III252. R31 (11) Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, Evangelista R32 A, Fattori R, Suzuki T, Oh JK, Moore AG, Malouf JF, Pape LA, Gaca C, Sechtem U, Lenferink R33 S, Deutsch HJ, Diedrichs H, Robles J, Llovet A, Gilon D, Das SK, Armstrong WF, Deeb GM, R34 Eagle KA. The International Registry of Acute Aortic Dissection (IRAD): new insights into an R35 old disease. JAMA 2000;283:897-903. R36 (12) Umana JP, Lai DT, Mitchell RS, Moore KA, Rodriguez F, Robbins RC, Oyer PE, Dake MD, R37 Shumway NE, Reitz BA, Miller DC. Is medical therapy still the optimal treatment strategy for R38 patients with acute type B aortic dissections? J Thorac Cardiovasc Surg 2002;124:896-910. R39

67 Chapter 5

R1 (13) Estrera AL, Miller CC, Goodrick J, Porat EE, Achouh PE, Dhareshwar J, Meada R, R2 Azizzadeh A, Safi HJ. Update on outcomes of acute type B aortic dissection. Ann Thorac Surg R3 2007;83:S842-S845. R4 (14) McPhee J, Eslami MH, Arous EJ, Messina LM, Schanzer A. Endovascular treatment of ruptured R5 abdominal aortic aneurysms in the United States (2001-2006): a significant survival benefit R6 over open repair is independently associated with increased institutional volume. J Vasc Surg R7 2009;49:817-826. R8 (15) Egorova N, Giacovelli J, Greco G, Gelijns A, Kent CK, McKinsey JF. National outcomes for R9 the treatment of ruptured abdominal aortic aneurysm: comparison of open versus endovascular R10 repairs. J Vasc Surg 2008;48:1092-100, 1100. R11 (16) Giles KA, Hamdan AD, Pomposelli FB, Wyers MC, Dahlberg SE, Schermerhorn ML. R12 Population-based outcomes following endovascular and open repair of ruptured abdominal R13 aortic aneurysms. J Endovasc Ther 2009;16:554-564. R14 (17) Wanhainen A, Bylund N, Bjorck M. Outcome after abdominal aortic aneurysm repair in Sweden R15 1994-2005. Br J Surg 2008;95:564-570. R16 (18) Schermerhorn ML, Giles KA, Hamdan AD, Dalhberg SE, Hagberg R, Pomposelli F. Population- R17 based outcomes of open descending thoracic aortic aneurysm repair. J Vasc Surg 2008;48:821- R18 827. R19 (19) Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection: R20 increasing prevalence and improved outcomes reported in a nationwide population-based study R21 of more than 14,000 cases from 1987 to 2002. Circulation 2006;114:2611-2618. R22 (20) Arthurs ZM, Starnes BW, Sohn VY, Singh N, Martin MJ, Andersen CA. Functional and R23 survival outcomes in traumatic blunt thoracic aortic injuries: An analysis of the National Trauma R24 Databank. J Vasc Surg 2009;49:988-994. R25 (21) Mehta RH, O’Gara PT, Bossone E, Nienaber CA, Myrmel T, Cooper JV, Smith DE, Armstrong R26 WF, Isselbacher EM, Pape LA, Eagle KA, Gilon D. Acute type A aortic dissection in the elderly: R27 clinical characteristics, management, and outcomes in the current era. J Am Coll Cardiol R28 2002;40:685-692. R29 (22) Newman AB, Arnold AM, Burke GL, O’Leary DH, Manolio TA. Cardiovascular disease and R30 mortality in older adults with small abdominal aortic aneurysms detected by ultrasonography: R31 the cardiovascular health study. Ann Intern Med 2001;134:182-190. R32 (23) Freiberg MS, Arnold AM, Newman AB, Edwards MS, Kraemer KL, Kuller LH. Abdominal R33 aortic aneurysms, increasing infrarenal aortic diameter, and risk of total mortality and incident R34 cardiovascular disease events: 10-year follow-up data from the Cardiovascular Health Study. R35 Circulation 2008;117:1010-1017. R36 (24) Mukherjee D, Evangelista A, Nienaber CA, Sechtem U, Suzuki T, Trimarchi S, Llovet A, Myrmel R37 T, O’Gara PT, Fang J, Cooper JV, Oh JK, Januzzi JL, Hutchison S, Fattori R, Pape LA, Isselbacher R38 EM, Eagle KA. Implications of periaortic hematoma in patients with acute aortic dissection R39 (from the International Registry of Acute Aortic Dissection). Am J Cardiol 2005;96:1734-1738.

68 Age-related Decision Making in Complicated Acute Type B Aortic Dissection

(25) Nienaber CA, Fattori R, Mehta RH, Richartz BM, Evangelista A, Petzsch M, Cooper JV, Januzzi R1 JL, Ince H, Sechtem U, Bossone E, Fang J, Smith DE, Isselbacher EM, Pape LA, Eagle KA. R2 Gender-related differences in acute aortic dissection. Circulation 2004;109:3014-3021. R3 (26) Mikhail GW. Coronary revascularisation in women. Heart 2006 May;92 Suppl 3:iii19-iii23. R4 (27) Malenka DJ, O’Connor GT, Quinton H, Wennberg D, Robb JF, Shubrooks S, Kellett MA, R5 Jr., Hearne MJ, Bradley WA, VerLee P. Differences in outcomes between women and men R6 associated with percutaneous transluminal coronary angioplasty. A regional prospective study of R7 13,061 procedures. Northern New England Cardiovascular Disease Study Group. Circulation R8 1996;94:II99-104. R9 (28) Norman PE, Powell JT. Abdominal aortic aneurysm: the prognosis in women is worse than in R10 men. Circulation 2007;115:2865-2869. R11 (29) Schlosser FJ, Vaartjes I, van der Heijden GJ, Moll FL, Verhagen HJ, Muhs BE, de Borst GJ, Tiel R12 Groenestege AT, Kardaun JW, de BA. Mortality after elective abdominal aortic aneurysm repair. 5 R13 Ann Surg 2010;251:158-164. R14 (30) Khoynezhad A, Donayre CE, Omari BO, Kopchok GE, Walot I, White RA. Midterm results of R15 endovascular treatment of complicated acute type B aortic dissection. J Thorac Cardiovasc Surg R16 2009;138:625-631. R17 (31) Patel HJ, Williams DM, Meerkov M, Dasika NL, Upchurch GR, Jr., Deeb GM. Long-term results R18 of percutaneous management of malperfusion in acute type B aortic dissection: implications for R19 thoracic aortic endovascular repair. J Thorac Cardiovasc Surg 2009;138:300-308. R20 (32) Kische S, Ehrlich MP, Nienaber CA, Rousseau H, Heijmen R, Piquet P, Ince H, Beregi JP, R21 Fattori R. Endovascular treatment of acute and chronic aortic dissection: midterm results from R22 the Talent Thoracic Retrospective Registry. J Thorac Cardiovasc Surg 2009;138:115-124. R23 (33) Feezor RJ, Martin TD, Hess PJ, Jr., Beaver TM, Klodell CT, Lee WA. Early outcomes after R24 endovascular management of acute, complicated type B aortic dissection. J Vasc Surg R25 2009;49:561-566. R26 (34) Verhoye JP, Miller DC, Sze D, Dake MD, Mitchell RS. Complicated acute type B aortic R27 dissection: midterm results of emergency endovascular stent-grafting. J Thorac Cardiovasc Surg R28 2008;136:424-430. R29 (35) Duebener LF, Lorenzen P, Richardt G, Misfeld M, Notzold A, Hartmann F, Sievers HH, Geist V. R30 Emergency endovascular stent-grafting for life-threatening acute type B aortic dissections. Ann R31 Thorac Surg 2004;78:1261-1266. R32 R33 R34 R35 R36 R37 R38 R39

69

Chapter 6

Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection: Insights from the International Registry of Acute Aortic Dissection (IRAD)

Circulation. 2010 Sep 28;122(13):1283-1289.

Santi Trimarchi, MD,1 Kim A. Eagle, MD,2 Christoph A. Nienaber, MD,3 Reed E. Pyeritz, MD,4 Frederik H.W. Jonker, MD,5 Toru Suzuki, MD, 6 Patrick T. O’Gara, MD,7 Stuart J. Hutchinson,8 MD, Vincenzo Rampoldi, MD,1 Viviana Grassi, MD,1 Eduardo Bossone, MD,9 Bart E. Muhs, MD, PhD,5 Arturo Evangelista, MD,10 Thomas T. Tsai, MD,2 Jim B. Froehlich, MD, 2 Jeanna V. Cooper, MS,2 Dan Montgomery, MS,2 Gabriel Meinhardt, MD,11 Truls Myrmel, MD,12 Gilbert R. Upchurch, MD,2 Thoralf M. Sundt,13 MD, Eric M. Isselbacher, MD14

On behalf of the International Registry of Acute Aortic Dissection (IRAD) investigators

1 Policlinico San Donato IRCCS, San Donato Milanese, Italy 2 University of Michigan Health System, Ann Arbor, USA 3 University Hospital Rostock, Rostock, Germany 4 University of Pennsylvania, Philadelphia, PA, USA 5 Yale University School of Medicine, New Haven, Conn, USA 6 University of Tokyo, Tokyo, Japan 7 Brigham and Women’s Hospital, Boston, Mass, USA 8 St Michael’s Hospital, University of Toronto, Toronto, Canada 9 National Research Council, Lecce, Italy 10 Hospital General Universitari Vall d’Hebron, Barcelona, Spain 11 Robert-Bosch Krankenhaus, Struttgart, Germany 12 Tromsø University Hospital, Tromsø, Norway 13 Mayo Clinic, Rochester, Minn, USA 14 Massachusetts General Hospital, Boston, Mass, USA Chapter 6

R1 AbSTRACT R2 R3 background R4 In patients with acute type B aortic dissection (ABAD), presence of recurrent or refractory R5 pain and/or refractory hypertension on medical therapy are sometimes used as an indication R6 for invasive treatment. The International Registry of Acute Aortic Dissection (IRAD) was used R7 to investigate the impact of refractory pain and/or refractory hypertension on the outcomes of R8 ABAD. R9 R10 Methods and results R11 365 patients affected by uncomplicated ABAD, enrolled in the IRAD from 1996-2004, were R12 categorized according to risk profile in two groups. Patients with recurrent and/or refractory pain R13 or refractory hypertension (group I, n=69), and patients not presenting clinical complications R14 at presentation (group II, n=296). “High risk” patients with one or more of the following: R15 shock, periaortic hematoma, spinal cord ischemia, pre-op mesenteric ischemia/infarction, acute R16 renal failure, and limb ischemia were excluded from this analysis. In-hospital management and R17 outcomes were compared between both groups. R18 The overall in-hospital mortality was 6.5 %, and was increased in group I compared to group II, R19 17.4 % vs. 4.0%, p=0.0003. The in-hospital mortality after medical management was significantly R20 increased in group I compared with group II (35.6% vs. 1.5%, p=0.0003). Mortality rates R21 after surgical (20% vs. 28%, p=0.74), or endovascular management (3.7% vs. 9.1%, p=0.50), R22 did not differ significantly between group I and group II, respectively. A multivariable logistic R23 regression model confirmed that recurrent and/or refractory pain or refractory hypertension was R24 an independent predictor of in-hospital mortality (OR 3.31, 95% CI 1.04-10.45, p=0.041). R25 R26 Conclusions R27 In ABAD uncomplicated patients, medical therapy is associated with excellent in-hospital R28 outcomes. Recurrent pain and/or refractory hypertension appeared as clinical signs associated R29 with increased in-hospital mortality, particularly when managed medically. R30 These observations suggest that aortic intervention, such as via an endovascular approach, may R31 be indicated in this intermediate risk group. R32 R33 R34 R35 R36 R37 R38 R39

72 Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection

INTRODUCTION R1 R2 Acute type B aortic dissection (ABAD) is associated with various clinical complications which R3 affect in-hospital outcomes. Currently, surgical or endovascular approaches are often advocated R4 for complications of ABAD, including shock, major organ malperfusion, limb ischemia, R5 periaortic bleeding, and rapidly expanding false lumen.1-8 Such complications are associated with R6 in-hospital mortality rates between 20% and 30% after surgery1, 7, 9-11 and 10% to 20% after R7 endovascular management.7, 12-18 However, some clinical conditions, such as recurrent/refractory R8 pain or refractory hypertension, may be signs of extending dissection or impending rupture R9 but have not resulted in any hemodynamic alteration or organ ischemia. The optimal approach R10 for these patients, either medical, endovascular stenting, or surgery, is still debated.19 Thoracic R11 endovascular aortic repair (TEVAR) is frequently adopted, but the actual evidence to support R12 this practice for ABAD patients with recurrent pain and/or hypertension is limited. R13 In order to better define the importance of refractory pain and/or refractory hypertension in R14 ABAD, and the optimal approach for these patients, we performed a comprehensive analysis of R15 patients presenting ABAD with only pain or refractory hypertension, but no other complications, 6 R16 enrolled in the International Registry of Acute Aortic Dissection (IRAD). In-hospital outcomes R17 of medical, endovascular or surgical management were compared between ABAD patients R18 presenting with and without pain and/or refractory hypertension. R19 R20 R21 METHODS R22 R23 Patient selection R24 Patients presenting ABAD enrolled in the IRAD between January 1996 and December 2004 R25 were investigated. IRAD is an ongoing multi-national multi-center registry that includes enrolled R26 patients with acute aortic dissection at 24 large referral centers (IRAD centers; online supplemental R27 material). The rationale and methods used in IRAD have previously been described.20 Acute type R28 B aortic dissection was defined as any acute aortic dissection involving the descending aorta R29 without any entry tear in the ascending aorta and/or in the aortic arch, presenting within 14 R30 days of symptom onset. Intramural hematoma (IMH) was defined as presence of a regionally R31 thickened aortic wall in the absence of evidence of a double lumen and/or intimal flap regardless R32 of imaging modality.11 For this analysis, both classic dissection and acute IMH were included in R33 the study cohort. R34 Patients were categorized according to risk profile in two groups. Patients with recurrent/ R35 refractory pain or refractory hypertension but no other clinical complications, were defined as R36 intermediate risk patients (group I). Patients without any clinical complications at presentation R37 were categorized as low risk or uncomplicated (group II). “High risk” patients with one or R38 R39

73 Chapter 6

R1 more of the following complications: shock, periaortic hematoma, spinal cord ischemia, pre- R2 op mesenteric ischemia/infarction, acute renal failure, limb ischemia were excluded from this R3 analysis (n=191). In-hospital outcomes according to presence and type of complication (low risk R4 vs. intermediate risk) and method of treatment were analyzed. The study was approved by the R5 institutional review committee at all participating IRAD institutions. R6 R7 Data collection R8 Data were collected using a standardized data form of 290 clinical variables including patient R9 demographics, history, clinical presentation, physical findings, imaging studies, management, in- R10 hospital mortality and adverse events. Completed data forms were forwarded to the coordinating R11 center at the University of Michigan. Data forms were reviewed for internal validity and R12 completeness of data, and were then entered into an Access database. For this analysis, 365 R13 ABAD patients were analyzed, of which 69 met our criteria as intermediate risk. R14 R15 Statistical Analysis R16 Data are shown as frequencies and percentages, and mean + standard deviation or as a median R17 if appropriate. Categorical variables were compared between ABAD patients presenting with R18 and without refractory pain and/or refractory hypertension using the chi-square test and Fisher’s R19 exact test when appropriate. Continuous variables with an approximately normal distribution R20 were compared between both groups using the Students t-test; other continuous variables were R21 investigated using the Mann-Whitney U test. A multiple logistic regression model was fitted R22 adjusting for the known predictors of in-hospital mortality for ABAD.8 A P value < 0.05 was R23 considered significant. SAS 8.1 software (Cary, NC) was used for all analyses. The authors had R24 full access to the data and take full responsibility for its integrity. All authors have read and agree R25 to the manuscript as written. R26 R27 R28 RESUlTS R29 R30 Patient population R31 Between 1996 to 2004, 69 patients (18.9%) with ABAD presenting with refractory pain and/ R32 or hypertension but no other significant clinical complication were identified (group I). During R33 the same period, 296 patients (81.1%) with uncomplicated ABAD were observed (group II). R34 The mean age was 63.5 ±14 years, and 32.6% (n=119) were females. There were no significant R35 differences between groups in demographics or patient’s history (Table 1), except for pre-existing R36 hypertension (89.7% vs. 72.7%, p=0.003) and Marfan syndrome (7.3% vs. 2.1%, p= 0.03) R37 which were more frequently present in group I. R38 R39

74 Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection

In group I (intermediate group), patients presented more frequently with an abrupt onset of pain, R1 migrating pain, and radiating pain (Table 1). Patients in group I underwent a higher number of R2 diagnostic imaging tests (2.34 vs. 1.99; p=0.02), including aortography and magnetic resonance R3 imaging (MRI). Complete thrombosis of the false lumen was more frequently detected in group R4 II (GI 1.8% vs. GII 15.9%; p=0.005), while patients in Group I tended to have more frequently R5 partial false lumen thrombosis. A trend for involvement of arch vessels and abdominal vessels was R6 observed in patients with pain and/or hypertension (Table 1). In addition, in group I, a larger R7 mean aortic diameter (4.75 cm vs. 4.32 cm; p=0.08), as well as a higher incidence of descending R8 thoracic aorta > 6 cm (16.4% vs. 6.7%; p=0.02) were detected (Table 2). Patients in group II R9 were more likely to have a normal chest X-ray (CXR), while widened mediastinum was more R10 often seen in group I. R11 R12 Table 1: Demographics, history of patients and clinical presentation R13 Variable Overall Group I Group II p-value R14 N (%) Intermediate Non-Complicated R15 N (%) N (%) 6 R16 No Patients 365 (100) 69(18.9) 296 (81.1) Demographics R17 Age-mean (years, ± SD), 63.5 (±13.8) 63.5 (±15.0) 64.3 (±13.1) 0.71 R18 Age > 70 years 145 (39.7) 28 (40.6) 117 (39.5) 0.87 R19 Gender female 119 (32.6) 19 (27.5) 100 (33.8) 0.31 R20 Transferred from other hospital 238 (67.6) 50 (74.6) 188 (66.0) 0.17 R21 Etiology and patients’ history Marfan syndrome 11 (2.1) 5 (7.3) 6 (2.1) 0.03 R22 Hypertension 274 (75.9) 61 (89.7) 213 (72.7) 0.003 R23 Atherosclerosis 118 (33.4) 22 (33.3) 96 (33.4) 0.98 R24 Bicuspid aortic valve 4 (1.7) 0 (0.0) 4 (2.1) 0.08 R25 Iatrogenic dissection 12 (3.5) 0 (0.0) 12 (4.3) 0.41 Prior aortic dissection 29 (8.1) 7 (10.3) 22 (7.6) 0.47 R26 Prior aortic aneurysm 72 (20.1) 13 (19.1) 59 (20.3) 0.82 R27 Diabetes 17 (4.8) 3 (4.4) 14 (4.9) 0.86 R28 Prior cardiac surgery 69 (19.8) 13 (19.7) 56 (19.9) 0.96 R29 Clinical presentations and signs Abrupt onset of pain 294 (83.3) 59 (92.2) 235 (81.3) 0.03 R30 Migrating pain 68 (20.0) 22 (35.5) 46 (16.6) 0.0008 R31 Radiating pain 127 (36.8) 32 (51.6) 95 (33.6) 0.007 R32 Time interval until diagnosis (h) 11.4 19.7 10.5 0.23 R33 Intermediate includes those with recurrent pain, refractory pain, and refractory hypertension, and absence R34 of complications such as pre-op mesenteric isch/infarct, acute renal failure, limb ischemia, spinal cord R35 ischemia; signs of aortic dissection shock; imaging results periaortic hematoma. Non-complicated includes those with none of intermediate and/or complicatation conditions. Time interval until diagnosis describes R36 the median time interval from onset of symptoms until the diagnosis in hours. SD: standard deviation. R37 R38 R39

75 Chapter 6

R1 Table 2: Diagnostic imaging studies and findings R2 Variable Overall Group I Group II p-value R3 N (%) Intermediate Non-complicated N (%) N (%) R4 Diagnostic imaging studies R5 No. of studies per patient 2.06 2.34 1.99 0.02 R6 Any imaging study R7 TEE 229 (70.9) 43 (68.2) 186 (71.5) 0.60 R8 CT 345 (96.1) 65 (95.6) 280 (96.2) 0.80 MRI 85 (27.9) 26 (43.3) 59 (24.1) 0.002 R9 Aortogram 91 (29.7) 26 (41.9) 65 (26.6) 0.01 R10 First Diagnostic Test R11 TEE/TTE 50 (15.0) 10 (15.6) 40 (14.9) 0.87 R12 CT 265 (79.6) 54 (84.4) 211 (78.4) 0.28 MRI / Aortogram 17 (5.1) 0 (0.0) 17 (6.3) 0.03 R13 Diagnostic imaging findings R14 Arch vessel involvement 17 (5.6) 6 (10.0) 11 (4.5) 0.09 R15 Abdominal vessel involvement 107 (29.6) 25 (36.2) 82 (28.1) 0.18 R16 Intramural hematoma 40 (11.4) 6 (8.8) 34 (12.0) 0.45 R17 False lumen thrombosis 150 (53.2) 28 (50.0) 122 (54.0) 0.59 Complete thrombosis 37 (13.1) 1 (1.8) 36 (15.9) 0.005 R18 Partial thrombosis 113 (40.1) 27 (48.1) 86 (38.1) 0.16 R19 Mean aortic arch diameter (cm, ±SD) 3.68 (±0.9) 3.62 (±1.0) 3.72 (±0.9) 0.56 R20 Aortic arch >6 cm 6 (29.3) 2 (4.8) 4 (2.4) 0.42 R21 Mean desc. aorta diameter (cm, ±SD) 4.41 (±1.3) 4.75 (±1.7) 4.32 (±1.2) 0.08 Descending aorta >6 cm 23 (8.8) 9 (16.4) 14 (6.7) 0.02 R22 Intimal tear desc aorta 111 (39.2) 27 (43.5) 84 (38.0) 0.42 R23 Site origin left subcl 179 (53.4) 30 (44.8) 149 (51.7) 0.30 R24 Site origin desc aorta 87 (26.0) 16 (23.9) 71 (24.6) 0.89 R25 Site origin abdominal 19 (5.7) 3 (4.5) 16 (5.6) 0.72 Chest X Ray R26 Normal 76 (22.3) 8 (12.5) 68 (24.5) 0.03 R27 Pleural effusion 46 (14.0) 12 (19.0) 34 (12.8) 0.20 R28 Widened mediastinum 163 (48.5) 44 (67.7) 119 (43.9) 0.006

R29 TEE: trans-esophageal echocardiography, TTE: trans-thoracic echocardiography, CT: computerized R30 tomography, MRI: magnetic resonance imaging. SD: standard deviation. R31 R32 In-Hospital Management and Outcomes R33 In total, 75.9% of patients were managed medically, 13.7% underwent surgery, and 6.5% of R34 patients were treated with endovascular methods. Patients in group I were more frequently R35 managed with surgery (36.2% vs. 8.4%, p<0.001) or endovascular methods (39.1% vs. 3.7%, R36 p<0.001) than patients in group II (Table 3). Medical management was more often offered to R37 patients in group II compared to group I (87.8% vs. 24.6%, p<0.001). In group I, the median R38 time interval between onset of symptoms and any invasive treatment was longer compared to the R39

76 Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection

low risk group (240 vs. 100 hours, p=0.005), as well as for the interval between diagnosis and R1 any invasive treatment (236 vs. 72 hours, p=0.004, Table 3). R2 The overall in-hospital mortality was 6.5 %, but was significantly increased in group I compared R3 to group II, 17.4 % vs. 4.0%, p=0.0003 (Table 3, Figure 1). Among group I, the in-hospital R4 mortality rate was 35.6% after medical management, 20% after surgical management and 3.7% R5 after endovascular management (p=0.019). The in-hospital mortality after medical management R6 was significantly higher in group I compared with group II (35.6% vs. 1.5%, p=0.0003, Figure R7 2). Among the six patients who expired after medical management in group 1, aortic rupture R8 was the cause of death in four cases. Mortality rates after surgical (20% vs. 28%, p=0.74), or R9 endovascular management (3.7% vs. 9.1%, p=0.50), did not differ significantly between the two R10 groups (Table 3). R11 R12 R13 R14 R15 6 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 Figure 1. Overall in-hospital mortality rates in the low-risk and intermediate risk groups R30 The intermediate risk group consists of ABAD patients with recurrent/refractory pain or refractory hypertension but no other clinical complications R31 R32 R33 R34 R35 R36 R37 R38 R39

77 Chapter 6

R1 Table 3. In-hospital management and mortality of patients with intermediate risk of type B R2 dissection R3 Variable Overall Group I Group II p-value R4 N (%) Intermediate Non-complicated N (%) N (%) R5 Definitive Management R6 Medical Rx 277 (75.9) 17 (24.6) 260 (87.8) <0.001 R7 Surgery 50 (13.7) 25 (36.2) 25 (8.4) <0.001 R8 Endovascular 38 (10.4) 27 (39.1) 11 (3.7) <0.001 R9 Time interval until invasive treatment From onset of symptoms (h) 211.5 240.0 99.9 0.005 R10 From diagnosis (h) 168.0 236.2 72.0 0.004 R11 Mortality 24 (6.5) 12 (17.4) 12 (4.0) 0.0003 R12 Medical 10 (3.7) 6 (35.6) 4 (1.5) 0.0003 R13 Surgical 12 (24.0) 5 (20.0) 7 (28.0) 0.74 Endovascular 2 (5.2) 1 (3.7) 1 (9.1) 0.50 R14 R15 The time interval until invasive treatment describes the median time interval in hours between onset of R16 symptoms or diagnosis until invasive treatment. R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 Figure 2. In-hospital mortality rates in the low-risk and intermediate risk groups after medical management The intermediate risk group consists of ABAD patients with recurrent/refractory pain or refractory R37 hypertension but no other clinical complications R38 R39

78 Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection

We used the multiple logistic regression model to examine the relationship of refractory pain R1 and/or refractory hypertension, and mortality after adjusting for the effects of known predictors R2 of in-hospital mortality in the overall cohort of 365 ABAD patients. The model suggested that R3 recurrent and/or refractory pain or refractory hypertension (intermediate risk group) was an R4 independent predictor of in-hospital mortality (OR 3.31, p=0.041). In this cohort, age ≥ 70 R5 years (OR 5.11, p=0.004) and absence of chest pain at admission (OR 3.49, p=0.048) were R6 predictors of death as well (Table 4). A plot of the observed vs. the predicted mortality confirmed R7 that these three factors discriminate well in death prediction and that there was little departure R8 from a good fit with the data; Hosmer and Lemeshow Test: Chi-square (5df) = 5.05, p=0.655 R9 (Figure 3). R10 R11 Table 4. Multivariable logistic regression model for known predictors of in-hospital mortality.8 R12 Variable Odds ratio 95% CI P Value R13 Age ≥70 years 5.11 1.70 – 15.39 0.004 R14 Female gender 1.04 0.37 – 2.95 0.932 R15 Hypotension 3.27 0.33 – 31.77 0.306 6 Absence of chest pain 3.49 1.01 – 12.09 0.048 R16 Intermediate risk group 3.31 1.04 – 10.45 0.041 R17 Abdominal vessel involvement 0.78 024 – 2.54 0.684 R18 Hosmer and Lemeshow Test: Chi-square (5df) = 5.05, p=0.655. R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 Figure 3. Plot of observed versus predicted deaths using the multivariable logistic regression model R38 Hosmer-Lemeshow Chi-sq (5 df): 5.05, p = 0.655 R39

79 Chapter 6

R1 DISCUSSION R2 R3 Patients presenting with ABAD have traditionally been categorized as uncomplicated, for which R4 medical treatment has been accepted as an adequate mode of therapy with mortality rates R5 between 1% and 6%, and complicated with features such as rupture, spinal cord ischemia, acute R6 renal failure, mesenteric or limb ischemia, all of which typically require surgical or endovascular R7 intervention and even with optimal treatment are associated with mortality rates of 20% to 30% R8 after sugery1, 7, 9-11 and 10% to 20% after endovascular management.7, 12-18 R9 For those patients in the low risk group, we observed that medical management is associated with R10 excellent in-hospital results (mortality rate 1.5%), appearing as the safest in-hospital therapy in R11 the absence of classic complications, refractory hypertension and/or pain. The significance of R12 ABAD with refractory hypertension and/or refractory or recurrent pain, in the absence of other R13 complications, is currently less well defined. Although some authors have suggested that these R14 signs/symptoms may not result in poorer outcomes and may be treated with medical management R15 and careful monitoring alone,19 others have argued that refractory pain or hypertension are R16 foreshadowing impending rupture and adverse outcomes, and therefore should be considered for R17 more aggressive intervention.5, 11, 21 R18 The IRAD database offers a unique opportunity to analyze the outcomes in large numbers of R19 this subset of ABAD patients. In the present analysis, we observed an in-hospital mortality of R20 17% among patients with refractory hypertension and/or pain, which was significantly higher R21 compared to patients without these symptoms (4%, p=0.0003), but consistently lower than R22 the in-hospital mortality of “classic” complicated ABAD.1, 7-11 After adjusting for known R23 risk factors,8 refractory pain and/or hypertension was an independent predictor for in-hospital R24 mortality. ABAD patients presenting with refractory hypertension and/or pain symptoms, in R25 the absence of other complications, therefore are at intermediate risk for an adverse in-hospital R26 outcome. R27 We observed several differences in patient’s history, presentation and diagnostic imaging findings R28 that may have contributed to the poor outcomes of the patients with refractory hypertension R29 and/or pain which have also been reported by others.8, 9, 22 These include a history of Marfan R30 syndrome (p=0.03), a larger descending thoracic aortic diameter (p=0.08), partial thrombosis of R31 the false lumen (p=0.16), abdominal vessel involvement (p=0.18), and a widened mediastinum R32 on CXR (p=0.006). Interestingly, patients in the intermediate group also presented more R33 frequently with an abrupt onset of pain (p=0.03), and with migrating pain (p=0.0008), which R34 on univariate analyses were associated with decreased mortality,8 perhaps because such clinical R35 signs led to an earlier diagnosis and a more prompt therapy. R36 R37 R38 R39

80 Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection

In the present study, more invasive treatment of ABAD with refractory hypertension and/or pain R1 was associated with improved outcomes, while approximately one third of the patients managed R2 medically expired. In these patients, the most common cause of death was aortic rupture. In the R3 intermediate risk group, the observed differences in mortality between endovascular and medical R4 management could reflect the effectiveness of endovascular methods, although a selection bias R5 may be present and patients treated with medical management may have had more unfavorable R6 characteristics such as a higher age or no chest pain at admission (table 4). Recent reports have R7 suggested that endovascular management of complicated ABAD provides a better survival than R8 medical treatment or open surgery.7, 12-18 Endovascular approaches are increasingly becoming R9 the first line treatment for complicated ABAD cases, and are increasingly used for uncomplicated R10 and chronic dissections as well.7, 14 This report would suggest that this trend may also be R11 beneficial for patients in the intermediate risk group presenting with refractory hypertension R12 and/or pain. R13 R14 limitations R15 Several considerations are important when interpreting the results of the present study. IRAD 6 R16 is an observational study in which participating centers have different approaches to diagnosis R17 and management, creating potential biases in patient selection, which may be minimized in one R18 single center series. Patients were not randomized to a predetermined management strategy, and R19 the results rather reflect a retrospective observation. In the absence of large randomized trials, R20 and given rapid evolution of endovascular aortic stent graft therapy for acute aortic syndromes, R21 there remains some uncertainty as to the optimal strategy to manage this subset of patients. R22 For this evaluation, we included patients that presented with ABAD between 1996 and 2004. R23 Endovascular methods and treatment strategies have advanced in recent years, and current results R24 of endovascular management may be improved compared with our results. This may further R25 support an endovascular approach for ABAD patients with recurrent pain and/or hypertension. R26 Furthermore, use of in-hospital mortality as an endpoint is necessary and important to patients, R27 however it is not sufficient for full a evaluation of outcomes of patients with acute type B aortic R28 dissection. In particular, long-term survival and quality of life are critical to deciding what is truly R29 the “best” strategy for various cohorts. Further studies are needed to address the best therapeutic R30 approaches for patients presenting with refractory hypertension and/or pain. However, the R31 realistic option to randomize these patients for medical vs. invasive management appears very R32 difficult, given the rarity of the condition, the tendency for each patient to present a truly unique R33 clinical and pathophysiologic state, and the fact that interventional and/or surgical approaches R34 represent a rescue strategy for failed response to medical treatment alone. R35 R36 R37 R38 R39

81 Chapter 6

R1 CONClUSION R2 R3 In ABAD completely uncomplicated patients, medical therapy was associated with excellent in- R4 hospital outcomes. Contrariwise, the presence of recurrent pain and/or refractory hypertension R5 were associated with increased in-hospital mortality, particularly when managed medically. These R6 observations suggest that aortic intervention, such as via an endovascular approach, may be R7 indicated in this intermediate risk group. R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

82 Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection

REfERENCES R1 R2 (1) Elefteriades JA, Lovoulos CJ, Coady MA, Tellides G, Kopf GS, Rizzo JA. Management of R3 descending aortic dissection. Ann Thorac Surg. 1999; 67:2002-2005. R4 (2) Glower DD, Fann JI, Speier RH, Morrison L, White WD, Smith LR, Rankin JS, Miller DC, R5 Wolfe WG. Comparison of medical and surgical therapy for uncomplicated descending aortic R6 dissection. Circulation. 1990; 82:IV39-IV46. R7 (3) Masuda Y, Yamada Z, Morooka N, Watanabe S, Inagaki Y. Prognosis of patients with medically R8 treated aortic dissections. Circulation. 1991; 84:III7-13. R9 (4) Lansman SL, McCullough JN, Nguyen KH, Spielvogel D, Klein JJ, Galla JD, Ergin MA, Griepp R10 RB. Subtypes of acute aortic dissection. Ann Thorac Surg. 1999; 67:1975-1978. R11 (5) Lansman SL, Hagl C, Fink D, Galla JD, Spielvogel D, Ergin MA, Griep RB. Acute type B aortic R12 dissection: surgical therapy. Ann Thorac Surg. 2002; 74:S1833-S1835. R13 (6) Umana JP, Lai DT, Mitchell RS, Moore KA, Rodriguez F, Robbins RC, Oyer PE, Dake MD, R14 Shumway NE, Reitz BA, Miller DC. Is medical therapy still the optimal treatment strategy for R15 patients with acute type B aortic dissections? J Thorac Cardiovasc Surg. 2002; 124:896-910. 6 R16 (7) Fattori R, Tsai TT, Myrmel T, Evangelista A, Cooper JV, Trimarchi S, Li J, Lovato L, Kische S, R17 Eagle KA, Isselbacher EM, Nienaber CA. Complicated acute type B dissection: is surgery still R18 the best option?: a report from the International Registry of Acute Aortic Dissection. JACC R19 Cardiovasc Interv. 2008; 1:395-402. R20 (8) Suzuki T, Mehta RH, Ince H, Nagai R, Sakomura Y, Weber F, Sumiyoshi T, Bossone E, Trimarchi R21 S, Cooper JV, Smith DE, Isselbacher EM, Eagle KA, Nienaber CA. Clinical profiles and outcomes R22 of acute type B aortic dissection in the current era: lessons from the International Registry of R23 Aortic Dissection (IRAD). Circulation. 2003;108:II312-II317. R24 (9) Trimarchi S, Nienaber CA, Tsai TT. In Search For Optimal Treatment Of Complicated Acute R25 Type B Aortic Dissections: Insights from The International Registry Of Aortic Dissection R26 (IRAD). Circulation. 2006; 114:432-433. 31-10-2006. Abstract. R27 (10) Estrera AL, Miller CC, Goodrick J, Porat EE, Achouh PE, Dhareshwar J, Meada R, Azizzadeh R28 A, Safi HJ. Update on outcomes of acute type B aortic dissection. Ann Thorac Surg. 2007; R29 83:S842-S845. R30 (11) Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Bossone E, Tolva V, Deeb MG, R31 Upchurch GR, Jr., Cooper JV, Fang J, Isselbacher EM, Sundt TM, III, Eagle KA. Role and results R32 of surgery in acute type B aortic dissection: insights from the International Registry of Acute R33 Aortic Dissection (IRAD). Circulation. 2006; 114:I357-I364. R34 (12) Khoynezhad A, Donayre CE, Omari BO, Kopchok GE, Walot I, White RA. Midterm results of R35 endovascular treatment of complicated acute type B aortic dissection. J Thorac Cardiovasc Surg. R36 2009; 138:625-631. R37 R38 R39

83 Chapter 6

R1 (13) Patel HJ, Williams DM, Meerkov M, Dasika NL, Upchurch GR, Jr., Deeb GM. Long-term results R2 of percutaneous management of malperfusion in acute type B aortic dissection: implications for R3 thoracic aortic endovascular repair. J Thorac Cardiovasc Surg. 2009; 138:300-308. R4 (14) Kische S, Ehrlich MP, Nienaber CA, Rousseau H, Heijmen R, Piquet P, Ince H, Beregi JP, R5 Fattori R. Endovascular treatment of acute and chronic aortic dissection: midterm results from R6 the Talent Thoracic Retrospective Registry. J Thorac Cardiovasc Surg. 2009; 138:115-124. R7 (15) Feezor RJ, Martin TD, Hess PJ, Jr., Beaver TM, Klodell CT, Lee WA. Early outcomes after R8 endovascular management of acute, complicated type B aortic dissection. J Vasc Surg. 2009; R9 49:561-566. R10 (16) Verhoye JP, Miller DC, Sze D, Dake MD, Mitchell RS. Complicated acute type B aortic R11 dissection: midterm results of emergency endovascular stent-grafting. J Thorac Cardiovasc Surg. R12 2008; 136:424-430. R13 (17) Schoder M, Czerny M, Cejna M, Rand T, Stadler A, Sodeck GH, Gottardi R, Loewe C, Lammer R14 J. Endovascular repair of acute type B aortic dissection: long-term follow-up of true and false R15 lumen diameter changes. Ann Thorac Surg. 2007; 83:1059-1066. R16 (18) Duebener LF, Lorenzen P, Richardt G, Misfeld M, Notzold A, Hartmann F, Sievers HH, Geist V. R17 Emergency endovascular stent-grafting for life-threatening acute type B aortic dissections. Ann R18 Thorac Surg. 2004; 78:1261-1266. R19 (19) Januzzi JL, Movsowitz HD, Choi J, Abernethy WB, Isselbacher EM. Significance of recurrent R20 pain in acute type B aortic dissection. Am J Cardiol. 2001; 87:930-933. R21 (20) Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, Evangelista R22 A, Fattori R, Suzuki T, Oh JK, Moore AG, Malouf JF, Pape LA, Gaca C, Sechtem U, Lenferink R23 S, Deutsch HJ, Diedrichs H, Robles J, Llovet A, Gilon D, Das SK, Armstrong WF, Deeb GM, R24 Eagle KA. The International Registry of Acute Aortic Dissection (IRAD): new insights into an R25 old disease. JAMA. 2000; 283:897-903. R26 (21) Hata M, Sezai A, Niino T, Yoda M, Wakui S, Unosawa S, Umeda T, Shimura K, Osaka S, R27 Furukawa N, Kimura H, Minami K. Prognosis for patients with type B acute aortic dissection: R28 risk analysis of early death and requirement for elective surgery. Circ J. 2007; 71:1279-1282. R29 (22) Tsai TT, Evangelista A, Nienaber CA, Myrmel T, Meinhardt G, Cooper JV, Smith DE, Suzuki R30 T, Fattori R, Llovet A, Froehlich J, Hutchison S, Distante A, Sundt T, Beckman J, Januzzi JL, Jr., R31 Isselbacher EM, Eagle KA. Partial thrombosis of the false lumen in patients with acute type B R32 aortic dissection. N Engl J Med. 2007; 357:349-359. R33 R34 R35 R36 R37 R38 R39

84 Chapter 7

Acute abdominal aortic dissection: Insight from the International Registry of Acute Aortic Dissection (IRAD)

Journal of Vascular Surgery 2007 Nov;46(5):913-919.

Santi Trimarchi, MD,1 Thomas Tsai, MD,2 Kim A. Eagle, MD,2 Eric M. Isselbacher, MD,3 Jim Froehlich, MD,2 Jeanna V. Cooper, MS,2 Vincenzo Rampoldi, MD,1 Gilbert R. Upchurch Jr, MD,2

On behalf of the International Registry of Acute Aortic Dissection (IRAD) investigators

1 Policlinico San Donato IRCCS, San Donato Milanese, Italy 2 University of Michigan Health System, Ann Arbor, USA 3 Massachusetts General Hospital, Boston, USA Chapter 7

R1 AbSTRACT R2 R3 background R4 Isolated acute dissection of the abdominal aorta is an unusual event that may present with several R5 different clinical scenarios. Because its incidence is low, the natural history is unknown. We R6 report data from the International Registry of Acute Aortic Dissection (IRAD), the largest group R7 of patients treated for acute aortic dissections. The aim of this study was to identify clinical R8 characteristics, therapeutic approaches, risk factors for mortality, in-hospital outcome, and long- R9 term results of this cohort, thus clarifying its natural history. R10 R11 Methods R12 A comprehensive analysis of 290 clinical variables on 18 patients affected by isolated acute R13 abdominal aortic dissection (IAAAD) was performed. Among 1417 patients enrolled in the R14 IRAD from 1996 to 2003, 532 (37.5%) had an acute type B dissection, of which 18 (1.3%) R15 had an IAAAD. The mean age was 67.7 ± 13.3 years, with a male predominance (n = 12, 67%). R16 Aortic aneurysms pre-existed in 5 patients (28%). IAAAD was iatrogenic in 2 cases (11%). R17 R18 Results R19 Compared with patients with type B aortic dissections, abdominal pain, mesenteric ischemia R20 or infarction, limb ischemia, and hypotension as initial clinical signs were significantly more R21 frequent in patients with IAAAD, whereas chest pain was more typical in patients with type R22 B dissections. No neurologic symptoms, such as ischemic spinal cord damage or ischemic R23 peripheral neuropathy, occurred in the IAAAD cohort. The 18 IAAAD patients were medically, R24 surgically, or percutaneously managed in 12 (66.6%), five (27.8%), and one (5.6%) cases, R25 respectively. The overall in-hospital mortality rate was 5.6% (n = 1). The patient who died was R26 medically managed. No deaths were reported among patients who underwent surgery or had an R27 endovascular procedure, irrespective of their preoperative status. A mean follow-up of 5 years R28 (range, 1 month to 9 years) was completed for 71% (12 of 17) of the patients. Four patients R29 (33.3%) died during the 9-year follow-up period. Overall survival was 93.3% ± 12.6% at 1 year R30 and 73.3% ± 27.2% at 5 years. All patients who died during the follow-up period had in-hospital R31 medical management (P = .04). R32 R33 Conclusions R34 IAAAD is a condition that may present differently compared with classic type B aortic dissections. R35 IAAAD patients treated with surgical or endovascular procedures had a lower unadjusted in- R36 hospital and long-term mortality rate compared with medically managed patients. On the basis R37 of the present natural history report, continued surveillance appears mandatory. To improve the R38 life expectancy of patients with IAAAD, aggressive surgical or endovascular management seems R39 justified.

86 Acute Abdominal Aortic Dissection

INTRODUCTION R1 R2 Isolated acute abdominal aorta dissection (IAAAD) is a rare event, usually related to spontaneous, R3 traumatic, or iatrogenic causes.1-3 The clinical presentation of IAAAD may be associated with R4 abdominal pain, visceral ischemia, acute renal failure, and limb ischemia. 1-4 We report data from R5 the International Registry of Acute Aortic Dissection (IRAD). These data represent a unique R6 opportunity to study the largest group of consecutive patients treated for aortic dissection R7 by 18 aortic referral centers throughout the world. The aim of this study was to clarify the R8 natural history of this specific cohort, identifying clinical characteristics, therapeutic approaches, R9 risk factors for mortality, in-hospital outcome, and long-term results. Also documented were R10 differences in demographic, history, clinical presentation, diagnostic imaging findings, indication R11 for surgical or endovascular treatment, and in-hospital management compared with the classic R12 type B dissection. R13 R14 R15 METHODS R16 R17 Patient selection 7 R18 Patients having IAAAD enrolled in the IRAD between January 1, 1996, and December 31, 2003 R19 were included in the present analysis. IRAD is an ongoing multinational multicenter registry R20 started in 1996 that includes patients with acute aortic dissection at 18 large IRAD referral R21 centers (Appendix, online only). The rationale and design of this registry, which excludes patients R22 with traumatic dissections, has previously been described.5 IAAAD was defined as any acute aortic R23 dissection involving the aorta below the diaphragm only and presenting ≤14 days of symptom R24 onset.5,6 Intramural hematoma (IMH), defined as presence of a regionally thickened aortic wall R25 in the absence of evidence of a double lumen or intimal flap, or both, regardless of imaging R26 modality,7 was also considered in the study cohort. A comprehensive analysis of 290 clinical R27 variables was completed on patients classified as having spontaneous or iatrogenic IAAAD. At R28 each enrolling hospital, study investigators obtained approval from their ethics committee or R29 institutional review board to participate in IRAD and its follow-up study. R30 R31 Data collection R32 Data were collected prospectively after presentation or retrospectively from hospital records R33 using a standardized data form of 290 clinical variables, including patient demographics, history, R34 clinical presentation, physical findings, imaging studies, medical and surgical management, R35 in-hospital mortality, and adverse events. Yearly follow-up data were obtained after discharge R36 using a standardized form that includes clinical, imaging, and vital signs data. Completed data R37 forms were forwarded to the coordinating center at the University of Michigan. Data forms R38 R39

87 Chapter 7

R1 were reviewed for analytical internal validity and scanned electronically into an Access database R2 (Microsoft Corp, Redmond, Wash). For this analysis, 18 patients with isolated acute aortic R3 dissection confined to the subdiaphragmatic aortic level were analyzed. Long-term follow up was R4 completed on 71% of the enrolled patients. R5 R6 Imaging R7 IAAAD was identified using one diagnostic examination in 16.7% (n = 3) of patients or two or R8 more in 83.3% (n = 15). Computed tomography (CT) imaging was used in 16 patients (88.9%), R9 being used as first diagnostic exam in 13 (72.2%), and an angiogram was used in nine (50%), of R10 which in two it was first exam. Magnetic resonance imaging (MRI) was used in four patients, as R11 was ultrasound imaging, which was the first diagnostic method in three patients. Involvement of R12 visceral or renal was specifically evidenced with diagnostic examinations in five patients R13 (MRI, 2; spiral CT, 2; ultrasound, 1). Among this cohort, one patient was treated surgically R14 after presenting with visceral and limb ischemia. All of these patients were discharged from the R15 hospital. R16 R17 Statistical analysis R18 Data are shown as frequencies and percentages, and mean ± standard deviation or as a median R19 (interquartile range). Missing data were not defaulted to negative, and denominators reflect only R20 reported cases. Associations of death among nominal variables were compared using the χ²-test R21 and the Fisher exact test, when appropriate, and among continuous variables using the Student t R22 test. Long-term survival using Kaplan-Maier methods was reported. R23 R24 R25 RESUlTS R26 R27 Patient population R28 Of 1417 patients with IAAAD enrolled in IRAD from 1996 to December 2003, 885 (62.5%) R29 had a type A dissection and 532 (37.5%) had a type B dissection, and 18 of these patients (1.3%) R30 had an acute IAAAD. They were compared with 514 patients with classic type B dissection. The R31 mean age in the IAAAD cohort was 67.7 ± 13.3 years, with men pre- dominating (n = 12, 67%). R32 Thirteen patients (72%) were transferred to an IRAD center from a referral hospital for definitive R33 treatment. R34 History of hypertension was present in 14 patients (78%), pre-existing atherosclerosis in seven R35 (41%), and diabetes in one (5.6%). An aortic aneurysm repair was previously performed in five R36 patients (28%). Five patients had an abdominal aortic graft replacement, and one patient had R37 a previous femorofemoral bypass Dacron graft. The extent of aneurysm repair was unavailable R38 from the database. R39

88 Acute Abdominal Aortic Dissection

IAAAD was considered to be an iatrogenic complication of a cardiac catheterization in two R1 patients (11%). None of the patients in the IAAAD cohort had Marfan syndrome. In addition, R2 none of the patients sustained any prior aortic dissection or had undergone aortic re- placement R3 for aneurysm or dissection (Table I). R4 R5 Table 1. Demographics and history of patients with abdominal aortic dissection R6 Abdominal Type B P-Value R7 (n =18) dissection R8 Variable* (n = 514) R9 Demographics Age y (mean ± SD) 67.7 ± 13.3 64.0 ± 13.6 .15 R10 Male Gender 12 (66.7) 355 (69.1) .82 R11 Transferred 13 (72.2) 337 (67.8) .69 R12 Etiology, patient history R13 Hypertension 14 (77.8) 391 (76.9) .93 R14 Athersclerosis 7 (41.2) 180 (35.8) .65 Diabetes 1 (5.6) 33 (6.50 .86 R15 Previous aortic aneurysm repair 5 (27.8) 92 (18.3) .3 R16 Iatrogenic dissection 2 (11.1) 11 (2.3) .07 R17 Marfan syndrome 0 (0.0) 18 (3.5) .41 R18 Prior aortic dissection 0 (0.0) 42 (8.3) .2 7 Prior aortic replacement for aneurysm and/or dissection 0 (0.0) 70 (14.2) .15 R19 Previous surgical coronary revascularization 2 (11.1) 21 (4.3) .19 R20 R21 *Categoric data are expressed as number (percentage) R22 Presenting and diagnostic signs and symptoms R23 Abdominal pain (P =.01), mesenteric ischemia or infarction (P = .03), and limb ischemia (P = .04) R24 were significantly more frequent in the IAAAD cohort than in the group with type B dissection. R25 In IAAAD, a clinical hypertensive state (systolic blood pressure >150/90 mmHg) appeared at R26 presentation in nine patients (53%). Hypotension (systolic blood pressure <90/60 mmHg) was R27 not initially detected in the total cohort, but appeared in 2 patients during hospitalization. This R28 condition was more often present in IAAAD patients (P .03; Table II). A CT scan was used to R29 detect the abdominal dissections in 16 patients (88.9%), and its use was initially adopted as the R30 first diagnostic exam in 13 (72.2%). Involvement of visceral or renal arteries was present in five R31 patients: the diagnosis was made with MRI in two, spiral CT in two, and ultrasound imaging R32 in one. R33 An aortic aneurysm was detected in seven patients (41%). An intramural hematoma was detected R34 in 1 patient (6%), and findings of patent, partial thrombosis, or complete thrombosis of the false R35 lumen were evidenced in 9 (53%), 6 (35%), and 2 (12%) patients, respectively (Table II). R36 R37 R38 R39

89 Chapter 7

R1 Table 2. Clinical characteristics and diagnostic findings of patients with abdominal aortic R2 dissection R3 Abdominal Type B dissection P-Value R4 Variable, No. (%) (n =18) (n = 514) R5 Characteristics of pain Abrupt onset 14 (82.4) 430 (86.0) .72 R6 Back pain 10 (62.5) 339 (67.6) .66 R7 Chest pain 7 (41.2) 365 (72.1) .01 R8 Abdominal pain 11 (68.7) 176 (35.8) .01 R9 Leg pain 3 (21.4) 42 (8.7) .12 Clinical signs R10 Mesenteric ischemia or infarction 3 (21.4) 22 (4.7) .03 R11 Acute renal failure 2 (13.3) 57 (12.1) .7 R12 Pulse deficit 5 (29.4) 82 (17.8) .22 R13 Limb ischemia 4 (25.0) 36 (7.7) .04 Ischemic spinal cord damage 0 (0.00) 16 (3.2) .99 R14 Syncope 1 (5.9) 18 (3.6) .48 R15 Hypertension 9 (52.9) 342 (68.1) .15 R16 Hypotension 2 (11.8) 14 (2.8) .03 R17 Any diagnostic test showed R18 Aortic aneurysm 7 (41.2) 162 (35.5) .61 Intramural hematoma 1 (5.6) 61 (12.4) .71 R19 Patentl false lumen thrombosis 9 (52.9) 185 (46.2) .58 R20 Partial false lumen thrombosis 6 (35.3) 160 (40.0) .69 R21 Complete false lumen thrombosis 2 (11.8) 55 (13.7) .99 R22 Periaortic hematoma 1 (6.2) 88 (18.5) .32 R23 R24 Surgical or endovascular interventions were performed for aortic rupture, visceral ischemia, limb R25 ischemia, acute renal failure, extension of dissection compromising additional arteries, recurrent R26 or refractory pain, or both, and untreatable hypertension (Table III). In almost half of the R27 invasively treated patients, more than one indication (mean, 2.3) was reported. Three patients R28 with IAAAD presented with visceral ischemia, and surgical aortic replacement was performed in R29 two. An aortic fenestration affecting the visceral segment was performed in one patient, followed R30 by placement of an abdominal aortic endograft. R31 R32 Management R33 Management of IAAAD patients was medical in 12 (66.6%), surgical in five (27.8%) and R34 percutaneous in one (5.6%). Initial medical therapy was with a -blocker in 15 patients (88%) and R35 nitroprusside and a calcium-channel blocker in four (27%). Vasodilators were more commonly R36 used in type B dissections (P = .01), but were also used in three IAAAD patients (19%). R37 Vasopressors were needed in one patient (7%) that was surgically treated. An abdominal aortic R38 dissection developed during cardiac catheterization in a second patient with hypotension. He was R39

90 Acute Abdominal Aortic Dissection

treated medically, but died 2 days latter of cardiac-related complications. Among the six patients R1 who had invasive therapy, procedures were delayed beyond the day of presentation in three cases, R2 with a medium observed interval time between the onset of symptoms and the treatment of R3 14 days: two patients had refractory hypertension not responsive to maximal medical therapy R4 with pain, and one presented with excessive comorbidities (Table III). The abdominal aorta was R5 replaced in five patients, and one patient also had a surgical peripheral vessel management. Aortic R6 stenting with fenestration was performed in one patient. No reoperations were required. R7 R8 Table 3. In-hospital management indications for treatments, and long-term medical therapy of R9 patients with abdominal aortic dissection R10 Abdominal Type B dissection P-Value R11 Variable, No. (%) (n =18) (n = 514) R12 Management R13 Medical 12 (66.6) 358 (69.7) .79 Surgical 5 (27.8) 85 (16.5) .21 R14 Endovascular 1 (5.6) 71 (13.8) .31 R15 Indications for surgical or endovascular treatment R16 Aortic rupture 1 (5.6) 21 (4.2) .79 R17 Visceral ischemia 3 (21.4) 30 (33.3) .18 R18 Acute renal failure 2 (13.3) 72 (15.2) .99 7 Limb ischemia 4 (23.5) 25 (29.4) .17 R19 Extension of dissection into visceral and/or renal arteries 3 (40.0) 65 (54.6) .66 R20 Recurrent or refractory pain 5 (31.2) 55 (50.4) .99 R21 Untreatable hypertension 1 (16.6) 21 (26.6) .99 R22 Initial medical therapy β-blocker 15 (88.2) 394 (80.1) .55 R23 Nitroprusside 4 (26.7) 190 (40.3) .28 R24 Calcium-channel blocker 4 (26.7) 124 (28.4) .99 R25 Vasodilatator 3 (18.8) 229 (50.4) .01 R26 Vasopressor 1 (6.7) 16 (3.7) .45 Delayed procedures 3 (50.0) 89 (63.1) .67 R27 Chronic medical treatment at time of discharge R28 β-blocker 14 (87.5) 401 (88.3) .99 R29 Calcium-channel blocker 9 (60.0) 266 (61.9) .99 R30 Vasodilatator 1 (6.7) 142 (34.3) .03 R31 Vasopressor 2 (13.3) 217 (50.8) .01 R32 Hospital outcomes R33 The overall in-hospital mortality was 5.6% (n = 1). Considering the limited number of patients, R34 no variables were significantly associated with in-hospital mortality. The patient who died had R35 electro-cardiographic (ECG) signs suggestive of a prior myocardial infarction. This patient R36 sustained an iatrogenic dissection during cardiac catheterization, associated with limb ischemia R37 and hypotension. He was medically managed, but died 2 days after this event. No deaths were R38 R39

91 Chapter 7

R1 reported among patients who underwent a surgical or endovascular procedure, irrespective of R2 their preoperative status. Mean hospital length of stay for survivors was 16.9 days: 12.9 days for R3 patients medically managed, and 25 days for those treated surgically. Among IAAAD patients R4 who survived, 88% were discharged with β-blocker therapy, and 60% received calcium-channel R5 blockers. Angiotensin-converting enzyme inhibitors and other vasodilators were more often R6 recommended in those with type B dissection (P = .01 and P = .03, respectively). R7 R8 follow-up R9 In IRAD, a yearly follow-up is conducted with ultrasound scan, CT scan, or MRI and aortogram, R10 depending on the center. Median follow-up was completed on 71% (12 of 17) of discharged R11 surviving patients, and averaged 5 years, (range, 1 month to 9 years). Four patients died (33%) R12 during the 9-year follow-up period. The mean survival rate was 93% ± 13% at 1 year and 73% R13 ± 27% at 5 years. In two patients, death occurred after 1 month and 5.1 years, respectively, and R14 was related to complications from the abdominal dissection. In particular, one patient died 1 R15 month after hospital discharge. She originally presented with an ischemic lower limb, acute renal R16 failure, signs of mesenteric ischemia, and ECG finding of ischemia. She was managed in-hospital R17 with β-blockers and calcium-channel blockers, and was discharged to home after 9 days in stable R18 condition. After 1 month, however, she had a second acute hospitalization for issues related to R19 an incident dissection with worsening acute renal failure. Two other patients died of unknown R20 causes at 1.6 and 7.5 years. Recurrence of new symptoms or abdominal pain was observed in 1 R21 patient and did not correlate with late death. No further aortic enlargement was reported. R22 All patients who died during the follow-up period had in-hospital medical management alone R23 (Fig). The patients who died in follow-up were more likely to be older (mean age, 71 years), to have R24 a known aortic aneurysm (2 of 4), to have a history of hypertension (3 of 4) and atherosclerosis R25 (2 of 4), and to have an altered ECG (3 of 4), with two of those showing signs of ischemia. One R26 of the patients who died during follow-up was the only IAAAD patient with a history of diabetes. R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

92 Acute Abdominal Aortic Dissection

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 figure. Kaplan-Meier survival curve for patients presenting with isolated acute aortic dissection of the R18 abdominal aorta. 7 *Denotes standard error that exceeds 10%. R19 R20 R21 DISCUSSION R22 R23 IAAADs are a rare event. In the current series, 1.3% of all IRAD enrolled patients were identified R24 as having an IAAAD, and previous reports have confirmed the low incidence of the disease.1-4 A R25 recent review of the literature by Farber et al. 1 2004 yielded only 51 patients with this segmental R26 dissection. Because of the large size of the IRAD database, the present experience represents the R27 largest series reported in the literature. R28 In the IRAD IAAAD cohort, history of hypertension was present in more than two-thirds of R29 the patients, similar to rates of type A5 and B8 acute aortic dissections. However, atherosclerosis R30 was present more often in the IAAAD compared with type B, as was the incidence of iatrogenic R31 origin.8 In the IRAD IAAAD experience, two patients sustained a dissection of iatrogenic origin R32 during a cardiac catheterization. Both had a history of atherosclerosis and ischemic cardiac R33 disease, and one died in the hospital. Perhaps patients with atherosclerosis more frequently R34 develop an iatrogenic dissection secondary to irregular calcifications in the aortic intimal wall.9, 10 R35 In such patients, the presence of associated cardiac comorbidities could compromise the chances R36 for a positive outcome, regardless of medical or surgical management. R37 R38 R39

93 Chapter 7

R1 An association with abdominal aneurysms was detected in 28% of the IRAD IAAAD patients, R2 similar to other experiences.1 Aortic dilatation can be the consequence of an aortic dissection R3 and seems not associated with specific new symptoms. 1, 10-12 Detection of an aortic aneurysm is R4 associated with invasive therapy 3,4,13 although in IRAD it was not associated with increased in- R5 hospital mortality. No deaths occurred among the five IRAD IAAAD patients who had an aortic R6 repair; however, one had a contained ruptured with periaortic hematoma that was complicated R7 by hypotension. In the IRAD IAAAD group, periaortic hematoma was detected in one patient R8 (6.2%), while it was present in 19% in the overall IRAD type B group, and was associated with R9 a 32.4% mortality rate.14 It appears reasonable that the coexistence of IAAAD and AAA leads to R10 higher risk for aortic rupture.3,4,13 R11 As expected, patients presenting with hypotension or shock have a poorer outcome. It was 12% R12 in IAADS in IRAD, lower than the rate of 17% to 28% experienced in other groups.1,3,4,15 This R13 high-risk condition was observed in two patients, one of which had an iatrogenic dissection R14 during cardiac catheterization and didn’t survive, whereas the second patient had a ruptured R15 aortic aneurysm that was successfully operated on. As IRAD previously reported in type A and R16 B aortic dissections, the presence of shock is associated with a high mortality, regardless of the R17 treatment choice. The presence of shock obviously may create a bias in evaluating therapeutic R18 options because patient survival is so limited.15, 16 R19 The proximal entry site of the dissection in IAAAD has been observed mostly below the origin R20 of the renal arteries, 1-4 with leg ischemia present in 19% of cases.1-3 In the IRAD IAAAD cohort, R21 it is not possible to estimate if the proximal entry tear was above or below the renal arteries, but R22 a detection of distal ischemia was often present. However, in IRAD, the IAAAD patients had R23 more complex with clinical presentations that included mesenteric ischemia/infarction and acute R24 renal failure. R25 Medical therapies were adopted in two-thirds of patients and surgical therapies in one-third. One R26 patient died in-hospital 2 days after an iatrogenic IAAAD, and another died 1 month after hospital R27 discharge. Both had been medically managed for the presence of ischemic cardiac comorbidities. R28 Abdominal aortic surgery in IAAAD resulted in better results compared with those of thoracic or R29 thoracoabdominal aortic replacement for acute type B dissections, where IRAD recently showed R30 a 29% in-hospital mortality rate. 16 This observation supports a more aggressive invasive surgical R31 or endovascular approach in patients with IAAAD compared with those with a classic type B R32 dissection. Standard surgical abdominal aortic replacement or endovascular management might R33 be performed in those IAAAD patients who are extremely complicated or who have excessive R34 comorbidi- ties, in which invasive management is not considered, as IRAD showed for those R35 having ischemic cardiac diseases. In this latter group surgery is often contraindicated. R36 IRAD IAAAD long-term survival showed that the follow-up mortality rates are high in patients R37 discharged from the hospital, approaching 33% of patients within 8 years. Similar to that for R38 descending thoracic aortic dissection,17-19 30% to 60% of follow-up deaths are from aortic- R39

94 Acute Abdominal Aortic Dissection

related complications such as rupture, extension of dissection, and perioperative mortality from R1 subsesequent repairs.18,19 All IRAD IAAAD patients who died during follow-up received in- R2 hospital medical management, and all were medically treated to lower heart rate, blood pressure, R3 and inotropy. No late deaths or aortic complications were observed among the six patients who R4 were invasively treated with surgery or an endograft. R5 Our study did not contain complete information on cause-specific mortality; however, similar R6 to type B dissections, we observed that comorbidities, such as a history of atherosclerosis and a R7 diseased aorta (i.e., aneurysm) are associated with an accelerated mortality seen during follow- R8 up.19,20 Although IRAD is an observational registry and therefore indications for treatment are R9 not supported from randomized controls, observations from this database could be helpful for R10 physicians who encounter this condition. Data from IRAD suggest the approach to these patients R11 in the presence of unstable clinical conditions should involve prompt surgical or endovascular R12 treatment. R13 Although not supported because of the small sample size, we suggest that an endovascular R14 approach should be considered primarily in the presence of comorbidities such as a history of R15 ischemic cardiac diseases. In stable patients, initial therapy with a β-blocker is recommended, R16 followed by serial follow-up using spiral CT or MRI of the proximal intimal tear and the distal R17 re-entries to consider the technical feasibility of endograft therapy. In the patients with anatomic 7 R18 contraindications to endovascular repair, standard surgical abdominal aortic replacement with R19 extension into iliac or femoral arteries can be performed safely. The use ofβ -blockers can be R20 recommended on the basis of the long-term IRAD results with acute type A and B-dissections. R21 20,21 All of these patients, regardless the type of dissection and the in- hospital management, R22 should be followed up at least yearly with imaging of the dissection. R23 R24 limitations R25 IRAD is an observational series of patients treated in tertiary referral centers, and therefore, R26 its findings may not necessarily be applicable to the total population. In-hospital death is the R27 outcome variable that was assessed in this registry analysis, and it is not sufficient for a full R28 evaluation of outcomes of patients with IAAAD, which should also take into consideration R29 nonfatal adverse events, patient functional status, and use of resources. IRAD has no standard R30 protocol for diagnosis, management, or imaging of dissections, and treatments therefore reflect R31 individual approaches. R32 In addition, all IRAD IAAAD patients in the present study were symptomatic; therefore this R33 study does not lend insight in how to manage the asymptomatic patient with an incidentally R34 discovered IAAAD. R35 Finally, we only had information on follow-up mortality and did not have information on R36 all cause of death. We were also unable to evaluate other end points such as freedom from R37 reoperation, rupture, or redissection. R38 R39

95 Chapter 7

R1 CONClUSIONS R2 R3 IAAAD is a rare condition that presents with significant clinical variability. In the IRAD R4 population, some complications, such as abdominal pain, mesenteric ischemia or infarction, R5 and limb ischemia, were more frequent in patients with IAAAD compared with patients with R6 type B dissection. In the present IRAD experience, IAAAD patients treated with surgical or R7 endovascular procedures had a lower mortality, both in-hospital and at long-term, compared R8 with those patients who had medical management. R9 The natural history of IRAD IAAAD suggests that these patients present similarly to type B R10 dissections, particularly in terms of factors associated with long-term risk of death. This R11 observation leads to the recommendation for serial CT scan surveillance. R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

96 Acute Abdominal Aortic Dissection

REfERENCES R1 R2 1. Farber A, Lautherbach SR, Wagner WH, Cossman DV, Long B, Cohen JL, et al. Spontaneous R3 infrarenal abdominal aortic dissection presenting as claudication: case report and review of the R4 literature. Ann Vasc Surg 2004;18:4-10. R5 2. Hirst A, Jonhs V, Kime W. Dissecting aneurysms of the aorta: a review of 505 cases. Medicine R6 1958;37:217-79. R7 3. Farber A, Wagner WH, Cossman DV, Cohen JL, Walsh DB, Fillinger MF, et al. Isolated R8 dissection of the abdominal aorta: clinical presenta- tion and therapeutic options. J Vasc Surg R9 2002;36:205-10. R10 4. Mozes G, Gloviczki P, Park WM, Schultz HL, Andrews JC. Spontane- ous dissection of the R11 infrarenal abdominal aorta. Semin Vasc Surg 2002;15:128-36. R12 5. Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, et al. The R13 International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. R14 JAMA 2000;283:897-903. R15 6. Daily PO, Trueblood HW, Stinson EB, Wuerflein RD, Shumway NE. Management of acute R16 aortic dissections. Ann Thorac Surg 1970;10: 237-47. R17 7. Erbel R, Alfonso F, Boileau C, Dirsch O, Eber B, Haverich A, et al. Task Force on Aortic 7 R18 Dissection, European Society of Cardiology. Diagnosis and management of aortic dissection. R19 Eur Heart J 2001; 22:1642-81. R20 8. Suzuki T, Metha RH, Huseyin I, Nagai R, Sakomura Y, Weber F, et al. Clinical Profiles and R21 Outcomes of Acute Type B Aortic Dissection in the Current Era: Lesson From the International R22 Registry of Aortic Dissec- tion (IRAD). Circulation 2003;108(suppl II):II 312-7. R23 9. Hagl C, Ergin MA, Galla JD, Spielvogel D, Lansman S, Squitieri RP, et al. Delayed chronic type R24 A dissection following CABG: implica- tions for evolving techniques of revascularization. J Card R25 Surg 2000;15:362-7. R26 10. DeBakey ME, McCollum CH, Crawford ES, Morris, GC, Howell J, Noon GP. Dissection and R27 dissecting aneurysms of the aorta: twenty- year follow-up of five hundred twenty seven patients R28 treated surgically. Surgery 1982;92:1118-33. R29 11. Roberts CS, Roberts WC. Aortic dissection with the entrance tear in the abdominal aorta. Am R30 Heart J 1991;121:1834-5. R31 12. Becquemin JP, Deleuze P, Watelet J, Testard J, Melliere D. Acute and chronic dissections of the R32 abdominal aorta: clinical features and treat- ment. J Vasc Surg 1990;11:397-402. R33 13. Cambria RP, Brewster DC, Moncure AC, Steinberg FL, Abbott WM. Spontaneous aortic R34 dissection in the presence of coexistent or previously repaired atherosclerotic aortic aneurysm. R35 Ann Surg 1988;208:619-624. R36 14. Trimarchi S, Nienaber CA, Tsai T, Llovet A, Froehlich J, Rampoldi V, et al. In search for optimal R37 treatment of complicated acute type B aortic dissection: insights from the International Registry R38 of Acute Aortic Dissection (IRAD). Circulation 2006;114:432-3. R39

97 Chapter 7

R1 15. Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Mehta RH, et al. Contemporary R2 results of surgery in acute type A aortic dissection: the IRAD (International Registry of Acute R3 Aortic Dissec- tion) experience. J Thorac Cardiovasc Surg 2005;129:112-22. R4 16. Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Tolva V, et al. Role and results of R5 surgery in acute type B aortic dissection -insights from the International Registry of Acute Aortic R6 Dissection (IRAD). Circulation 2006;114(1 suppl):I357-64. R7 17. Elefteriades JA, Lovoulos CJ, Coady MA, Tellides G, Kopf GS, Rizzo JA. Management of R8 descending aortic dissection. Ann Thorac Surg 1999;67:2002-5; discussion 2014-9. R9 18. Gysi J, Schaffner T, Mohacsi P, Aeschbacher B, Althaus U, Carrel T. Early and late outcome of R10 operated and non-operated acute dissection of the descending aorta. Eur J Cardiothorac Surg R11 1997;11:1163-9; discussion 1169-70. R12 19. Umana JP, Lai DT, Mitchell RS, Moore KA, Rodriguez F, Robbins RC, et al. Is medical therapy R13 still the optimal treatment strategy for patients with acute type B aortic dissections? J Thorac R14 Cardiovasc Surg 2002;124:896-910. R15 20. Tsai T T, Fattori R, Trimarchi S, Nienaber C, Myrmel T, Hutchinson S, et al. Long-term survival R16 in patients with type B aortic dissection: insights from the International Registry of Acute Aortic R17 Dissection (IRAD). Circulation 2005 112(suppl II):II-535. R18 21. Tsai TT, Fattori R, Trimarchi S, Nienaber CA, Myrmel T, Hutchinson S, et al. Long-term survival R19 in patients with type A aortic dissection: insights from the International Registry of Acute Aortic R20 Dissection (IRAD). Circulation 2006;114 (I suppl):I350-6. R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

98 Chapter 8

Aortic expansion after uncomplicated acute type B aortic dissection

(Submitted)

Frederik HW Jonker, MD, PhD,1 Santi Trimarchi, MD,2 Vincenzo Rampoldi, MD,2 Himanshu Patel, MD,3 Patrick O’Gara, MD,4 Mark D. Peterson, MD, PhD,5 Rossella Fattori, MD,6 Frans L. Moll, MD, PhD,7 Matthias Voehringer, MD,8 Reed E Pyeritz, MD, PhD,9 Stuart Hutchison, MD,10 Dan Montgomery, MS, 2 Eric M. Isselbacher, MD, 3 Christoph A. Nienaber, MD,11 Kim A Eagle, MD 2

1 School of Medicine,Yale University, New Haven, Conn, USA 2 Policlinico San Donato IRCCS, San Donato Milanese, Italy 3 University of Michigan, Ann Arbor, Mich, USA 4 Brigham and Women’s Hospital, Boston, Mass, USA 5 St. Michael’s Hospital, University of Toronto, Canada 6 S. Orsola-Malpighi University Hospital, Bologna, Italy 7 University Medical Center Utrecht, the Netherlands 8 Robert Bosch Hospital, Stuttgart, Germany 9 University of Pennsylvania, School of Medicine, Philadelphia, Pa, USA 10 University of Calgary, Calgary, Alberta, Canada 11 University of Rostock, Rostock, Germany Chapter 8

R1 AbSTRACT R2 R3 Introduction R4 A considerable number of patients with acute type B aortic dissection (ABAD) treated with R5 medical management alone will develop aortic enlargement during follow-up (FU), which could R6 potentially lead to aortic aneurysm and rupture. The purpose of this study was to investigate R7 predictors of aortic expansion among ABAD patients enrolled in the International Registry of R8 Acute Aortic Dissection (IRAD). R9 R10 Methods R11 We analyzed 191 ABAD patients treated with medical therapy alone enrolled in the IRAD R12 between 1996 and 2010, with available descending aortic diameter measurements at admission, R13 and at 6, 12 and/or 24 months after the acute event. The annual aortic expansion rate was R14 calculated for all patients and multivariate regression analysis was used to investigate factors R15 affecting the expansion rate. R16 R17 Results R18 Aortic expansion was observed in 59% of patients (mean expansion rate 1.7 ±7 mm/year); no R19 ruptures or deaths occurred during the FU period. In multivariate analysis, white race (regression R20 coefficient [RC] 4.6, 95%CI 1.4 to 7.7) and an initial aortic diameter <4.0cm (RC 6.3, 95%CI R21 4.0 to 8.6) were associated with increased aortic expansion. Female gender (RC -3.8, 95%CI R22 -6.1 to -1.4), intramural hematoma (RC -3.8, 95%CI -6.5 to -1.1), and use of calcium channel R23 blockers (RC -3.8, 95%CI -6.2 to -1.3) were associated with decreased aortic expansion. The R24 mean expansion rate for patients with calcium channel blockers was 0.5 mm/year compared to R25 3.9 mm/year for those without (p=0.005). R26 R27 Conclusions R28 A small aortic diameter at presentation of ABAD is associated with increased aortic expansion R29 during FU, while female gender and IMH were associated with less aortic expansion. We R30 observed decreased aortic expansion among patients treated with calcium channel blockers, these R31 data raise the possibility that the use of calcium channel blockers following ABAD may reduce R32 the rate of aortic expansion and therefore further investigation is warranted. R33 R34 R35 R36 R37 R38 R39

100 Aortic Expansion after Uncomplicated Acute Type B Aortic Dissection

INTRODUCTION R1 R2 The majority of patients with acute type B aortic dissection (ABAD) are treated with medical R3 therapy alone. Surgical or endovascular treatment is usually reserved only for patients with R4 complications at presentation such as rupture, enlarging aneurysm, retrograde dissection R5 or malperfusion syndromes.1-5 In-hospital outcomes are generally acceptable in patients with R6 uncomplicated ABAD, with up to 90% of patients surviving discharge after receiving effective R7 antihypertensive therapy.5-8 However, 1 in 4 ABAD patients discharged from the hospital alive R8 will expire within three years.9, 10 Probably the most important sequel during follow-up after R9 ABAD treated with medical therapy alone is aortic enlargement, which could lead to aneurysm R10 formation and rupture.6, 9, 11, 12 Currently it is unclear which patients are more likely to experience R11 aortic enlargement after medical therapy. Better understanding of the factors that affect the aortic R12 expansion rate after ABAD could help in selecting patients that require close radiologic follow- R13 up, and perhaps may lead to future therapies to slow aortic expansion. This may theoretically R14 reduce the number of patients developing aortic rupture, improving the long-term prognosis of R15 ABAD. The purpose of this study is to investigate predictors of aortic expansion during follow-up R16 among ABAD patients enrolled in the International Registry of Acute Aortic Dissection (IRAD). R17 R18 R19 METHODS R20 8 R21 Patient selection R22 IRAD is an ongoing multi-national multi-center registry that enrolls patients with acute aortic R23 dissection at 28 large referral centers (Appendix 1), the inception and structure of IRAD has R24 been described previously.13 All patients presenting ABAD treated with medical therapy alone R25 enrolled in the IRAD between 1996 and October 2010 were identified. ABAD patients that R26 expired during hospitalization were excluded. Subsequently, only those patients with available R27 descending aortic diameter measurements at presentation, and during follow-up at 6, 12 and/or R28 24 months after the acute event, were included for analysis. Patients were excluded if the follow- R29 up aortic measurements were performed after 24 months, to warrant a relatively homogenous R30 study cohort with comparable follow-up intervals. Patients that had undergone descending aortic R31 intervention prior to the follow-up measurements were excluded as well. R32 R33 Data collection R34 In IRAD, patient and procedural data are collected using forms with over 290 variables, which R35 are submitted to the IRAD coordinating center at the University of Michigan, and checked R36 for face validity and analytical internal validity. The initial and follow-up descending aortic R37 diameters are measured by computed tomography (CTA) and/or magnetic resonance imaging R38 R39

101 Chapter 8

R1 (MRI). If patients underwent multiple imaging studies, the study that reflected the largest aortic R2 diameter was selected for analysis. The maximum aortic diameter was measured at cross-sectional R3 images perpendicular to the long axis of the descending aorta. In IRAD, all initial aortic diameter R4 measurements were obtained after aortic dissection had occurred. R5 R6 Statistical analysis R7 Data analysis was performed by a statistician (D.M.) with the use of SPSS statistical analysis R8 software (SPSS Inc., Chicago, Ill.). Summary statistics are presented as frequencies and percentages R9 for categorical variables and mean ±SD for continuous variables. The annual aortic expansion R10 rate was calculated for all patients as follows: (maximum diameter at follow-up – maximum R11 initial diameter) / length of follow-up in years. If multiple aortic measurements were performed R12 during follow-up, the most recent follow-up aortic measurements were used for calculating the R13 annual aortic expansion rate. Multivariate linear regression analysis was used to investigate the R14 effects of demographics, patient history, and imaging findings on the annual aortic expansion R15 rate. Age, gender, and variables with a P value < 0.2 in univariate linear regression analysis were R16 integrated in the multivariate linear regression model to calculate independent effects on the R17 aortic expansion rate. A P value < .05 was considered significant. The authors had full access to R18 and took full responsibility for the integrity of the data. All authors have read and agree to the R19 manuscript as written. R20 R21 R22 RESUlTS R23 R24 baseline characteristics R25 In total, 191 patients with acute type B aortic dissection managed with medical treatment only R26 were included for analysis. The mean age of patients was 62.1 ±13 years, 69% were males. At R27 discharge, 93% were prescribed Beta-blockers, and 65% of patients received calcium channel R28 blockers (table 1). The mean descending thoracic aortic diameter at first admission, measured R29 after the acute event, was 4.2 ±1.1cm. The initial descending aortic diameter was smaller than R30 4.0cm in 42% of patients (table 2). A patent false lumen was present in 48% of patients, R31 remaining patients had a partially or completely thrombosed false lumen (table 2). Intramural R32 hematoma (IMH) was observed in 37% of patients, of which 17% had an isolated IMH and R33 20% had combined IMH and ABAD. R34 R35 R36 R37 R38 R39

102 Aortic Expansion after Uncomplicated Acute Type B Aortic Dissection

Table 1. Demographics and Patient History R1 Variable N = 191 (%) R2 Demographics R3 Age (sd) 62.1 (± 13) R4 Female gender 59 (30.9) Race R5 White 161 (86.1) R6 African American 4 (2.1) R7 Asian 18 (9.6) R8 Hispanic 4 (2.1) R9 History Marfan’s syndrome 1 (2.6) R10 Aortic valve disease 9 (4.8) R11 Hypertension 133 (70.0) R12 Atherosclerosis 58 (30.5) R13 Diabetes 9 (4.8) Prior CABG 4 (2.2) R14 Prior Cath/Angiography 6 (4.8) R15 Medication at discharge R16 Beta-blocker 172 (92.5) R17 ACE-inhibitor 97 (53.6) Calcium channel blocker 119 (65.4) R18 R19 CABG coronary artery bypass grafting R20 Aortic expansion during follow-up 8 R21 The median length of follow-up was 2.0 years (IR 1 to 2 years). Aortic expansion during follow- R22 up was observed in 59% (n=113) of patients. Overall, the mean descending aortic diameter R23 increased from 4.2 cm to 4.5 cm (p<0.001); the mean annual expansion rate was 1.7 ±7.6 mm/ R24 year. The aortic expansion rate showed an approximately normal distribution, ranging from -22 R25 mm/year to +24 mm/year (figure 1). No aortic ruptures occurred during follow-up in the cohort. R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

103 Chapter 8

R1 Table 2. Imaging findings at initial presentation R2 N (%) R3 Descending aortic diameter 4.19 (± 1.1) R4 Descending < 3.5 39 (20.4) Descending 3.5 – 3.9 41 (21.5) R5 Descending 4.0 – 4.4 41 (21.5) R6 Descending 4.5 – 4.9 24 (12.6) R7 Descending ≥ 5.0 46 (24.1) R8 Aortic arch diameter 3.53 (± 0.77) False lumen thrombosis R9 Patent false lumen 73 (48.0) R10 Partial thrombosis 53 (34.9) R11 Complete thrombosis 26 (17.1) R12 Intramural hematoma 68 (36.8) Isolated IMH 12 (17.3) R13 Combined IMH and ABAD 36 (19.5) R14 R15 IMH intramural hematoma, ABAD acute type B aortic dissection R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 Figure 1. Distribution of annual aortic expansion rates R34 R35 Predictors of aortic expansion R36 In univariate linear regression analysis, the following variables significantly affected the aortic R37 expansion rate: female gender (RC -2.8, p=0.018), white race (RC 5.6, p<0.001), Asian race R38 (RC -5.5, p=0.003), use of calcium channel blockers (RC -3.4, p=0.005), presence of intramural R39

104 Aortic Expansion after Uncomplicated Acute Type B Aortic Dissection

hematoma (RC -3.7, p=0.003), and an initial descending aortic diameter < 4.0cm (RC 3.7 R1 p=0.001, table 3). As the initial aortic diameter increased, the annual aortic expansion rate R2 decreased (figure 2). Overall, the expansion rate for patients with an initial aortic diameter < 4.0 R3 cm was 3.8 mm, as compared to 0.2 mm/year for aortas ≥ 4.0cm (p=0.001). Patients with isolated R4 aortic dissection exhibited on average 3.1 mm aortic expansion per year, as compared with 0.6 R5 mm/year for isolated IMH, and -1.8mm/year for combined IMH and ABAD (p=0.002). The R6 mean annual expansion rate for patients prescribed calcium channel blockers was 0.5 mm/year R7 compared to 3.9 mm/year for those without (p=0.005). R8 R9 Table 3. Univariate effects on the aortic expansion rate (mm/year) R10 Variable RC 95% CI 95% CI Rate (no) Rate (yes) P Value R11 Low high mm/year mm/year R12 Demographics R13 Age (sd) -0.05 -0.13 0.04 ------.271 Female gender -2.82 -5.14 -0.50 2.57 -0.25 .018 R14 Race R15 White 5.64 2.64 8.64 -3.06 2.58 <0.001 R16 African American -6.18 -13.56 1.21 1.93 -4.25 .101 R17 Asian -5.49 -9.05 -1.93 2.32 -3.17 .003 R18 Hispanic -3.24 -10.66 4.18 1.87 -1.38 .390 History R19 Marfan -0.22 -7.06 6.61 1.72 1.50 .949 R20 Aortic valve disease 2.20 -2.96 7.35 1.64 3.83 .401 R21 Hypertension -0.11 -2.50 2.28 1.80 1.68 .927 8 R22 Atherosclerosis 0.39 -1.98 2.77 1.60 1.99 .744 Diabetes -2.01 -7.18 3.16 1.79 -0.22 .443 R23 Prior CABG 4.16 -3.57 11.88 1.72 5.88 .290 R24 Prior Cath/Angiography 2.26 -4.60 9.11 2.41 4.67 .516 R25 Medication at discharge R26 Beta-blocker 3.39 -0.81 7.58 -1.43 1.96 .113 ACE-inhibitor -0.17 -2.44 2.10 1.66 1.50 .883 R27 Calcium channel blocker -3.39 -5.72 -1.06 3.90 0.51 .005 R28 Initial aortic diameter < 4.0cm 3.67 1.53 5.82 0.16 3.83 .001 R29 False lumen thrombosis R30 Patent false lumen 2.38 -0.13 4.89 0.93 3.31 .063 R31 Partial thrombosis -1.47 -4.12 1.18 2.59 1.11 .274 Complete thrombosis -1.83 -5.18 1.53 2.39 0.56 .283 R32 Intramural hematoma* -3.70 -6.06 -1.33 3.11 -0.58 .002 R33 *Intramural hematoma (IMH) included both patients with an isolated IMH as combined ABAD and IMH R34 R35 R36 R37 R38 R39

105 Chapter 8

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 Figure 2. Initial descending aortic diameter and mean annual expansion rate R14 R15 In multivariate regression analysis, white race (RC 4.6, 95%CI 1.4 to 7.7, p=0.005) and an R16 initial aortic diameter <4.0cm (RC 6.3, 95%CI 4.0 to 8.6, p<0.001) were associated with R17 increased aortic expansion. The following variables were independently associated with decreased R18 aortic expansion during follow-up: female gender (RC -3.8, 95%CI -6.1 to -1.4), intramural R19 hematoma (RC -3.8, 95%CI -6.5 to -1.1), and calcium channel blockers (RC -3.8, 95%CI -6.2 R20 to -1.3) (table 4). Patients treated with calcium channel blocker therapy tended to have improved R21 survival during follow-up (p=0.115, figure 3). R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 Figure 3. Survival of patients treated with and without calcium channel blocker therapy R39

106 Aortic Expansion after Uncomplicated Acute Type B Aortic Dissection

Table 4. Independent predictors of the aortic expansion rate (mm/year) R1 Variable Regression Standardized 95% CI 95% CI P Value R2 Coefficient Coefficient low high R3 Age (years) 0.07 0.11 -0.02 0.15 0.134 R4 Female gender -3.77 -0.23 -6.13 -1.42 0.002 White race 4.58 0.21 1.41 7.75 0.005 R5 Initial aortic diameter < 4.0cm 6.30 0.40 4.03 8.58 <0.001 R6 Patent false lumen 0.53 0.05 -1.47 3.17 0.468 R7 Intramural hematoma* -3.85 -0.21 -6.54 -1.15 0.005 R8 Beta-blocker therapy 7.47 0.22 -2.71 12.23 0.057 R9 Calcium-channel blocker therapy -3.77 -0.22 -6.20 -1.33 0.003 R10 *Intramural hematoma (IMH) included both patients with an isolated IMH as combined ABAD and IMH R11 R12 DISCUSSION R13 R14 Although early results of uncomplicated ABAD treated with medical management alone are R15 encouraging, up to 60% of such patients in IRAD exhibited aortic expansion during follow-up. R16 The overall expansion rate was around 1.7 mm/year in IRAD, similar to previous analyses which R17 found expansion rates around 1.3 to 2.6 mm/year at the level of the descending thoracic aorta.11, R18 12, 14 R19 The annual expansion rate after ABAD varies considerably among patients, so identifying R20 patients at risk for increased aortic expansion appears crucial. Our analysis revealed multiple 8 R21 patient characteristics that may influence the aortic expansion rate, including demographic, R22 morphologic and therapeutic factors. The expansion rate was increased among male patients as R23 compared to females, which may be explained by anatomic differences between both genders. R24 Women are typically smaller in size than men, and female aortic measurements may be reduced R25 as well, with smaller aortic expansion rates. Racial differences in aortic expansion were observed R26 as well. Caucasians exhibited on average an increased expansion rate, while aortic enlargement R27 occurred less frequently among Asian patients. In abdominal aortic aneurysm (AAA), racial R28 disparities appear to exist as well, since the prevalence of AAA is considerably lower in the Asian R29 population, as compared to the Caucasian population.15, 16 R30 In addition, morphologic characteristics of the aortic pathology affected the aortic enlargement R31 during follow-up. Presence of intramural hematoma was associated with a lower expansion R32 rate as compared to isolated aortic dissections. In intramural hematoma, there is typically no R33 intimal tear which may be beneficial for the hemodynamic forces that are normally experienced R34 in the false lumen. In patients with classic type B dissection, a patent false lumen may increase R35 the aortic expansion rate, as demonstrated previously.11, 14, 17 In contrast, in a thrombosed false R36 lumen, the aorta frequently becomes as a single lumen channel, with dissected layers which R37 become a single layer, increasing the vessel wall strength.11 A previous IRAD study found an R38 R39

107 Chapter 8

R1 increased long-term mortality rate among patient with a partially thrombosed false lumen, R2 possibly because of occlusion of distal reentry tears of the false lumen by a partial thrombus.18 R3 This could theoretically result in increased pressure in the false lumen, and elevated expansion R4 and rupture risks. We could not confirm this hypothesis in the present analysis. R5 Surprisingly, an aortic diameter > 4.0cm at first admission was associated with decreased aortic R6 expansion, while smaller aortic dissections tended to expand faster during follow-up. Sueyoshi R7 et al previously found a higher aortic expansion rate in patients with relatively smaller aortic R8 diameters as well, although this failed to reach statistical significance in their analysis.11 In aortic R9 aneurysm, in which the vessel wall is still intact, aortic expansion increases exponentially with R10 increasing diameter, according to Laplace’s law.19, 20 An explanation for this diameter paradox R11 remains unclear at this moment. The risk of developing aortic dissection generally is thought to R12 increase as the aorta enlarges,21 although the majority of type A and B aortic dissection still occur R13 in aortic diameters inferior to the threshold for elective intervention (5.5cm).22, 23 Patients with R14 relatively normal aortic measurements developing aortic dissection may have more severe aortic R15 wall abnormalities, possibly caused by connective tissue disorders, as compared to those suffering R16 aortic dissection in enlarged aortas. The poor condition of the aortic wall in these patients with R17 small aortic diameters may theoretically also result in increased risks of aortic enlargement during R18 follow-up after dissection. R19 Most patients suffering uncomplicated ABAD are managed with medical management alone. In R20 IRAD, 93% of ABAD patients were prescribed Beta-blockers, and about 65% of patients received R21 calcium channel blocker therapy. We observed that those patients treated with calcium channel R22 blockers exhibited a significantly lower aortic expansion rate during follow-up after ABAD R23 compared to those without calcium channel blockers (0.5mm/year vs. 3.9mm/year), even after R24 adjusting for other risk factors. Calcium antagonists generally block the slow calcium channels, R25 decreasing the influx of extracellular calcium, which results in suppression of the sinoatrial and R26 atrioventricular nodes, vascular dilatation with decreased blood pressure, and reduced cardiac R27 contractility. Adequate blood pressure regulation and subsequent decreased pressure on the aortic R28 wall through calcium channel blockade may theoretically decrease risks of aortic enlargement. R29 Decreased aortic expansion observed among patients treated with calcium channel blockers might R30 lead to a lower risk of aneurysm formation and rupture. A recent IRAD investigation showed R31 that use of calcium channel blockers was independently associated with improved survival during R32 follow-up among the IRAD cohort of ABAD patients.24 The present analysis with a smaller R33 cohort of ABAD patients with available growth rates showed a similar trend towards decreased R34 follow-up mortality among patients treated with calcium channel blockers. Unfortunately, exact R35 causes of death during follow-up were not available for all patients, so it remains unclear if the R36 survival benefit is related to decreased aortic expansion among patients treated with calcium R37 channel blockers. R38 R39

108 Aortic Expansion after Uncomplicated Acute Type B Aortic Dissection

While many predictive factors of aortic expansion found in the current analysis, like demographic R1 and morphologic characteristics of ABAD patients, cannot be altered by physicians, medical R2 therapy clearly could be prescribed after ABAD. However, the exact role of calcium channel R3 blockers in aortic dissection remains undefined at this moment and caution is needed when R4 interpreting the current observations. Additional research, preferably a randomized controlled R5 trial, may be needed to assess the exact benefits and potential adverse effects of calcium channel R6 blockers after ABAD, before any recommendations can be made. R7 The IRAD database contains the largest series of patients with acute type B aortic dissection to R8 date, which provided a unique opportunity to investigate predictors of aortic expansion after R9 aortic dissection. However, as with all observational studies, this investigation has limitations that R10 must be kept in mind when the data are interpreted. Descending aortic diameter measurements R11 during follow-up were sometimes not available, so the aortic expansion rate could not be R12 calculated for all patients in IRAD. Unfortunately, data related to the specific calcium channel R13 blockers prescribed, medication dosages or patient adherence were not recorded in IRAD, so R14 we could not stratify for these variables. In addition, the mortality data available to us did not R15 include information on the cause of death for all patients, so potential improvements in survival R16 among patients that received calcium channel blockers remain unexplained. R17 R18 R19 CONClUSION R20 8 R21 A small aortic diameter at presentation of ABAD is associated with increased aortic expansion R22 during FU, while female gender and IMH were associated with less aortic expansion. We observed R23 decreased aortic expansion among patients treated with calcium channel blockers. These data R24 raise the possibility that the use of calcium channel blockers following ABAD may reduce the R25 rate of aortic expansion and therefore further investigation is warranted. R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

109 Chapter 8

R1 REfERENCES R2 R3 (1) Elefteriades JA, Hartleroad J, Gusberg RJ, Salazar AM, Black HR, Kopf GS, Baldwin JC, R4 Hammond GL. Long-term experience with descending aortic dissection: the complication- R5 specific approach. Ann Thorac Surg 1992 January;53(1):11-20. R6 (2) Golledge J, Eagle KA. Acute aortic dissection. Lancet 2008 July 5;372(9632):55-66. R7 (3) Masuda Y, Yamada Z, Morooka N, Watanabe S, Inagaki Y. Prognosis of patients with medically R8 treated aortic dissections. Circulation 1991 November;84(5 Suppl):III7-13. R9 (4) Glower DD, Fann JI, Speier RH, Morrison L, White WD, Smith LR, Rankin JS, Miller DC, R10 Wolfe WG. Comparison of medical and surgical therapy for uncomplicated descending aortic R11 dissection. Circulation 1990 November;82(5 Suppl):IV39-IV46. R12 (5) Hata M, Shiono M, Inoue T, Sezai A, Niino T, Negishi N, Sezai Y. Optimal treatment of R13 type B acute aortic dissection: long-term medical follow-up results. Ann Thorac Surg 2003 R14 June;75(6):1781-4. R15 (6) Elefteriades JA, Lovoulos CJ, Coady MA, Tellides G, Kopf GS, Rizzo JA. Management of R16 descending aortic dissection. Ann Thorac Surg 1999 June;67(6):2002-5. R17 (7) Suzuki T, Mehta RH, Ince H, Nagai R, Sakomura Y, Weber F, Sumiyoshi T, Bossone E, Trimarchi R18 S, Cooper JV, Smith DE, Isselbacher EM, Eagle KA, Nienaber CA. Clinical profiles and outcomes R19 of acute type B aortic dissection in the current era: lessons from the International Registry of R20 Aortic Dissection (IRAD). Circulation 2003 September 9;108 Suppl 1:II312-II317. R21 (8) Trimarchi S, Eagle KA, Nienaber CA, Pyeritz RE, Jonker FH, Suzuki T, O’Gara PT, Hutchinson R22 SJ, Rampoldi V, Grassi V, Bossone E, Muhs BE, Evangelista A, Tsai TT, Froehlich JB, Cooper R23 JV, Montgomery D, Meinhardt G, Myrmel T, Upchurch GR, Sundt TM, Isselbacher EM. R24 Importance of refractory pain and hypertension in acute type B aortic dissection: insights from R25 the International Registry of Acute Aortic Dissection (IRAD). Circulation 2010 September R26 28;122(13):1283-9. R27 (9) Tsai TT, Fattori R, Trimarchi S, Isselbacher E, Myrmel T, Evangelista A, Hutchison S, Sechtem U, R28 Cooper JV, Smith DE, Pape L, Froehlich J, Raghupathy A, Januzzi JL, Eagle KA, Nienaber CA. R29 Long-term survival in patients presenting with type B acute aortic dissection: insights from the R30 International Registry of Acute Aortic Dissection. Circulation 2006 November 21;114(21):2226- R31 31. R32 (10) Tsai TT, Evangelista A, Nienaber CA, Trimarchi S, Sechtem U, Fattori R, Myrmel T, Pape L, R33 Cooper JV, Smith DE, Fang J, Isselbacher E, Eagle KA. Long-term survival in patients presenting R34 with type A acute aortic dissection: insights from the International Registry of Acute Aortic R35 Dissection (IRAD). Circulation 2006 July 4;114(1 Suppl):I350-I356. R36 (11) Sueyoshi E, Sakamoto I, Hayashi K, Yamaguchi T, Imada T. Growth rate of aortic diameter in R37 patients with type B aortic dissection during the chronic phase. Circulation 2004 September R38 14;110(11 Suppl 1):II256-II261. R39

110 Aortic Expansion after Uncomplicated Acute Type B Aortic Dissection

(12) Song JM, Kim SD, Kim JH, Kim MJ, Kang DH, Seo JB, Lim TH, Lee JW, Song MG, Song JK. R1 Long-term predictors of descending aorta aneurysmal change in patients with aortic dissection. J R2 Am Coll Cardiol 2007 August 21;50(8):799-804. R3 (13) Hagan PG, Nienaber CA, Isselbacher EM, Bruckman D, Karavite DJ, Russman PL, Evangelista R4 A, Fattori R, Suzuki T, Oh JK, Moore AG, Malouf JF, Pape LA, Gaca C, Sechtem U, Lenferink R5 S, Deutsch HJ, Diedrichs H, Robles J, Llovet A, Gilon D, Das SK, Armstrong WF, Deeb GM, R6 Eagle KA. The International Registry of Acute Aortic Dissection (IRAD): new insights into an R7 old disease. JAMA 2000 February 16;283(7):897-903. R8 (14) Kimura N, Tanaka M, Kawahito K, Yamaguchi A, Ino T, Adachi H. Influence of patent false R9 lumen on long-term outcome after surgery for acute type A aortic dissection. J Thorac Cardiovasc R10 Surg 2008 November;136(5):1160-6, 1166. R11 (15) Spark JI, Baker JL, Vowden P, Wilkinson D. Epidemiology of abdominal aortic aneurysms in the R12 Asian community. Br J Surg 2001 March;88(3):382-4. R13 (16) Salem MK, Rayt HS, Hussey G, Rafelt S, Nelson CP, Sayers RD, Naylor AR, Nasim A. Should R14 Asian men be included in abdominal aortic aneurysm screening programmes? Eur J Vasc Endovasc R15 Surg 2009 December;38(6):748-9. R16 (17) Sueyoshi E, Sakamoto I, Uetani M. Growth rate of affected aorta in patients with type B partially R17 closed aortic dissection. Ann Thorac Surg 2009 October;88(4):1251-7. R18 (18) Tsai TT, Evangelista A, Nienaber CA, Myrmel T, Meinhardt G, Cooper JV, Smith DE, Suzuki R19 T, Fattori R, Llovet A, Froehlich J, Hutchison S, Distante A, Sundt T, Beckman J, Januzzi JL, Jr., R20 Isselbacher EM, Eagle KA. Partial thrombosis of the false lumen in patients with acute type B 8 R21 aortic dissection. N Engl J Med 2007 July 26;357(4):349-59. R22 (19) Powell JT, Sweeting MJ, Brown LC, Gotensparre SM, Fowkes FG, Thompson SG. Systematic R23 review and meta-analysis of growth rates of small abdominal aortic aneurysms. Br J Surg 2011 R24 May;98(5):609-18. R25 (20) Schlosser FJ, Tangelder MJ, Verhagen HJ, van der Heijden GJ, Muhs BE, van der GY, Moll R26 FL. Growth predictors and prognosis of small abdominal aortic aneurysms. J Vasc Surg 2008 R27 June;47(6):1127-33. R28 (21) Elefteriades JA. Natural history of thoracic aortic aneurysms: indications for surgery, and surgical R29 versus nonsurgical risks. Ann Thorac Surg 2002 November;74(5):S1877-S1880. R30 (22) Pape LA, Tsai TT, Isselbacher EM, Oh JK, O’Gara PT, Evangelista A, Fattori R, Meinhardt G, R31 Trimarchi S, Bossone E, Suzuki T, Cooper JV, Froehlich JB, Nienaber CA, Eagle KA. Aortic R32 diameter >or = 5.5 cm is not a good predictor of type A aortic dissection: observations from R33 the International Registry of Acute Aortic Dissection (IRAD). Circulation 2007 September R34 4;116(10):1120-7. R35 R36 R37 R38 R39

111 Chapter 8

R1 (23) Trimarchi S, Jonker FH, Hutchison S, Isselbacher EM, Pape LA, Patel HJ, Froehlich JB, Muhs R2 BE, Rampoldi V, Grassi V, Evangelista A, Meinhardt G, Beckman J, Myrmel T, Pyeritz RE, R3 Hirsch AT, Sundt Iii TM, Nienaber CA, Eagle KA. Descending aortic diameter of 5.5 cm or R4 greater is not an accurate predictor of acute type B aortic dissection. J Thorac Cardiovasc Surg R5 2011 May 16. R6 (24) Suzuki T, Isselbacher EM, Nienaber CA, Pyeritz RE, Eagle KA, Tsai TT, Cooper JV, Januzzi JL, R7 Jr., Braverman AC, Montgomery DG, Fattori R, Pape L, Harris KM, Booher A, Oh JK, Peterson R8 M, Ramanath VS, Froehlich JB. Type-Selective Benefits of Medications in Treatment of Acute R9 Aortic Dissection (from the International Registry of Acute Aortic Dissection [IRAD]). Am J R10 Cardiol 2011 September 23. R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

112 Chapter 9

Long-Term Outcomes Of Surgical Aortic Fenestration for Complicated Acute Type B Aortic Dissections

Journal of Vascular Surgery 2010 Aug;52(2):261-6.

Santi Trimarchi, MD,1 Frederik H.W. Jonker, MD, 2 Bart E Muhs, MD, PhD, 2 Viviana Grassi, MD,1 Paolo Righini, MD,1Gilbert R. Upchurch, MD, 3 Vincenzo Rampoldi, MD, 1

1 Policlinico San Donato IRCCS, San Donato Milanese, Italy 2 School of Medicine,Yale University, New Haven, Conn, USA 3 University of Michigan, Ann Arbor, Mich, USA Chapter 9

R1 AbSTRACT R2 R3 Introduction R4 Surgical aortic fenestration can be used for treating ischemic complications of acute type B aortic R5 dissection (ABAD). The purpose of this study is to describe our surgical suprarenal and infrarenal R6 aortic fenestration technique, and to report the long-term outcomes of this approach in the R7 management of complicated ABAD. R8 R9 Methods R10 We retrospectively analyzed the in-hospital and long-term outcomes of 18 patients treated R11 with either suprarenal (group 1, n=10) or infrarenal surgical fenestration (group 2, n=8) for R12 complicated ABAD between 1988 and 2002. Median age was 60 years (interquartile range 14; R13 range 48 to 82 years) and 89% (n=16) were males. Patients in group 1 were approached through R14 a thoraco-abdominal incision in the 10th intercostal space, while patients in group 2 underwent R15 a midline laparotomy. A longitudinal aortotomy was performed and the true and false lumens R16 were identified, followed by widely resection of the intimal membrane. R17 R18 Results R19 The in-hospital mortality was 22% (n=4), which included 2 deaths in group 1 (20%) and 2 deaths R20 in group 2 (25%). In the remaining patients full visceral, renal and lower extremity function was R21 recovered. Median follow-up of surviving patients was 10.0 years (interquartile range 12.5; range R22 0.5 to 20 years). During follow-up, none of the patients developed complications related to the R23 initial surgical fenestration procedure and no significant dilatations of the treated aortic segments R24 were noted. Three out of 18 ABAD patients expired during the follow-up period due to causes R25 non-related to the aortic dissection or surgical procedure. R26 R27 Conclusions R28 Surgical aortic fenestration represents an effective and durable option for treating ischemic R29 complications of ABAD. Actually, this conservative surgical technique may serve as the alternative R30 treatment in case of contra-indications or failure of endovascular management of complicated R31 ABAD. R32 R33 R34 R35 R36 R37 R38 R39

114 Surgical aortic fenestration for complicated acute type B aortic dissections

INTRODUCTION R1 R2 Acute type B aortic dissection (ABAD) is a serious cardiovascular emergency in which morbidity R3 and mortality is often related to the presence of complications at clinical presentation.1-3 Visceral, R4 renal and limb ischemia occur in up to 30% of ABAD patients, and these are associated with R5 higher in-hospital mortality.4-9 Surgical or endovascular intervention for ABAD are indicated in R6 case of malperfusion syndromes, extending dissection and/or aortic rupture. R7 Surgical aortic fenestration was the first operation for the management of complicated aortic R8 dissection, and was performed for the first time by Gurin et al in 1935.10 The technique is R9 based on the concept to create a single lumen which resolves the malperfusion, limiting risks R10 of intestinal infarction, acute renal failure and limb ischemia.11-13 Although surgical aortic R11 fenestration has been used for treating complicated ABAD for many decades, limited experience R12 has been reported and the long-term outcomes are unknown. R13 In the last decade, percutaneous approaches, such as endovascular fenestration or stenting, have R14 gained popularity for the acute management of complicated ABAD. The initial results of these R15 less invasive approaches have been encouraging, however concerns remain regarding the required R16 major expertise and the long-term outcomes.14-16 R17 The purpose of the present study is to describe our surgical suprarenal and infrarenal aortic R18 fenestration techniques, and to report the long-term outcomes of this approach for the R19 management of complicated ABAD. R20 R21 R22 METHODS R23 9 R24 Patient selection R25 Overall, 60 patients with ABAD were admitted at the Cardiovascular Center “E. Malan” of the R26 Policlinico San Donato IRCCS (, Italy) between 1988 and 2002, of which 65% (n=39) R27 were managed medically and 35% (n=21) required surgical intervention due to complicated R28 ABAD. Surgery consisted of open surgical aortic fenestration in 86% (n=18), and the remaining R29 three patients were treated with femoro-femoral bypass. After 2002 no further patients with R30 complicated ABAD underwent surgical aortic fenestration, except for one patient with an R31 iatrogenic dissection associated with acute renal and limb ischemia. This last patient was not R32 included in the present evaluation because of her relatively short follow-up (6 months). Since R33 2002 endovascular management became our first choice of treatment for complicated ABAD. R34 The median age of the 18 patients treated with surgical aortic fenestration was 60 years (interquartile R35 range 14 years; range 48 to 82 years) and 89% (n=16) were males. The majority of these patients R36 (n=15, 83%) required immediate surgical intervention because of ischemic complications, R37 including ischemia of the lower extremities (n=9), of which one showed preoperative paraplegia, R38 R39

115 Chapter 9

R1 renal ischemia (n=7), and visceral ischemia (n=4). Four of these patients had multiple ischemic R2 complications. Three ABAD patients had different indications for surgical aortic fenestration and R3 were treated within 48 hours. Indications for intervention included iatrogenic ABAD, refractory R4 hypertension, and refractory pain, of which the last two had a concomitant abdominal aortic R5 aneurysm and a hepatic artery aneurysm. R6 R7 Surgical technique R8 The treated aortic segment was normal in size in all 18 patients with ABAD at surgery. The R9 surgical aortic fenestration was performed at the suprarenal level in 10 cases (group 1) and at R10 the infrarenal level in 8 cases (group 2). Patients in group 1 were approached through a thoraco- R11 abdominal incision in the 10th intercostal space. Than the thoraco-abdominal aorta was isolated, R12 a supra-diaphragmatic clamp was placed without using any extra-corporeal support. Distally the R13 aorta was clamped at the level of its bifurcation. A longitudinal aortotomy was performed, and R14 the true and false lumens were identified. The intimal membrane was widely resected, leaving R15 behind the adherent intima. The dissected flap was proximally and distally resected in a triangular R16 shape, in order to maximize the single aortic lumen (figure 1 A). If the dissection extended into R17 the visceral and/or renal arteries, a similar technique was adopted. The longitudinal aortotomy R18 was then sutured and reinforced with strips of Teflon, which resulted in a slightly reduced aortic R19 diameter (figure 1 B). R20 In group 2, the patients were approached through a median laparotomy. The aorta was clamped R21 at the suprarenal level as well as at the iliac arteries, and then opened longitudinally. After R22 identification of the true and false lumen, the dissected flap was resected longitudinally for 2-3 R23 cm, from above to just below the level of the renal arteries. Similar to group 1, the intimal flap R24 was resected proximally in a triangular shape, in order to maximize the single aortic lumen. R25 The distal extent of the intimal flap was fixed to adventitia, and the aortic wall was reinforced R26 both internally and externally with strips of Teflon. Aortic replacement with a Dacron graft R27 was performed in those cases with complete infrarenal aortic occlusion and/or abdominal aortic R28 aneurysm (figures 2 A,B,C,D). R29 The in-hospital course was followed on the basis of clinical signs and ultrasound. Additional CT R30 imaging was only used in case of suspected complications. Long-term follow-up was conducted R31 annually using ultrasound and CT imaging. R32 R33 R34 R35 R36 R37 R38 R39

116 Surgical aortic fenestration for complicated acute type B aortic dissections

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 Figure 1 A,B. Surgical technique of suprarenal aortic fenestration R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 9 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 Figure 2 A, B, C, D. Surgical technique of infrarenal aortic fenestration R38 R39

117 Chapter 9

R1 RESUlTS R2 R3 In-hospital outcomes R4 The overall in-hospital mortality was 22% (n=4), which included 2 deaths in group 1 (20%) R5 due to myocardial infarction and multi-organ failure, and 2 deaths in group 2 (25%) due to R6 myocardial infarction and an aortoduodenal fistula with hemorrhagic shock. Three out of four R7 deaths occurred within the first week after surgery. In the remaining patients full visceral, renal R8 and lower extremity function was recovered. R9 R10 follow-up outcomes R11 Median follow-up of surviving patients was 10.0 years (interquartile range 12.5; range 0.5 to R12 20 years). During follow-up, none of the patients developed complications related to the initial R13 surgical fenestration procedure. Three out of 18 ABAD patients expired during the follow-up R14 period, due to causes non-related to the aortic dissection or surgical procedure (figure 3). Causes R15 of death were stroke, acute myocardial infarction, and multi-organ failure (MOF) after aortic R16 replacement for a descending thoracic aneurysm in an aortic segment above the fenestrated one, R17 not previously treated. R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 Figure 3. Survival after surgical aortic fenestration for complicated ABAD R34 R35 The surgically managed aortic segments were followed using ultrasound and CTA, and no R36 significant dilatations of the aortic segments were noted (figure 4). The median diameter of the R37 treated aortic tract at the latest follow-up moment was 38 mm (interquartile range 7; range 29 R38 to 43 mm). Six new aortic aneurysms were diagnosed in five ABAD patients at distant locations R39 from the aortic segments that were initially treated. Five of these aneurysms were located at the

118 Surgical aortic fenestration for complicated acute type B aortic dissections

descending thoracic aorta (figure 5), and one at the distal aortic arch, three of these aneurysms R1 underwent open surgical repair. R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 Figure 4. Computed tomography angiography (CTA) showing no aortic dilatation 6 years after suprarenal R18 fenestration (patient 18). R19 R20 R21 R22 R23 9 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 Figure 5. Computed tomography angiography (CTA) image of a patient with a contained rupture (white R37 arrow) of a descending thoracic aortic aneurysm 15 years after the initial fenestration procedure (patient R38 11). R39

119 Chapter 9

R1 DISCUSSION R2 R3 Conservative medical treatment is generally accepted as an adequate therapy in stable patients R4 with ABAD without evidence of complications.1-3, 17, 18 Immediate intervention is required if R5 ABAD is complicated by acute aortic enlargement, extension of the dissection, or visceral, renal R6 or limb ischemia.4-9 Surgical aortic fenestration was first performed in 1935 by Gurin et al,10 R7 and this technique can be used for the management of acute malperfusions caused by ABAD in R8 normal sized aortas.3, 19-21 During fenestration, a suprarenal or infrarenal single lumen is created R9 which resolves the malperfusion and prevents further visceral, renal or limb ischemia. R10 Surgical aortic fenestration is less invasive than total aortic replacement. Extra-corporeal support R11 is not required, and in contrast to aortic replacement, surgical fenestration allows preservation of R12 the intercostal arteries which facilitates adequate spinal cord perfusion and results in decreased R13 risks on paraplegia. However, in case of extensive dissections or acute aortic enlargement, total R14 aortic replacement is mandatory.3, 18, 22-25 Although surgical aortic fenestration has been used for R15 treating complicated ABAD for many decades, limited experience and long-term outcomes have R16 been reported. R17 In the present study, we observed that surgical aortic fenestration is an effective and durable R18 approach for managing complicated cases of ABAD. After a median follow-up of 10 years, no R19 complications related to the initial conservative surgical procedure occurred in the surviving R20 patients, and no significant aneurysmal enlargement has been diagnosed in the aortic segments R21 that were treated surgically. New aortic aneurysms occurred at distant locations from the aortic R22 segments that were initially treated with aortic fenestration. R23 More recently, minimal invasive techniques like endovascular aortic fenestration and endovascular R24 stenting have gained enormous popularity for the treatment of complicated ABAD. The initial R25 results of endovascular management of complicated ABAD have been promising, demonstrating R26 30-day mortality rates around 10% to 20%.14-16, 26-28 Endovascular management is therefore R27 becoming the first line treatment for complicated ABAD. However, a suitable vascular anatomy, R28 adequate operator experience, and improvement of current stent-grafts may be required for R29 acceptable results.15, 29 Additionally, continued enlargement of the true and false lumen after R30 successful endovascular treatment is frequently seen14, 16 and long-term results of endovascular R31 management of complicated ABAD remain unknown. R32 To our knowledge, the present study represents the largest series with long-term follow-up R33 outcomes of patients treated with surgical aortic fenestration for complicated ABAD. Although R34 the number of patients of the present study is still limited, and the operative mortality may R35 be slightly increased compared with results after endovascular management, the presented R36 conservative surgical approach appears a valuable and durable alternative for the management of R37 ABAD in the present endovascular era. At our institutions, endovascular management has become R38 the procedure of choice in case of complicated ABAD. However, surgical aortic fenestration R39

120 Surgical aortic fenestration for complicated acute type B aortic dissections

represents the alternative treatment for complicated ABAD in case of contra-indications or R1 failure of endovascular management. R2 R3 R4 CONClUSION R5 R6 Surgical aortic fenestration represents an effective and durable option for treating ischemic R7 complications of acute type B aortic dissections. In this series, no ABAD-related complications R8 or significant aneurysmal enlargement occurred during long-term follow-up after suprarenal or R9 infrarenal aortic fenestration. Actually, this conservative surgical technique may be used as the R10 alternative treatment in case of contra-indications or failure of endovascular management of R11 complicated ABAD. R12 R13 R14 ACkNOwlEDGEMENTS R15 R16 The authors thank and dedicate this article to Ugo Ruberti, MD, Professor of Surgery and Past- R17 Director of the Cardiovascular Center “Edmondo Malan” at the University of Milan, Italy, who R18 initiated the surgical fenestration experience for treating ischemic complication in acute B aortic R19 dissection in the eighties. R20 R21 R22 R23 9 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

121 Chapter 9

R1 REfERENCES R2 R3 (1) Anagnostopoulos CE, Prabhakar MJ, Kittle CF. Aortic dissections and dissecting aneurysms. Am R4 J Cardiol 1972 Aug;30(3):263-73. R5 (2) Wheat MW, Jr. Acute dissecting aneurysms of the aorta: diagnosis and treatment--1979. Am R6 Heart J 1980 Mar;99(3):373-87. R7 (3) Elefteriades JA, Lovoulos CJ, Coady MA, Tellides G, Kopf GS, Rizzo JA. Management of R8 descending aortic dissection. Ann Thorac Surg 1999 Jun;67(6):2002-5. R9 (4) Cambria RP, Brewster DC, Gertler J, Moncure AC, Gusberg R, Tilson MD, et al. Vascular R10 complications associated with spontaneous aortic dissection. J Vasc Surg 1988 Feb;7(2):199-209. R11 (5) DeBakey ME, McCollum CH, Crawford ES, Morris GC, Jr., Howell J, Noon GP, et al. R12 Dissection and dissecting aneurysms of the aorta: twenty-year follow-up of five hundred twenty- R13 seven patients treated surgically. Surgery 1982 Dec;92(6):1118-34. R14 (6) Fann JI, Sarris GE, Mitchell RS, Shumway NE, Stinson EB, Oyer PE, et al. Treatment of R15 patients with aortic dissection presenting with peripheral vascular complications. Ann Surg 1990 R16 Dec;212(6):705-13. R17 (7) Suzuki T, Mehta RH, Ince H, Nagai R, Sakomura Y, Weber F, et al. Clinical profiles and outcomes R18 of acute type B aortic dissection in the current era: lessons from the International Registry of R19 Aortic Dissection (IRAD). Circulation 2003 Sep 9;108 Suppl 1:II312-II317. R20 (8) Henke PK, Williams DM, Upchurch GR, Jr., Proctor M, Cooper JV, Fang J, et al. Acute limb R21 ischemia associated with type B aortic dissection: clinical relevance and therapy. Surgery 2006 R22 Oct;140(4):532-9. R23 (9) Slater EE, DeSanctis RW. The clinical recognition of dissecting aortic aneurysm. Am J Med 1976 R24 May 10;60(5):625-33. R25 (10) Gurin D, Bulmer JW, Derby R. Dissecting aneurysm of the aorta: diagnosis and operative relief R26 of acute arterial obstruction due to this cause. N Y State J Med 1935;35:1200-2. R27 (11) Stone CD, Greene PS, Gott VL, Frank S, Williams GM. Single-stage repair of distal aortic arch R28 and thoracoabdominal dissecting aneurysms using aortic tailoring and circulatory arrest. Ann R29 Thorac Surg 1994 Mar;57(3):580-7. R30 (12) Webb TH, Williams GM. Abdominal aortic tailoring for renal, visceral, and lower extremity R31 malperfusion resulting from acute aortic dissection. J Vasc Surg 1997 Sep;26(3):474-80. R32 (13) Matsuyama K, Narita Y, Usui A, Akita T, Oshima H, Ueda Y. Entry closure and aortic tailoring R33 for chronic type B aortic dissection. Asian Cardiovasc Thorac Ann 2008 Jun;16(3):249-51. R34 (14) Beregi JP, Haulon S, Otal P, Thony F, Bartoli JM, Crochet D, et al. Endovascular treatment of R35 acute complications associated with aortic dissection: midterm results from a multicenter study. R36 J Endovasc Ther 2003 Jun;10(3):486-93. R37 (15) Duebener LF, Lorenzen P, Richardt G, Misfeld M, Notzold A, Hartmann F, et al. Emergency R38 endovascular stent-grafting for life-threatening acute type B aortic dissections. Ann Thorac Surg R39 2004 Oct;78(4):1261-6.

122 Surgical aortic fenestration for complicated acute type B aortic dissections

(16) Park KB, Do YS, Kim SS, Kim DK, Choe YH. Endovascular treatment of acute complicated R1 aortic dissection: long-term follow-up of clinical outcomes and CT findings. J Vasc Interv Radiol R2 2009 Mar;20(3):334-41. R3 (17) Tsai TT, Trimarchi S, Nienaber CA. Acute aortic dissection: perspectives from the International R4 Registry of Acute Aortic Dissection (IRAD). Eur J Vasc Endovasc Surg 2009 Feb;37(2):149-59. R5 (18) Pradhan S, Elefteriades JA, Sumpio BE. Utility of the aortic fenestration technique in the R6 management of acute aortic dissections. Ann Thorac Cardiovasc Surg 2007 Oct;13(5):296-300. R7 (19) Panneton JM, Teh SH, Cherry KJ, Jr., Hofer JM, Gloviczki P, Andrews JC, et al. Aortic R8 fenestration for acute or chronic aortic dissection: an uncommon but effective procedure. J Vasc R9 Surg 2000 Oct;32(4):711-21. R10 (20) Hsu RB, Ho YL, Chen RJ, Wang SS, Lin FY, Chu SH. Outcome of medical and surgical R11 treatment in patients with acute type B aortic dissection. Ann Thorac Surg 2005 Mar;79(3):790- R12 4. R13 (21) Elefteriades JA, Hammond GL, Gusberg RJ, Kopf GS, Baldwin JC. Fenestration revisited. A safe R14 and effective procedure for descending aortic dissection. Arch Surg 1990 Jun;125(6):786-90. R15 (22) Estrera AL, Miller CC, III, Chen EP, Meada R, Torres RH, Porat EE, et al. Descending thoracic R16 aortic aneurysm repair: 12-year experience using distal aortic perfusion and cerebrospinal fluid R17 drainage. Ann Thorac Surg 2005 Oct;80(4):1290-6. R18 (23) Coselli JS, Bozinovski J, LeMaire SA. Open surgical repair of 2286 thoracoabdominal aortic R19 aneurysms. Ann Thorac Surg 2007 Feb;83(2):S862-S864. R20 (24) Schepens MA, Heijmen RH, Ranschaert W, Sonker U, Morshuis WJ. Thoracoabdominal R21 aortic aneurysm repair: results of conventional open surgery. Eur J Vasc Endovasc Surg 2009 R22 Jun;37(6):640-5. R23 (25) Maniar HS, Sundt TM, III, Prasad SM, Chu CM, Camillo CJ, Moon MR, et al. Delayed 9 R24 paraplegia after thoracic and thoracoabdominal aneurysm repair: a continuing risk. Ann Thorac R25 Surg 2003 Jan;75(1):113-9. R26 (26) Fattori R, Tsai TT, Myrmel T, Evangelista A, Cooper JV, Trimarchi S, et al. Complicated acute R27 type B dissection: is surgery still the best option?: a report from the International Registry of R28 Acute Aortic Dissection. JACC Cardiovasc Interv 2008 Aug;1(4):395-402. R29 (27) Slonim SM, Miller DC, Mitchell RS, Semba CP, Razavi MK, Dake MD. Percutaneous balloon R30 fenestration and stenting for life-threatening ischemic complications in patients with acute aortic R31 dissection. J Thorac Cardiovasc Surg 1999 Jun;117(6):1118-26. R32 (28) Kische S, Ehrlich MP, Nienaber CA, Rousseau H, Heijmen R, Piquet P, et al. Endovascular R33 treatment of acute and chronic aortic dissection: midterm results from the Talent Thoracic R34 Retrospective Registry. J Thorac Cardiovasc Surg 2009 Jul;138(1):115-24. R35 (29) Lupattelli T, Garaci FG, Basile A, Casini A, Dalainas I, Minnella DP, et al. Emergency stent R36 grafting of type B aortic dissection: technical considerations. Emerg Radiol 2008 Nov;15(6):375- R37 82. R38 R39

123

Chapter 10

Circulating Transforming Growth Factor-Beta Levels in Acute Aortic Dissection

Journal of the American College of Cardiology. 2011 Aug 9;58(7):775.

Toru Suzuki, MD,1* Santi Trimarchi, MD,2* Daigo Sawaki, MD,1* Viviana Grassi, MD,2 Elena Costa, MD, 2 Vincenzo Rampoldi, MD,2 Ryozo Nagai, MD,1 Kim Eagle, MD,3

1 The University of Tokyo, Tokyo, Japan 2 Policlinico San Donato IRCCS, San Donato Milanese, Italy 3 University of Michigan, Ann Arbor, Mich, USA

* Toru Suzuki, Santi Trimarchi and Daigo Sawaki contributed equally to this work Chapter 10

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

126 Circulating transforming growth factor-Beta levels in acute aortic dissection

To the Editor: Circulating transforming growth factor (TGF)- beta has received recent attention R1 because it may potentially serve as a biomarker for therapeutic monitoring of aortic remodeling R2 processes in patients with Marfan syndrome.1,2 TGF-beta and its actions are regulated by the R3 elastic protein fibrillin-1. TGF-beta is bound in a complexed form to fibrillin-1, which, on R4 pathogenic insult, is disrupted to release and activate the TGF-beta molecule, which in turn R5 acts on vascular cells to induce aortic remodeling. Mutation of fibrillin-1 leading to aberrant R6 regulation of TGF-beta is thought to be a central pathogenic mechanism of Marfan syndrome.3 R7 Recent discoveries that renin-angiotensin system inhibitors directly act on dysregulation of TGF- R8 beta to affect aortic remodeling have opened up new therapeutic possibilities for this disease.1,4 R9 Whether TGF-beta is increased in aortic conditions in the non-Marfan adult, however, has R10 remained unknown. Here, we show that circulating TGF-beta levels are elevated in patients with R11 acute aortic dissection. Efforts have been made to develop diag- nostic biomarkers of this disease R12 because aortic dissection is a catastrophic aortic disease with high mortality and morbidity that R13 requires immediate diagnosis and treatment. Aortic dissection and rupture are also primary R14 cardiovascular manifestations in patients with Marfan syndrome who reach adolescence. R15 In 28 non-Marfan patients with acute aortic dissection defined as being within 24 h of symptom R16 onset, TGF-beta levels as measured by a commercially available assay (TGF-beta 1 Quan- tikine R17 immunoassay, R&D Systems, Minneapolis, Minnesota) showed elevations at 24.5 ± 12.9 R18 ng/ml (median 25.4 ng/ml) that are approximately 5-fold elevations compared with normal R19 controls (5.4 ± 2.8 ng/ml).5 When categorized according to dissection type, Stanford type A R20 dissections showed approximately 2-fold elevations (p =0.01, nonparametric Mann-Whitney U R21 test) in type A patients (28.5 ± 14.7 ng/ml [n = 20]; median 28.7 ng/ml) compared with type R22 B dissections (14.4 ± 6.1 ng/ml [n = 8]; median 14.6 ng/ml). Circulating TGF-beta levels were R23 therefore markedly elevated in patients with type A dissections, which are often associated with R24 a worse prognosis and require immediate medical attention. Collectively, our findings, although R25 preliminary, suggest that circulating TGF-beta levels may also serve as a potential biomarker of R26 aortic disease in the non-Marfan adult. 10 R27 Further questions that remain to be addressed are whether circulating TGF-beta levels can be used R28 to monitor remodeling dynamics in aortic conditions such as aortic dissection and aneurysms R29 in the non-Marfan adult and whether they will be responsive to pharmacological treatment. For R30 example, whether circulating TGF-beta levels will serve as a reliable surrogate biomarker to assess R31 aortic expansion in type B dissections that might predict the need for surgery and response to R32 medication during the course of treatment will need to be addressed. More sensitive assays maybe R33 neccesary to allow accurate assessment as necessary for therapeutic monitoring in the chronic R34 phase of disease progression. Nonethe- less, we believe that circulating TGF-beta is a promising R35 bio- marker for potential use in diagnostic and therapeutic assessment of aortic diseases. R36 R37 R38 R39

127 Chapter 10

R1 REfERENCES R2 R3 1. Matt P, Schoenhoff F, Habashi J, et al., GenTAC Consortium. Circulating transforming growth R4 factor-beta in Marfan syndrome. Circulation 2009;120: 526-31 –32. R5 2. Ahimastos AA, Aggarwal A, D’Orsa KM, et al. Effect of perindopril on large artery stiffness and R6 aortic root diameter in patients with Marfan syndrome: a randomized controlled trial. JAMA R7 2007; 298: 1539 – 47. R8 3. Judge DP, Dietz HC. Marfan’s syndrome. Lancet 2005;366:1965–76. R9 4. Brooke BS, Habashi JP, Judge DP, Patel N, Loeys B, Dietz HC 3rd. Angiotensin II blockade and R10 aortic-root dilation in Marfan’s syndrome. N Engl J Med 2008;358:2787–95. R11 5. Campistol JM, Inigo P, Jimenez W, et al. Losartan decreases plasma levels of TGF-beta1 in R12 transplant patients with chronic allograft nephropathy. Kidney Int 1999;56: 714-9. R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

128 Chapter 11

In search of blood tests for thoracic aortic diseases

Annals of Thoracic Surgery 2010 Nov;90(5):1735-42.

Santi Trimarchi, MD,1 Giuseppe Sangiorgi, MD,2 Xiangpeng Sang, BS,3 Vincenzo Rampoldi, MD,1 Toru Suzuki, MD,4 Kim A Eagle, MD,5 John A Elefteriades, MD,6

1 Policlinico San Donato IRCCS, San Donato Milanese, Italy 2 University of Modena, Modena, Italy 3 School of Medicine,Yale University, New Haven, Conn, USA 4 The University of Tokyo, Tokyo, Japan 5 University of Michigan, Ann Arbor, Mich, USA Chapter 11

R1 AbSTRACT R2 R3 A number of new diagnostic screening tools have been developed for the assessment of acute R4 and chronic diseases of the thoracic aorta. Although standardized blood-based tests, capable of R5 detecting individuals at risk for aortic aneurysm and dissection disease are not yet available, our R6 current knowledge is expanding at a rapid rate and the future is very promising. In this review, R7 the authors have provided an update of the contemporary knowledge on blood tests for detecting R8 thoracic aortic diseases in both preclinical and clinical settings, offering the potential to predict R9 adverse aortic events, such as enlargement, rupture, and dissection. R10 R11 Abbreviation list R12 AAA abdominal aortic aneurysm R13 AGGF angiogenic factor R14 AD acute dissection R15 CRP C-reactive protein R16 H-FABP heart-type fatty acid –binding protein R17 Lp(a) lipoprotein a R18 MMP matrix metalloproteinase R19 PAI plasminogen activator inhibitor R20 SMMHCP smooth-muscle myosin heavy chain protein R21 SEP serum elastin peptides R22 TAA thoracic aortic aneurysm R23 TAT thrombin anti-thrombin complex R24 TGF-β transforming growth factor β R25 TIMP tissue inhibitor of metalloproteinase R26 PCR polymerase chain reaction R27 RNA ribonucleic acid R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

130 In search of blood tests for thoracic aortic diseases

INTRODUCTION R1 R2 Advanced imaging technologies have allowed insight into the anatomy of thoracic aortic R3 aneurysms (TAA) and acute dissections (AD), providing a robust basis for diagnosing and R4 treating of aortic diseases, based on aortic size. However, these techniques have shortcomings R5 as screening methods, including expense, inconvenience, invasiveness, and potential harm, R6 including radiation exposure and renal injury. Although many cases of aortic aneurysms are R7 detected incidentally, when patients undergo imaging for other conditions, CT scanning and R8 MRI imaging are not felt to be appropriate as screening tests for the general population. The tests R9 do not perform as well in low-risk populations as they do in high-risk populations, nor is their R10 use for this purpose thought to be cost-effective.1, 2 Many patients who have an increased genetic R11 risk of developing aneurysms later in life may not have recognizable enlargement of the aorta at R12 the time of screening, even with state-of-the-art imaging technologies. Therefore we clearly need R13 better methods to screen for TAA. Since early diagnosis of aortic disease remains essential in R14 order to render meaningful progress in preventing acute aortic conditions. R15 Until now, routine laboratory tests have played only a minor role in the assessment of acute R16 and chronic diseases of the aorta. However, a number of new diagnostic screening tools have R17 been developed,3, 4 in order to evaluate the use of biochemical markers in the peripheral blood R18 as initial diagnostic screening. Ideally, a biochemical diagnostic test would be a chemical marker R19 reflective of the pathogenic activity, readily available, inexpensive, and non-invasive. A biological R20 test could be helpful in the initial patient triage at the emergency department of referral centers, R21 helping physicians determine whether aortic dissection should be considered as a possible cause R22 of thoracic pain. Other biomarkers might assist in routine office screening of well patients, and R23 for long term follow-up evaluation of disease progression in patients with aneurysm, offering the R24 potential to predict adverse aortic events, such as enlargement, rupture, and dissection. R25 Recently, several biological markers have been proposed for the diagnosis of acute and chronic R26 aortic diseases. Our aim is update the contemporary knowledge on blood tests for detecting R27 thoracic aortic diseases in both preclinical and clinical settings. R28 R29 11 R30 METHODS R31 R32 For this review, we queeried the PubMed, Cochrane Library CENTRAL and EMBASE databases R33 using the key words : Aortic aneurysm, Aortic dissection, Diagnosis, Biomarkers. R34 R35 D-Dimer. D-dimer is a degradation product of cross-linked fibrin detectable in plasma at levels R36 > 0.5 μg/ml fibrinogen equivalent units in nearly all patients with venous thromboembolism.5 R37 R38 R39

131 Chapter 11

R1 Despite being non-specific, since elevated plasma levels can be found with any intravascular R2 activation of the coagulation system with secondary fibrinolysis, (i.e. various forms of malignancies, R3 infections, renal disease, recent trauma, recent surgery, or disseminated intravascular coagulation), R4 D-dimer has become popular due mainly to its high negative predictive value to exclude deep R5 venous thromboembolism and pulmonary embolism.5-7 R6 Elevated plasma levels of D-dimer were found in small series of patients who experienced R7 aortic dissections.8-10 D-Dimer along with thrombin-antithrombin complex (TAT), which is R8 another hemostatic molecular marker, are useful in predicting morphological changes in acute R9 AD (including longitudinal extent and mortality), and both have been proposed as alternative R10 means of follow-up in patients with AD.9-11 D-dimer has near 100% sensitivity for acute aortic R11 dissection, but very low specificity. A positive test should raise concern about aortic dissection. R12 Conversely, the high negative predictive value of a negative D-dimer test renders the presence R13 of aortic dissection disease very unlikely. Recently, D-dimer levels have been proposed further R14 for risk stratifying patients with suspected aortic dissection; in addition to ruling out aortic R15 dissection (if applied within the first 24 hours after symptom onset).12 These analyses showed R16 that the widely used cut-off level of 500 ng/mL for ruling out pulmonary embolism also can R17 reliably rule out aortic dissection, with a negative likelihood ratio of 0.07 throughout the first R18 24 hours.12 R19 Increased levels of D-dimer have been also found in blood samples of abdominal aortic aneurysm R20 (AAA) endograft patients, where they appeared suitable as a useful marker for fixation problems R21 after endovascular AAA repair, potentially helping to rule out type I endoleak, thus excluding R22 patients from unnecessary invasive imaging tests.13, 14 (Monitoring after stent grafting can also R23 be done by non-invasive ultrasound.) The ability of D-dimer to detect endoleak is thought to R24 be due to the fact that endoleak leads to a blood-filled sac with continued coagulation and R25 thrombolysis.13 Outside of this usefulness for post-stent monitoring thus far, efforts to correlate R26 serum D-dimer or other biomarkers with rate of aneurysm expansion have shown either no or R27 weak clinical correlation.14 R28 Despite the extremely high sensitivity of D-dimer for aortic dissection, the question arises R29 whether a physician on the front lines, highly suspecting aortic dissection, should order a CT R30 scan if the D-dimer comes back negative. Despite the high sensitivity published in the literature, R31 the overall clinical picture and the physician’s clinical judgment remain paramount. R32 D-dimer rises early and predictably after aortic dissection onset (figure 1), but, because aortic R33 dissection must have already occurred for D-dimer to rise, D-dimer does not appear to be a R34 predictor of future dissection events in patients with aortic disease. R35 R36 R37 R38 R39

132 In search of blood tests for thoracic aortic diseases

R1 R2 R3 R4 R5 R6 Circulating levels Circulating R7 R8 0 hrs 6 hrs 12 hrs R9 T ime after onset R10 R11 D-dimer Calponin (smooth muscle troponin-like protein) R12 M yosin R13 C reatine kinase-B B isozyme R14 Figure 1: Circulating level of aortic biomarkers after acute dissection. (Modified from Suzuki T. R15 MMPs (MMP-9) levels are not reported because increase maximally 2 weeks after the acute symptom onset. R16 R17 R18 Matrix metalloproteinases (MMPs). This group of zinc enzymes, responsible for degradation R19 of extracellular matrix components, including fibronectin, collagen, elastin, and proteoglycans R20 in normal and pathologic conditions,15, 16 is synthesized by a variety of connective tissue, R21 parenchymal, and inflammatory cells. MMPs have been found to play an important role in the R22 pathogenesis of aortic aneurysms,17 in which an excess of several members of the MMP family, R23 including MMP-1, MMP-2, MMP-3, and MMP-9 have been demonstrated.15, 18 R24 Patients with AAA exhibit arterial dilation and altered matrix composition throughout the R25 vasculature.19 Phenotypically, there is dissolution and fragmentation of collagen and elastin, R26 which leads to expansion of the vessel wall that can no longer withhold the repetitive expansible R27 force of systolic contraction.16 More specifically, as it relates to MMPs, it has been shown that R28 R29 collagen neosynthesis may decrease with relatively increased collagen degredation in patients 11 with AAA. A meta-analysis showed that circulating MMP-9 concentrations are higher in patients R30 with AAA than those in subjects without AAA.20 R31 In Marfans’ patients affected by AD, tissue overexpression of MMP-1, MMP-2 and MMP- R32 3 have been demonstrated,21 as well as in patients affected by thoracic aneurysm and aortic R33 dissection compared with control specimens.22 A relative index of proteolytic state, the MMP-9/ R34 TIMP-1, ratio was increased in both aortic aneurysm and dissection groups compared with R35 control patients, suggesting an important role of metalloproteinases in the development and R36 progression of aortic disease.22 A temporal increase in MMPs after the acute phase of aortic R37 dissection has been reported,23 showing that plasma MMP-9 expression was maximally elevated R38 R39

133 Chapter 11

R1 at 2 weeks after the acute symptom onset. This suggests that proteolytic enzymes may play R2 an important role in the biology of the response of the aortic wall after the initial dissection R3 phase. This serologic finding is consistent with the histologic observation that local activation R4 of proteolytic enzymes near an inflammatory macrophage infiltrate characterizes the reparative R5 phase after acute dissection. It is also plausible that this remodelling phase may be associated R6 with the dilatation process, permitting stretching of the weakened aortic wall, with progressive R7 expansion and aneurysm formation.24 R8 Alterations in MMP levels in the plasma could also serve as a marker of persistent endoleakage R9 after endovascular AAA treatment. Indeed, the traditional diagnostic approach for endoleak R10 relies mostly on serial computed tomography scan examinations . It has been shown that plasma R11 MMP-9 and MMP-3 levels are significantly elevated in patients with AAA compared with R12 healthy control subjects and that after endovascular graft exclusion, these proteins decrease to R13 a level similar to that of patients who have undergone open surgical repair.25 Additionally, the R14 authors showed that failure of MMP levels to decrease after endovascular graft exclusion may R15 help to identify patients who have endoleak and are at risk for consequent aneurysm expansion. R16 Theoretically, endoleaks might determine a continued aneurysm wall tension and a related R17 increased inflammatory cell response, that maintains high level on MMPs. Conversely, a decrease R18 in the amount of circulating MMPs could represent a simple marker of successful aneurysm R19 exclusion.26 R20 The importance of the above findings should be viewed in the wide context of current AAA R21 treatment. The decision to operate on a patient should be made on an individual basis, since all R22 aneurysms are potential lethal and the rate of expansion and frequency of rupture are not fully R23 predictable in individual patients.27 Because plasma MMP determination by ELISA is a simple R24 and readily available technique, there is the potential for such studies to assess the response to R25 medical, surgical, or interventional treatment instead of monitoring patients with multiple CT R26 scan examinations which is currently the standard of care. Larger studies are needed to confirm R27 the utility of this approach in the various cohorts of patients with acute and chronic aortic R28 disease. R29 R30 Smooth-muscle myosin heavy chain protein (SMMHCP). Smooth muscle myosin heavy R31 chain and its isoforms are structural proteins found in smooth muscle cells.28, 29 Levels of R32 smooth-muscle myosin heavy chain protein can be used to diagnose aortic dissection soon after R33 symptom onset with the use of a rapid 30-minute assay.28-30 This method showed promising R34 results in the experimental studies that were reproduced in the clinical field. More specifically R35 this assay has been found to have moderately high sensitivity (90.9%) and acceptable specificity R36 (98% compared with healthy controls and 83% compared with patients with acute myocardial R37 infarction) in patients presenting within the first three hours of the onset of symptoms.30 It is R38 of note that the assay performed best in patients with proximal lesions and was less sensitive R39

134 In search of blood tests for thoracic aortic diseases

in patients who presented at a later point.30 At present,a reliable commercial platform for this R1 biomarker is not widely available. R2 R3 Calponin. Suzuki and colleagues proposed that an assay of the troponin counterpart of smooth R4 muscle might offer a diagnostic role in aortic dissection given the utility of cardiac troponin R5 as the most reliable biomarker of myocardial ischemia/infarction.31,32 Of the three isoforms of R6 calponin (acidic, basic and neutral), basic calponin (also called h1) is the most abundant and R7 specific isoform in smooth muscle, whereas acidic and neutral calponin (also called h2) are not R8 thought to be specific to smooth muscle (i.e. found in both smooth muscle and non-smooth R9 muscle tissue) and are less abundant than basic calponin in smooth muscle tissue.32 Preliminary R10 study suggests that calponin measurements allowed for detection of aortic dissection in patients R11 with a more delayed presentation (out to 12 hrs, figure 1), a welcome addition for diagnostic R12 use.31 Combined use of calponin and smooth muscle myosin heavy chain assays might allow for R13 improved detection of acute aortic dissection by biochemical means. This would be analogous R14 to use of multiple diagnostic biomarkers for diagnosis of acute myocardial infarction with the R15 initial peak in myoglobin being followed by later elevations in creatine kinase MB-isozyme R16 and troponins. An important corollary to the possibility of a set of biochemical tests for aortic R17 dissection concerns how such tools might interplay with imaging. Like D-dimer, smooth muscle R18 proteins may detect acute aortic dissection after it occurs but are not thought to be useful in R19 chronic monitoring or in predicting that dissection is likely to be forthcoming. R20 R21 Serum markers of elastin and collagen metabolism. Aortic dilatation in TAA is often R22 accompanied by structural modifications of the main components of the aortic tissue, elastin R23 and collagen. This includes a decrease in parallel beta-sheet structures such as elastin.33, 34 In TAA R24 and AAA, the structure and the amount of elastin is changed, and increased levels of elastase and R25 other elastolytic proteases in aneurysmal walls have been demonstrated.35, 36 With the study of the R26 serum elastin peptides (SEP), plasma elastin alpha1 antitrypsin complex, and protocollagen III- R27 N-terminal propeptide (all measured by enzyme linked immunoassay and indicative of increased R28 R29 elastolyis), it has been suggested that increased elastolysis is associated with increased AAA wall 11 distensibility. Also, increased collagen turnover is associated with reduced distensibility.37 All R30 these markers of elastin/collagen metabolism hold promise as potential biomarker for assessing R31 patients with or at risk for TAA and AAA. R32 R33 Cytokines and gene expression signature in peripheral blood. It has become increasingly R34 evident that the immune system plays a pivotal role in the development of aortic aneurysms.22, R35 38-42 Increased circulating levels of inflammatory cytokines such as interleukin IL-1, IL-6, tumor R36 necrosis factor-a, and interferon-γ have been found in patients with AAA.40-43 It has been R37 hypothesized that gene expression patterns in peripheral blood cells may reflect TAA disease R38 R39

135 Chapter 11

R1 status. Elefteriades and colleagues have identified a 41-gene signature in peripheral blood cells R2 that distinguishes TAA patients from controls.44 This set of 41 classifier genes, derived from R3 microarray analysis and further validated by TaqMan® real-time PCR assays, detects TAA disease R4 with an overall classification accuracy of 78–80%. This investigation provided a comprehensive R5 analysis of the gene expression profiles that characterize the peripheral blood cells of TAA R6 patients, which may reflect activities of immune cells at the site of the disease. In fact, many R7 of the biological pathways (and genes) identified in this study, such as interleukin signaling, R8 T-cell activation, and apoptosis, have been previously associated with development of aneurysm R9 diseases.38, 42, 45-49 The pathways implicated in the RNA signature test make biologic “sense” vis-à- R10 vis known aneurysm pathogenesis. R11 Preliminary observations also identified highly predictive distinct gene expression profiles for R12 subtypes of TAA, including ascending vs. descending TAA and familial vs. sporadic TAA.44 R13 These results provide further evidence supporting the current understanding that ascending R14 and descending thoracic aortic aneurysms may represent two very different diseases with widely R15 differing, embryology, pathophysiology, and clinical manifestations.50 The RNA signature R16 findings are also consistent with the increasing recognition that TAA disease is transmitted in R17 an inherited fashion in at least one-fifth of patients.51 Identification of involved RNA pathways R18 opens many interesting avenues to explore in future studies. An angiogenic factor, AGGF1 gene, R19 was identified as one of the potential signature genes distinguishing familial vs. sporadic TAA.44 R20 AGGF1 is located on chromosome 5q13.3, which has been mapped within the previously R21 identified 5q13–14 locus known to be associated with familial TAA.52, 53 Therefore, AGGF1 is R22 being evaluated as a candidate gene for TAA. R23 Gene expression profiles of peripheral blood cells may provide additional information to genetic R24 studies of TAA and help to identify potential disease-associated genes for familial TAA. In R25 particular, identification of a specific signature for sporadic TAA also holds promise in general R26 screening of TAA disease, because traditional radiographic imaging methods or genetic tests that R27 are potentially applicable to known affected families would not be cost-effective for screening R28 large populations. R29 The gene expression signature, as a dynamic phenomenon potentially reflecting the R30 pathophysiology of thoracic aortic aneurysm, also holds promise for advance detection of R31 impending rupture and dissection. R32 R33 Acute phase reactants. C-reactive protein (CRP) and white blood cells levels have been found to R34 be elevated in patients who presented with symptoms or rupture of an AAA.54, 55 Recently, it has R35 been shown that leukocyte count is associated with aortic arch plaque thickness and is especially R36 correlated with plaques > 4mm.56 Elevated serum CRP was highly correlated with asymptomatic R37 AAAs and was associated with aneurysmal size.57 Based on these data, high sensitivity CRP R38 may serve as a marker of AAA and it is conceivable that CRP produced in vascular tissue may R39

136 In search of blood tests for thoracic aortic diseases

contribute to aneurysm formation.54-57 It has been shown that CRP is produced, among other R1 disease states, in brain tissue of patients with Alzheimer’s disease,58 in coronary plaques,59 and R2 in myocardial infarction.60 These data suggest that elevations in CRP represent a generalized R3 reaction to the vascular injury and its measurement is not specific for early identification or R4 rupture prediction in aortic aneurysm disease. R5 R6 lipoprotein (a). Serum lipoprotein (Lp)(a) is a complex lipoprotein that consists of a molecule R7 in which the apoliporpotein B100 is linked to apoliprotein(a), comprising a structure similar to R8 plasminogen, which has been implicated in atherosclerotic and coronary and aortic diseases.61-64 R9 Althought Lp(a) plasma levels were found to be significantly increased in patients with AAA R10 (independently of cardiovascular risk factors and the extent of atherosclerosis),65 a causal R11 relationship between Lp(a) and AAA is not clear. The role of increased Lp(a) levels as biomarker R12 of thoracic and abdominal aortic diseases is debated.66-69 R13 R14 Activators and inhibitors of plasminogen, fibrinogen, and haemostatic markers. It has R15 been suggested that plasmin is a common activator of the known protelolytic systems involved R16 in aneurysmal degradation. Plasmin has been reported to be associated with the expansion of R17 AAA. More specifically, it has been demonstrated that the aortic matrix degradation in AAA R18 may be partly caused by an activation of plasminogen by tissue plasminogen activator.62, 70 R19 In addition, fibrinogen, which is an independent risk factor for coronary events and stroke, R20 has been found as an independent marker for thoracic aortic plaque related to the severity of R21 thoracic aortic atherosclerosis.59 Fibrinogen may act through many mechanisms affecting blood R22 viscosity, rheology, platelet aggregation, and endothelial function.63, 64, 70, 71 Fibrinolytic activity R23 and gene expression have been shown within the aortic wall of patients with asymptomatic R24 AAA.72 It has been demonstrated that preoperative haemostatic markers, particularly plasma R25 prothrombin fragment and plasminogen activator inhibitor (PAI) activity, may distinguish R26 acutely symptomatic ruptured from non-ruptured AAA.72, 73 Despite the importance of the above R27 findings, a rapid assay of PAI is not yet available, and the proper role for such tests in clinical R28 practice are not yet clear. R29 11 R30 Endothelin and hepatocyte growth factor. Endothelin is an endothelium–derived contracting R31 factor, which regulates vascular tone, and is a well known potent vasoconstrictor that plays R32 a major role in vasospasm following bleeding from cerebral aneurysms.74 It has long been R33 hypothesized that pathophysiological changes in the endothelium may play an important role in R34 the pathogenesis of aortic aneurysms. The endothelin pathway, including endothelin-converting R35 enzyme 2 (ECE-2), is up-regulated in TAA. The potential involvement of endothelin-converting R36 enzymes in the pathogenesis of thoracic aneurysms has not been extensively evaluated. Its release R37 is stimulated by a variety of factors such as angiotensin II, vasopressin, adrenaline, hypoxia, and R38 R39

137 Chapter 11

R1 vascular injury.75-78 Some of these processes may be implicated in the in the pathogenesis of R2 aortic aneurysms. Endothelin 1,2 levels were studied in patients with large and/or symptomatic R3 AAA and were found to be significantly increased in patients with large comparing to small R4 aneurysms, thus constituting a potential marker of aneurysm diameter.79 In addition, levels of R5 plasma hepatocyte growth factor, a member of the endothelial-specific growth factors with the R6 greatest mitogenic activity, have been proposed to predict the presence of atherosclerotic lesions R7 in the abdominal aorta and femoral arteries.80 Additional studies of alterations in endothelin R8 levels in aneurysms are needed to validate these preliminary observations. R9 R10 Homocysteine. It has been shown that increased plasma homocysteine is a risk factor for R11 coronary events, stroke, and peripheral atherosclerotic disease.81-83 In addition, plasma R12 homocysteine is significantly and independently correlated with the degree of atherosclerosis in R13 the thoracic aorta.83, 84 However, despite its established role as a marker of atherosclerotic lesions R14 in major arterial locations, any role as a useful adjunct in early diagnosis of aortic disease and its R15 complications is yet to be confirmed. R16 R17 TGf-β. TGF-β is a protein that controls cell cycle and apoptosis and is involved in many R18 pathways relevant to cell growth, differentiation and oncogenesis. Recent work has implicated R19 excess TGF-β activity in the pathophysiology of aneurysm formation in Marfan’s disease, and R20 a human trial of the angiotensin receptor blocker Losartan in which reduces TGF-B levels, is R21 underway in Marfan patients. Very recent evidence from Dietz et al has found that in a mice R22 model of Marfan syndrome serum levels of TGF-β correlate with rate of growth of the aorta. R23 They propose that this may became a useful clinical marker in humans.85 R24 R25 R26 Serum heart-type fatty acid-binding protein. Heart-type fatty acid –binding protein (H-FABP) R27 is a low molecular weight (15 kD) cytoplasmic protein, abundant in cardio-myocytes, known R28 to be a sensitive marker for early detection of myocardial injury. Elevated H-FABP serum levels R29 have also been reported in patients with AD.86 In one of these investigations, Hazui and co- R30 workers87 postulated that in patients with AD, elevated serum H-FABP concentration is possibly R31 caused by protein release not only from skeletal muscle and , but also from aortic R32 wall. The levels of H-FABP appeared to correlate with aortic flap extension, regardless of whether R33 or not patients had shock or renal dysfunction. Since the initial report studied a small number of R34 patients, this biomarker requires further study in order to be considered as a potential diagnostic R35 marker in AD. R36 R37 R38 R39

138 In search of blood tests for thoracic aortic diseases

CONClUSIONS R1 R2 The rapidly aging patient population in Western societies and the resultant increase in the aortic R3 disease, coupled with the unpredictably virulent clinical nature of aortic aneurysm, cry out for R4 advances in biomarker detection of aneurysm activity. Several biochemical markers potentially R5 may be predictive of clinical disease (table 1). Practical limitations to implementation of R6 biomarker-based diagnosis and prediction relate to the need that potential candidate assays be R7 readily employable in clinical settings, have commercial availability, standardized methodologies, R8 and have adequate clinical precision. It is hoped that, utilization of serologic biomarkers to R9 diagnose and follow aortic diseases may be coupled synergistically with imaging to improve R10 diagnostic accuracy and cost-effectiveness of aortic care. R11 As an example, Suzuki et al have proposed the utilization of a diagnostic algorithm consisting R12 of positive biochemical smooth muscle myosin heavy chain level associated with widened R13 mediastinum on chest X-ray, which may represent a tenable solution for early diagnosis of the R14 very early presenting cohort with suspected type A aortic dissection.88 R15 Use of biomarkers to detect aortic dissection after it occurs is already a clinical reality (esp. R16 D-Dimer). A bigger challenge is to detect impending dissection or rupture before they occur. R17 It is hoped that some of the reported potential biomarkers, such as the RNA Signature, may R18 permit screening diagnosis of asymptomatic TAA patients with high accuracy. These efforts R19 can potentially lead to a diagnostic “point-of-care” assay for identification of asymptomatic R20 TAA disease with a simple blood-based test. Clinicians may be able to use markers of tissue R21 inflammation, matrix degeneration, and fibrinolysis (biological processes associated with aortic R22 aneurysm) to determine the presence of TAA as well as the stage of pathogenic activity and the R23 subclass of the aneurysm (e.g. genetic or sporadic). The biomarkers of interest, it is hoped, may R24 also lead to prediction of growth, rupture, dissection and appropriateness of medical and surgical R25 intervention in patients with already identified aneurysm disease. Use of biomarkers to indicate R26 biological activity of the aneurysm and to guide management may be even more important R27 than diagnostic use. Furthermore, transcriptional programs in peripheral blood leading to the R28 R29 identification of these markers also provide insights into the mechanism of development of aortic 11 aneurysms. R30 The utilization of biomarkers could lead to further improvements in diagnostic pathways in R31 acute and chronic aortic diseases, highlighting potential targets for therapeutic intervention and R32 establishing uniform, evidence-based follow-up programs. Although a standardized blood-based R33 test or tests, capable of detecting individuals at risk for aortic aneurysm and dissection disease R34 is not still available, our current knowledge is expanding at a rapid rate and the future is very R35 promising. R36 R37 R38 R39

139 Chapter 11

R1 R2 R3 R4 R5 R6 R7 R8 R9 Potential diagnostic marker in acute diagnostic marker Potential dissection Implicated as possible “final common Implicated for aortic aneurysm disease pathway” Correlates with degree of atherosclerosis of atherosclerosis with degree Correlates in thoracic aorta Level correlates well with size of AAA with size well correlates Level They may distinguish acutely non-ruptured from symptomatic ruptured AAA Also elevated in atherosclerosis, limiting in atherosclerosis, Also elevated utility in TAA Wide range of tissues origin and disease Wide states 70% - 80% overall accuracy in diagnosis 70% - 80% overall of TAA Intimately implicated in pathophysiology Intimately of TAA Direct insight in pathogenesis, but Direct unexplained as a biomarker SMC troponin counterpart remains counterpartSMC troponin remains 24 hrs beyond diagnostic even It can diagnoses acute dissection only It onset symptom’s within 6 hours from Useful as marker to detect endoleaks as marker Useful follow-up. during EVAR Useful as marker to detect endoleaks as marker Useful follow-up. during EVAR Comments R10 R11 R12 R13 R14 R15 R16 ? Yes (in mice) ? ? ? ? Possible Possible ? ? No No Unknown No R17 of Growth, for Prediction Use Dissection Rupture, R18 R19 R20 R21 ? ? ? ? ? Yes (non-specific) Yes (non-specific) Yes ? ? No No Possible No R22 of Aneurysm R23 R24 R25 R26 R27 R28 R29 ? ? ? Yes Yes AAA) (ruptured ? Yes (aortic but non-specific) rupture ? ? ? Yes Yes 98% specific) (90.9% sensitive, Yes Yes (99.9% sensitivity, non-specific) R30 of AAD for Diagnosis Use for Diagnosis Use R31 ) R32 γ R33 R34 , interferon-

R35 α R36 R37 R38 β Heart-type fatty acid-binding proteinHeart-type (non-specific) Yes TGF- Homocysteine Endothelin Endothelin Factor Frowh Hepatocyte Hemostatic markers Hemostatic (plasmin, fibrinogen, plasma fragment, PAI) thrombin Lipoprotein (a) Lipoprotein CRP “RNA signature” Cytokines (IL-1, TNF Il-6, Elastin/Collagen Markers III) ANC, protocollagen PEA1, (SEP, Calponin SMMHCP MMPs D-Dimer Biomarker

R39 Utility and Knowledge Current Dissection: Aortic in Thoracic 1. Biomarkers Table

140 In search of blood tests for thoracic aortic diseases

REfERENCES R1 R2 (1) Barbant SD, Eisenberg MJ, Schiller NB. The diagnostic value of imaging techniques for aortic R3 dissection. Am Heart J 1992 August;124(2):541-3. R4 (2) Nienaber CA, von KY, Nicolas V et al. The diagnosis of thoracic aortic dissection by noninvasive R5 imaging procedures. N Engl J Med 1993 January 7;328(1):1-9. R6 (3) Hovsepian DM, Ziporin SJ, Sakurai MK, Lee JK, Curci JA, Thompson RW. Elevated plasma R7 levels of matrix metalloproteinase-9 in patients with abdominal aortic aneurysms: a circulating R8 R9 Potential diagnostic marker in acute diagnostic marker Potential dissection Implicated as possible “final common Implicated for aortic aneurysm disease pathway” Correlates with degree of atherosclerosis of atherosclerosis with degree Correlates in thoracic aorta

Level correlates well with size of AAA with size well correlates Level marker of degenerative aneurysm disease. J Vasc Interv Radiol 2000 November;11(10):1345-52. They may distinguish acutely non-ruptured from symptomatic ruptured AAA Also elevated in atherosclerosis, limiting in atherosclerosis, Also elevated utility in TAA Wide range of tissues origin and disease Wide states 70% - 80% overall accuracy in diagnosis 70% - 80% overall of TAA Intimately implicated in pathophysiology Intimately of TAA Direct insight in pathogenesis, but Direct unexplained as a biomarker SMC troponin counterpart remains counterpartSMC troponin remains 24 hrs beyond diagnostic even It can diagnoses acute dissection only It onset symptom’s within 6 hours from Useful as marker to detect endoleaks as marker Useful follow-up. during EVAR Useful as marker to detect endoleaks as marker Useful follow-up. during EVAR Comments (4) Suzuki T, Katoh H, Watanabe M et al. Novel biochemical diagnostic method for aortic dissection. R10 Results of a prospective study using an immunoassay of smooth muscle myosin heavy chain. R11 Circulation 1996 March 15;93(6):1244-9. R12 (5) Goldhaber SZ, Simons GR, Elliott CG et al. Quantitative plasma D-dimer levels among patients R13 undergoing pulmonary angiography for suspected pulmonary embolism. JAMA 1993 December R14 15;270(23):2819-22. R15 (6) Bick RL. Disseminated intravascular coagulation. Objective laboratory diagnostic criteria and R16 ? Yes (in mice) ? ? ? ? Possible Possible ? ? No No Unknown No Use for Prediction of Growth, of Growth, for Prediction Use Dissection Rupture, guidelines for management. Clin Lab Med 1994 December;14(4):729-68. R17 (7) Costantini V, Zacharski LR. Fibrin and cancer. Thromb Haemost 1993 May 3;69(5):406-14. R18 (8) Weber T, Hogler S, Auer J et al. D-dimer in acute aortic dissection. Chest 2003 May;123(5):1375- R19 8. R20 (9) Nomura F, Tamura K, Yoshitatsu M, Katayama A, Katayama K, Ihara K. Changes in coagulation R21 ? ? ? ? ? Yes (non-specific) Yes (non-specific) Yes ? ? No No Possible No of Aneurysm condition, cytokine, adhesion molecule after repair of type A aortic dissection. Eur J Cardiothorac R22 Surg 2004 August;26(2):348-50. R23 (10) Eggebrecht H, Naber CK, Bruch C et al. Value of plasma fibrin D-dimers for detection of acute R24 aortic dissection. J Am Coll Cardiol 2004 August 18;44(4):804-9. R25 (11) Iyano K, Kawada T, Aiba M, Takaba T. Correlation of hemostatic molecular markers and R26 morphology of the residual false lumen in chronic aortic dissection. Ann Thorac Cardiovasc Surg R27 2004 April;10(2):106-12. R28 (12) Suzuki T, Distante A, Zizza A et al. Diagnosis of acute aortic dissection by D-dimer: the R29 ? ? ? Yes Yes AAA) (ruptured ? Yes (aortic but non-specific) rupture ? ? ? Yes Yes 98% specific) (90.9% sensitive, Yes Yes (99.9% sensitivity, non-specific) Use for Diagnosis of AAD for Diagnosis Use for Diagnosis Use 11 International Registry of Acute Aortic Dissection Substudy on Biomarkers (IRAD-Bio) R30 experience. Circulation 2009 May 26;119(20):2702-7. R31 ) γ (13) Serino F, Abeni D, Galvagni E et al. Noninvasive diagnosis of incomplete endovascular aneurysm R32 repair: D-dimer assay to detect type I endoleaks and nonshrinking aneurysms. J Endovasc Ther R33 2002 February;9(1):90-7. R34 , interferon-

α (14) Urbonavicius S, Urbonaviciene G, Honore B, Henneberg EW, Vorum H, Lindholt JS. Potential R35 circulating biomarkers for abdominal aortic aneurysm expansion and rupture--a systematic R36 review. Eur J Vasc Endovasc Surg 2008 September;36(3):273-80. R37 β (15) Vine N, Powell JT. Metalloproteinases in degenerative aortic disease. Clin Sci (Lond) 1991 R38 Heart-type fatty acid-binding proteinHeart-type (non-specific) Yes TGF- Homocysteine Endothelin Endothelin Factor Frowh Hepatocyte Hemostatic markers Hemostatic (plasmin, fibrinogen, plasma fragment, PAI) thrombin Lipoprotein (a) Lipoprotein CRP “RNA signature” Cytokines (IL-1, TNF Il-6, Elastin/Collagen Markers III) ANC, protocollagen PEA1, (SEP, Calponin SMMHCP MMPs D-Dimer Biomarker Table 1. Biomarkers in Thoracic Aortic Dissection: Current Knowledge and Utility and Knowledge Current Dissection: Aortic in Thoracic 1. Biomarkers Table August;81(2):233-9. R39

141 Chapter 11

R1 (16) Knox JB, Sukhova GK, Whittemore AD, Libby P. Evidence for altered balance between matrix R2 metalloproteinases and their inhibitors in human aortic diseases. Circulation 1997 January R3 7;95(1):205-12. R4 (17) Gonzalez AA, Segura AM, Horiba K et al. Matrix metalloproteinases and their tissue inhibitors R5 in the lesions of cardiac and pulmonary sarcoidosis: an immunohistochemical study. Hum Pathol R6 2002 December;33(12):1158-64. R7 (18) Toshiharu I, Noriko A. Collagen and Elastin Degradation by Matrix Proteinases and Tissue R8 Inhibitors in the Lesions of Cardiac and Pulmonary Sarcoidosis: An Immunohistochemical R9 Study. Hum Pathol 2000;31:640-6. R10 (19) McDonnell S, Morgan M, Lynch C. Role of matrix metalloproteinases in normal and disease R11 processes. Biochem Soc Trans 1999 August;27(4):734-40. R12 (20) Takagi H, Manabe H, Kawai N, Goto SN, Umemoto T. Circulating matrix metalloproteinase-9 R13 concentrations and abdominal aortic aneurysm presence: a meta-analysis. Interact Cardiovasc R14 Thorac Surg 2009 September;9(3):437-40. R15 (21) Matrisian LM. The matrix-degrading metalloproteinases. Bioessays 1992 July;14(7):455-63. R16 (22) Koullias GJ, Ravichandran P, Korkolis DP, Rimm DL, Elefteriades JA. Increased tissue R17 microarray matrix metalloproteinase expression favors proteolysis in thoracic aortic aneurysms R18 and dissections. Ann Thorac Surg 2004 December;78(6):2106-10. R19 (23) Sangiorgi G, Trimarchi S, Mauriello A et al. Plasma levels of metalloproteinases-9 and -2 in the R20 acute and subacute phases of type A and type B aortic dissection. J Cardiovasc Med (Hagerstown R21 ) 2006 May;7(5):307-15. R22 (24) Zhang X, Shen YH, LeMaire SA. Thoracic aortic dissection: are matrix metalloproteinases R23 involved? Vascular 2009 May;17(3):147-57. R24 (25) Sangiorgi G, D’Averio R, Mauriello A et al. Plasma levels of metalloproteinases-3 and -9 as R25 markers of successful abdominal aortic aneurysm exclusion after endovascular graft treatment. R26 Circulation 2001 September 18;104(12 Suppl 1):I288-I295. R27 (26) Ferrans VJ. New insights into the world of matrix metalloproteinases. Circulation 2002 January R28 29;105(4):405-7. R29 (27) Mortality results for randomised controlled trial of early elective surgery or ultrasonographic R30 surveillance for small abdominal aortic aneurysms. The UK Small Aneurysm Trial Participants. R31 Lancet 1998 November 21;352(9141):1649-55. R32 (28) Katoh H, Suzuki T, Yokomori K et al. A novel immunoassay of smooth muscle myosin heavy R33 chain in serum. J Immunol Methods 1995 September 11;185(1):57-63. R34 (29) Katoh H, Suzuki T, Hiroi Y et al. Diagnosis of aortic dissection by immunoassay for circulating R35 smooth muscle myosin. Lancet 1995 January 21;345(8943):191-2. R36 (30) Suzuki T, Katoh H, Tsuchio Y et al. Diagnostic implications of elevated levels of smooth-muscle R37 myosin heavy-chain protein in acute aortic dissection. The smooth muscle myosin heavy chain R38 study. Ann Intern Med 2000 October 3;133(7):537-41. R39

142 In search of blood tests for thoracic aortic diseases

(31) Suzuki T, Distante A, Zizza A et al. Preliminary experience with the smooth muscle troponin-like R1 protein, calponin, as a novel biomarker for diagnosing acute aortic dissection. Eur Heart J 2008 R2 June;29(11):1439-45. R3 (32) Walsh MP. Calponin--knocked out but not down! J Physiol 2000 December 15;529 Pt 3:517. R4 (33) Iliopoulos DC, Kritharis EP, Giagini AT, Papadodima SA, Sokolis DP. Ascending thoracic aortic R5 aneurysms are associated with compositional remodeling and vessel stiffening but not weakening R6 in age-matched subjects. J Thorac Cardiovasc Surg 2009 January;137(1):101-9. R7 (34) Bonnier F, Rubin S, Debelle L et al. FTIR protein secondary structure analysis of human R8 ascending aortic tissues. J Biophotonics 2008 August;1(3):204-14. R9 (35) Satta J, Juvonen T, Haukipuro K, Juvonen M, Kairaluoma MI. Increased turnover of collagen in R10 abdominal aortic aneurysms, demonstrated by measuring the concentration of the aminoterminal R11 propeptide of type III procollagen in peripheral and aortal blood samples. J Vasc Surg 1995 R12 August;22(2):155-60. R13 (36) Sans M, Moragas A. Mathematical morphologic analysis of the aortic medial structure. R14 Biomechanical implications. Anal Quant Cytol Histol 1993 April;15(2):93-100. R15 (37) Wilson K, Bradbury A, Whyman M et al. Relationship between abdominal aortic aneurysm R16 wall compliance and clinical outcome: a preliminary analysis. Eur J Vasc Endovasc Surg 1998 R17 June;15(6):472-7. R18 (38) Tang PC, Yakimov AO, Teesdale MA et al. Transmural inflammation by interferon-gamma- R19 producing T cells correlates with outward vascular remodeling and intimal expansion of ascending R20 thoracic aortic aneurysms. FASEB J 2005 September;19(11):1528-30. R21 (39) Koullias G, Modak R, Tranquilli M, Korkolis DP, Barash P, Elefteriades JA. Mechanical R22 deterioration underlies malignant behavior of aneurysmal human ascending aorta. J Thorac R23 Cardiovasc Surg 2005 September;130(3):677-83. R24 (40) Walton LJ, Franklin IJ, Bayston T et al. Inhibition of prostaglandin E2 synthesis in abdominal R25 aortic aneurysms: implications for smooth muscle cell viability, inflammatory processes, and the R26 expansion of abdominal aortic aneurysms. Circulation 1999 July 6;100(1):48-54. R27 (41) Davis VA, Persidskaia RN, Baca-Regen LM, Fiotti N, Halloran BG, Baxter BT. Cytokine pattern R28 R29 in aneurysmal and occlusive disease of the aorta. J Surg Res 2001 December;101(2):152-6. 11 (42) Schonbeck U, Sukhova GK, Gerdes N, Libby P. T(H)2 predominant immune responses prevail R30 in human abdominal aortic aneurysm. Am J Pathol 2002 August;161(2):499-506. R31 (43) Jones KG, Brull DJ, Brown LC et al. Interleukin-6 (IL-6) and the prognosis of abdominal aortic R32 aneurysms. Circulation 2001 May 8;103(18):2260-5. R33 (44) Wang Y, Barbacioru CC, Shiffman D et al. Gene expression signature in peripheral blood detects R34 thoracic aortic aneurysm. PLoS One 2007;2(10):e1050. R35 (45) Henderson EL, Geng YJ, Sukhova GK, Whittemore AD, Knox J, Libby P. Death of smooth R36 muscle cells and expression of mediators of apoptosis by T lymphocytes in human abdominal R37 aortic aneurysms. Circulation 1999 January 5;99(1):96-104. R38 R39

143 Chapter 11

R1 (46) Dawson J, Cockerill GW, Choke E, Belli AM, Loftus I, Thompson MM. Aortic aneurysms R2 secrete interleukin-6 into the circulation. J Vasc Surg 2007 February;45(2):350-6. R3 (47) Shimizu K, Mitchell RN, Libby P. Inflammation and cellular immune responses in abdominal R4 aortic aneurysms. Arterioscler Thromb Vasc Biol 2006 May;26(5):987-94. R5 (48) Shimizu K, Shichiri M, Libby P, Lee RT, Mitchell RN. Th2-predominant inflammation and R6 blockade of IFN-gamma signaling induce aneurysms in allografted aortas. J Clin Invest 2004 R7 July;114(2):300-8. R8 (49) Middleton RK, Lloyd GM, Bown MJ, Cooper NJ, London NJ, Sayers RD. The pro-inflammatory R9 and chemotactic cytokine microenvironment of the abdominal aortic aneurysm wall: a protein R10 array study. J Vasc Surg 2007 March;45(3):574-80. R11 (50) Elefteriades JA. Natural History: Evidence-based Indications for Operation. In: Coselli J, editor. R12 Surgery of the Aortic Arch. 2007. R13 (51) Albornoz G, Coady MA, Roberts M et al. Familial thoracic aortic aneurysms and dissections- R14 -incidence, modes of inheritance, and phenotypic patterns. Ann Thorac Surg 2006 R15 October;82(4):1400-5. R16 (52) Guo D, Hasham S, Kuang SQ et al. Familial thoracic aortic aneurysms and dissections: genetic R17 heterogeneity with a major locus mapping to 5q13-14. Circulation 2001 May 22;103(20):2461- R18 8. R19 (53) Kakko S, Raisanen T, Tamminen M et al. Candidate locus analysis of familial ascending aortic R20 aneurysms and dissections confirms the linkage to the chromosome 5q13-14 in Finnish families. R21 J Thorac Cardiovasc Surg 2003 July;126(1):106-13. R22 (54) Lagrand WK, Visser CA, Hermens WT et al. C-reactive protein as a cardiovascular risk factor: R23 more than an epiphenomenon? Circulation 1999 July 6;100(1):96-102. R24 (55) Domanovits H, Schillinger M, Mullner M et al. Acute phase reactants in patients with abdominal R25 aortic aneurysm. Atherosclerosis 2002 August;163(2):297-302. R26 (56) Elkind MS, Sciacca R, Boden-Albala B, Homma S, Di Tullio MR. Leukocyte count is associated R27 with aortic arch plaque thickness. Stroke 2002 November;33(11):2587-92. R28 (57) Vainas T, Lubbers T, Stassen FR et al. Serum C-reactive protein level is associated with abdominal R29 aortic aneurysm size and may be produced by aneurysmal tissue. Circulation 2003 March R30 4;107(8):1103-5. R31 (58) Yasojima K, Schwab C, McGeer EG, McGeer PL. Human neurons generate C-reactive protein R32 and amyloid P: upregulation in Alzheimer’s disease. Brain Res 2000 December 22;887(1):80-9. R33 (59) Yasojima K, Schwab C, McGeer EG, McGeer PL. Generation of C-reactive protein and R34 complement components in atherosclerotic plaques. Am J Pathol 2001 March;158(3):1039-51. R35 (60) Yasojima K, Schwab C, McGeer EG, McGeer PL. Human heart generates complement R36 proteins that are upregulated and activated after myocardial infarction. Circ Res 1998 October R37 19;83(8):860-9. R38 R39

144 In search of blood tests for thoracic aortic diseases

(61) Tribouilloy C, Peltier M, Colas L et al. Fibrinogen is an independent marker for thoracic aortic R1 atherosclerosis. Am J Cardiol 1998 February 1;81(3):321-6. R2 (62) Lindholt JS, Jorgensen B, Shi GP, Henneberg EW. Relationships between activators and inhibitors R3 of plasminogen, and the progression of small abdominal aortic aneurysms. Eur J Vasc Endovasc R4 Surg 2003 June;25(6):546-51. R5 (63) Koenig W, Hombach V, Ernst E, Sund M, Mraz W, Keil U. Plasma viscosity as a cardiovascular R6 risk factor. Circulation 1992 September;86(3):1045. R7 (64) Koenig W, Sund M, Ernst E, Mraz W, Hombach V, Keil U. Association between rheology and R8 components of lipoproteins in human blood. Results from the MONICA project. Circulation R9 1992 June;85(6):2197-204. R10 (65) Schillinger M, Domanovits H, Ignatescu M et al. Lipoprotein (a) in patients with aortic R11 aneurysmal disease. J Vasc Surg 2002 July;36(1):25-30. R12 (66) Peltier M, Iannetta Peltier MC, Sarano ME, Lesbre JP, Colas JL, Tribouilloy CM. Elevated serum R13 lipoprotein(a) level is an independent marker of severity of thoracic aortic atherosclerosis. Chest R14 2002 May;121(5):1589-94. R15 (67) Papagrigorakis E, Iliopoulos D, Asimacopoulos PJ et al. Lipoprotein(a) in plasma, arterial wall, R16 and thrombus from patients with aortic aneurysm. Clin Genet 1997 November;52(5):262-71. R17 (68) Norrgard O, Angquist KA, Dahlen G. High concentrations of Lp(a) lipoprotein in serum are R18 common among patients with abdominal aortic aneurysms. Int Angiol 1988 January;7(1):46-9. R19 (69) Lindholt JS. Screening for abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 2003 R20 May;25(5):377-9. R21 (70) Ernst E. Fibrinogen, viscosity, and white blood cell count. Circulation 1992 May;85(5):1956. R22 (71) Ernst E, Resch KL. Fibrinogen as a cardiovascular risk factor: a meta-analysis and review of the R23 literature. Ann Intern Med 1993 June 15;118(12):956-63. R24 (72) Adam DJ, Haggart PC, Ludlam CA, Bradbury AW. Hemostatic markers before operation in R25 patients with acutely symptomatic nonruptured and ruptured infrarenal abdominal aortic R26 aneurysm. J Vasc Surg 2002 April;35(4):661-5. R27 (73) Adam DJ, Ludlam CA, Ruckley CV, Bradbury AW. Coagulation and fibrinolysis in patients R28 R29 undergoing operation for ruptured and nonruptured infrarenal abdominal aortic aneurysms. J 11 Vasc Surg 1999 October;30(4):641-50. R30 (74) Lin CL, Winardi W, Jeng AY, Kwan AL. Endothelin-converting enzyme inhibitors for the treatment R31 of subarachnoid hemorrhage-induced vasospasm. Neurol Res 2006 October;28(7):721-9. R32 (75) Cannan CR, Burnett JC, Jr., Brandt RR, Lerman A. Endothelin at pathophysiological R33 concentrations mediates coronary vasoconstriction via the endothelin-A receptor. Circulation R34 1995 December 1;92(11):3312-7. R35 (76) Lerman A, Holmes DR, Jr., Bell MR, Garratt KN, Nishimura RA, Burnett JC, Jr. Endothelin R36 in coronary endothelial dysfunction and early atherosclerosis in humans. Circulation 1995 R37 November 1;92(9):2426-31. R38 R39

145 Chapter 11

R1 (77) Matheis G, Haak T, Beyersdorf F, Baretti R, Polywka C, Winkelmann BR. Circulating endothelin R2 in patients undergoing coronary artery bypass grafting. Eur J Cardiothorac Surg 1995;9(5):269- R3 74. R4 (78) Filep JG, Bodolay E, Sipka S, Gyimesi E, Csipo I, Szegedi G. Plasma endothelin correlates with R5 antiendothelial antibodies in patients with mixed connective tissue disease. Circulation 1995 R6 November 15;92(10):2969-74. R7 (79) Treska V, Wenham PW, Valenta J, Topolcan O, Pecen L. Plasma endothelin levels in patients with R8 abdominal aortic aneurysms. Eur J Vasc Endovasc Surg 1999 May;17(5):424-8. R9 (80) Tateishi J, Waku S, Masutani M, Ohyanagi M, Iwasaki T. Hepatocyte growth factor as a R10 potential predictor of the presence of atherosclerotic aorto-iliac artery disease. Am Heart J 2002 R11 February;143(2):272-6. R12 (81) Coull BM, Malinow MR, Beamer N, Sexton G, Nordt F, de GP. Elevated plasma homocyst(e)ine R13 concentration as a possible independent risk factor for stroke. Stroke 1990 April;21(4):572-6. R14 (82) Graham IM, Daly LE, Refsum HM et al. Plasma homocysteine as a risk factor for vascular R15 disease. The European Concerted Action Project. JAMA 1997 June 11;277(22):1775-81. R16 (83) Konecky N, Malinow MR, Tunick PA et al. Correlation between plasma homocyst(e)ine and R17 aortic atherosclerosis. Am Heart J 1997 May;133(5):534-40. R18 (84) Tribouilloy CM, Peltier M, Iannetta Peltier MC, Trojette F, Andrejak M, Lesbre JP. Plasma R19 homocysteine and severity of thoracic aortic atherosclerosis. Chest 2000 December;118(6):1685-9. R20 (85) Matt P. Proteomics and the Role of Losartan. 2009. R21 (86) Seino Y, Ogata K, Takano T et al. Use of a whole blood rapid panel test for heart-type fatty R22 acid-binding protein in patients with acute chest pain: comparison with rapid troponin T and R23 myoglobin tests. Am J Med 2003 August 15;115(3):185-90. R24 (87) Hazui H, Negoro N, Nishimoto M et al. Serum heart-type fatty acid-binding protein concentration R25 positively correlates with the length of aortic dissection. Circ J 2005 August;69(8):958-61. R26 (88) Suzuki T, Trimarchi S, Smith D. Early Diagnosis of Acute Aortic Dissection: Identification R27 of Clinical Variables associated with Early Diagnosis and determination of the usefulnes of R28 biochemical diagnosis as shown by the International Regirstry of Acute Aortic Diseection (IRAD) R29 Database. Circulation 2004;110(3):370. R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

146 Chapter 12

Summary and General discussion Chapter 12

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

148 Summary and General discussion

SUMMARy R1 R2 Acute aortic dissection is a catastrophic event associated with high morbidity and mortality. In R3 the last years new methods of early diagnosis and management have improved the results. In the R4 same time, new perceptions have been recognized and utilized for preventing the acute aortic R5 event or as hypothesis for additional researches. Our studies showed as the diameter may not be R6 considered as a fundamental parameter to predict acute dissection, highlighting as it may occurs R7 frequently in patients with a normal aortic diameter and not affected by predisposing factors R8 like connective tissue disorders. In acute dissected patients requiring open or/and endovascular R9 approaches, results are strictly associated with pre-operative clinical conditions. In acute type B R10 aortic dissection (ABAD) patients, presence of refractory pain and hypertension in the absence R11 of any hemodynamic and/or ischemic complication, categorizes a group at intermediate risk R12 for in-hospital mortality. Such risk stratification is helpful in predicting the outcome and may R13 assist surgeon in advising patients and their families about the realistic chances of the operation. R14 Follow-up of those patients treated medically in the acute setting evidenced as aortic diameter of R15 less than 4 cm at presentation of ABAD was associated with increased aortic expansion during R16 follow-up while female gender, intramural hematoma and use of calcium channel blockers R17 predicted decreased aortic expansion. In B complicated patients who required open aortic R18 fenestration, we showed that the treated aortic segment do not dilate over time. Biomarkers in R19 aortic dissection are a research field of primary interest, however no standard blood-based test R20 are yet available for the diagnosis and follow-up of these patients. Although not specific, actually R21 only the D-dimer may be useful as serological diagnosis in the acute setting. R22 R23 The role of aortic diameter in acute dissection R24 The role of the initial aortic diameter was investigated with the use of the International Registry R25 of Acute Aortic Dissection (IRAD) database in Chapter 2. In our analysis of 613 dissection R26 patients, 81.6% of the patients with acute type B aortic dissection presented with a descending R27 aortic diameter smaller than 5.5 cm, and the median aortic diameter at which the dissection R28 occurred was 4.1 cm. Current guidelines recommend elective descending thoracic aortic R29 intervention in case of an aortic diameter larger than 5.5 cm or 6.0, in an effort to prevent R30 life-threatening complications, like acute dissection and aortic rupture.1-5 On this basis, elective R31 surgery is not indicated to prevent aortic B dissection. This calls to mind the notion that rupture 12 R32 and dissection presumably have different triggers, and current science is inadequate for risk R33 assessment and prevention. In addition, in the same cohort about one fifth of patients with acute R34 type B aortic dissection presented with a aortic diameter of less than 3.5 cm (Chapter 3). These R35 patients are more frequently females and younger, when compared to dissected patients with R36 larger diameter. In both research projects, the investigated IRAD ABAD patients did not show an R37 increased prevalence of Marfans, which is an important risk factor for aortic dissection, especially R38 R39

149 Chapter 12

R1 in young patients, perhaps because Marfan’s syndrome patients typically suffer from aortic root R2 enlargement and dissection, instead of descending thoracic aortic disease. Both studies clearly R3 showed that initial aortic diameter is not a valuable parameter in the early identification of R4 patients at risk for aortic dissection. R5 R6 Risk stratification for mortality in acute aortic dissection R7 The rise in the incidence of patients with aortic dissection is also correlated with the increasing R8 life expectancy in the general population. In-hospital mortality of acute type A aortic dissection R9 (AAD) is currently 15-30% and is unimproved over time.6-11 Since increasing age has proved R10 to be a independent predictor of in-hospital mortality after cardiovascular interventions, we R11 investigated in Chapter 4 the influence of age on the outcome in AAD patients. We sought at R12 what age the risks of aortic repair outweighed the risk of conservative treated AAD. Our analysis R13 in 936 AAD patients showed that age over 70 was an independent predictor for mortality (38.2% R14 vs 26.0%; P<.0001, odds ratio 1.73). Although the rate of surgical repair decreased with age, it R15 was associated with a significant lower mortality rate than medical treatment until the age of 80 R16 years. Octogenarians with AAD might benefit from ascending aortic open surgical repair and, R17 therefore, surgery should be considered in all patients with AAD. We also analyzed the outcomes R18 of 464 patients with complicated ABAD enrolled in IRAD, stratifying the patients by age and R19 management type (medical, surgical or endovascular treatment) (Chapter 5). The mean age R20 for patients treated with surgery was 60.5 years, compared with 62.1 years in the endovascular R21 group, and 66.8 years in the medical management group (p<0.001). The in-hospital mortality R22 rates for complicated patients ≤ 70 years vs. > 70 years were 7.7% vs. 35.7% for endovascular R23 treatment (p=0.012), 13.5% vs. 32.8% for surgical treatment (p=0.001), and 14.6% vs. 26.6% R24 for medical treatment (p=0.036). Age > 70 years (OR 6.6, 95% CI 1.9-22.8) resulted predictor R25 of mortality. The rate of surgical repair progressively decreased with age, while the rate of medical R26 management significantly increased with age, which may serve as an alternative to invasive R27 treatment in very elderly patients. The use of endovascular methods for complicated ABAD has R28 increased over the years, however, advanced age has a dramatic impact on the mortality of such R29 cohort patients, irrespective of the management type. R30 In Chapter 6 recurrent pain and refractory hypertension were investigated as predictors for in- R31 hospital mortality in ABAD. Patients presenting with recurrent and/or refractory pain, in absence R32 of any hemodynamic and/or ischemic complications, were compared with uncomplicated ABAD. R33 Overall in-hospital mortality was higher in patients with refractory pain and hypertension, R34 17.4% versus 4.0% (P = 0.0003), even more evident when patients were managed medically R35 (35.6 vs 1.5%; P=0.0003). Patients with such complications have been defined at intermediate R36 risk for mortality. Because of endovascular approaches are increasingly becoming the first line R37 treatment for ABAD, both in the acute and chronic setting, this report suggests that this trend R38 may also be beneficial for patients in the intermediate-risk group, who present with recurrent R39 pain and refractory hypertension.12, 13

150 Summary and General discussion

A different clinical subset of dissection patients consists of those presenting with isolated acute R1 abdominal aortic dissection (IAAAD). Because of the incidence is low, we tried to clarify the R2 natural history by identifying clinical characteristics, therapeutic approaches, risk factors, in- R3 hospital outcome, and long-term results of this cohort.14-16 In Chapter 7, 18 IAAD patients were R4 identified within the IRAD database, a total of 1.3% of all enrolled dissected patients. IAAD R5 showed a mean age of 67.7 ± 13.3 years and male predominance ( 67%). Initial clinical symptoms R6 were abdominal pain, mesenteric ischemia or infarction, limb ischemia and hypotension. Two R7 patients presented with an iatrogenic IAAD. Most patients were treated medically with good R8 results and in-hospital mortality was 5.6%. No deaths were reported among patients who R9 underwent surgery or had an endovascular procedure. Overall survival in these patients was R10 93.3% ±12.6% at 1 year and 73.3% ± 27.2% at 5 years. All patients who died during the follow- R11 up period, had an in-hospital medical management. Based on these results, aggressive surgical or R12 endovascular management combined with continued surveillance seems to be mandatory. R13 R14 Evolution of descending aortic dissection (or type b dissection) R15 In medically treated ABAD patients, the long-term outcome is strongly associated with the R16 development of aortic enlargement which can potentially lead to aortic aneurysm and eventual R17 rupture.17, 18 We conducted a study in the IRAD database to identify prognostic parameters in R18 this patient category, which is presented in Chapter 8. A total of 191 ABAD patients treated with R19 medical therapy alone and with available descending aortic diameter measurements at admission, R20 and at 6, 12 and/or 24 months after the acute event were identified. An aortic diameter of R21 less than 4 cm at presentation of ABAD was associated with increased aortic expansion during R22 follow-up (RC 6.3, 95%CI 4.0 to 8.6). On the contrary, female gender (RC -3.8, 95%CI -6.1 to R23 -1.4), intramural hematoma (RC -3.8, 95%CI -6.5 to -1.1), and use of calcium channel blockers R24 (RC -3.8, 95%CI -6.2 to -1.3) were associated with decreased aortic expansion. A decreased R25 aortic growth rate in patients treated with calcium channel blockers raise the possibility that the R26 use of calcium channel blockers is beneficial in all ABAD patients. R27 Since the introduction of endovascular techniques in the treatment of acute aortic dissection, R28 it has gained popularity due to its minimal invasive character.19, 20 However, not all patients R29 are considered suitable for endovascular intervention. In Chapter 9, we studied the long-term R30 outcome of open surgical fenestration performed in patients when percutaneous techniques were R31 still not available, contra-indicated or failed. A total of 18 patients with complicated ABAD 12 R32 were treated with surgical fenestration, of which 10 patients underwent a suprarenal and 8 R33 infrarenal fenestration. In-hospital mortality was 22% and during follow-up no surgical related R34 complications were reported. In particular, the aortic segment submitted to the aortotomy R35 did not dilate over time. These results showed that in case of contraindications or failure of R36 endovascular management, surgical fenestrations can serve as a durable treatment option in R37 complicated ABAD. R38 R39

151 Chapter 12

R1 biomarkers in aortic dissection R2 For the identification, risk stratification and therapeutic monitoring in aortic dissection, R3 biomarkers might play an important role. Prior studies in Marfan patients showed that aortic R4 remodeling was associated with increased circulating transforming growth factor (TGF)- beta.21, R5 22 In Chapter 10 we evaluated the TGF- beta levels in 28 non-Marfan patients with acute aortic R6 dissection. TGF-beta levels were increased approximately 5-fold in acute dissected patients R7 compared to normal controls, and twice as much in type A compared to type B patients. These R8 finding emphasized the potential benefit of TGF-beta levels as a biomarker. It remains unknown R9 if circulating TGF-beta levels can be used to monitor remodeling dynamics in acute aortic disease R10 and whether they will be responsive to pharmacological treatment. Further studies should be R11 awaited, to fully understand the potential of TGF-beta levels as biomarker in the diagnostic and R12 surveillance of these patients. R13 In Chapter 11, we updated the role of routine laboratory test in the assessment of acute and R14 chronic aortic dissections of the thoracic aorta. Multiple blood markers were identified as potential R15 predictors of aortic dissection and aortic growth rate: D-Dimer, Smooth Muscle Myosin Heavy R16 Chain Protein, Calponin, Serum Markers of Elastin and Collagen Metabolism, Cytokines and R17 Gene Expression Signature in Peripheral Blood, Acute Phase Reactant, Lipoprotein, Activators R18 and Inhibitors of Plasminogen, Fibrinogen,and Hemostatic Markers, Endothelin and Hepatocyte R19 Growth Factor, Homocysteine, TGF-beta and Serum Heart-Type Fatty Acid–Binding Protein. R20 These serologic biomarkers are potentially predictive of the clinical disease, however, actually, they R21 cannot be utilized to diagnose and follow aortic dissections, except the D-dimer in specifying R22 setting. It seems useful to rule out patients, if applied within the first 24 hours after symptom R23 onset using the cut-off level < 500 ng/mL, similarly for ruling out pulmonary embolism, and R24 may be reliable to rule in aortic dissection when > 1500 ng/mL if within 6 hours of symptoms’ R25 onset.23 Although a standardized blood-based test or tests, capable of detecting individuals at R26 risk for aortic aneurysm and dissection disease is not still available, our current knowledge is R27 expanding at a rapid rate and the future is very promising. R28 R29 future perspectives R30 As Charcot stated in the 19th century, ”disease is very old and nothing about it has changed. It is R31 we who change, as we learn to recognize what was formerly imperceptible.” 24 With this thesis we R32 tried to extend insights into acute aortic dissection, providing additional information both for a R33 better management of these patients and to generate innovative research hypothesis. R34 Although acute type B aortic dissection seems to occur at considerably smaller size than the R35 current threshold for elective descending thoracic aortic repair, lowering the cut-off for aortic R36 intervention is certainly not the solution for this problem. It is likely that many affected individuals R37 in the world have a descending aorta larger than 4 cm in diameter. Yet, we believe that only a R38 small fraction will ultimately dissect. The medical, emotional and financial risk of intervening on R39

152 Summary and General discussion

this population would be substantial and unpredictable. Before eventually genetic, proteomic, R1 and/or imaging profiles help to identify patients at high risk, a conservative approach is justified. R2 In order to prevent acute type B aortic dissection, further natural history studies are needed, as R3 well as research on genetic predisposition for thoracic aortic disease. These may reveal other risk R4 factors for aortic dissection besides increasing aortic diameter, resulting in better medical and/or R5 interventional criteria for prophylactic thoracic aortic repair. R6 New operative techniques have raised important prospective for better managing and results R7 in dissected type A and B patients with advanced age, focusing in particular on safer cerebral R8 protection and wide use of TEVAR. However, for surviving patients, it is important to consider R9 subsequent survival and quality of life. Recently Hata et al. underlined the impact of postoperative R10 complications in octogenarians with AAD on their families, such as stroke, depression, dementia R11 and/or bedridden status. 25 Several families who had to take responsibility of the patients that R12 survived the operation, voiced their complaints. Further study of this aspect of care and outcomes R13 is sorely needed. R14 In ABAD patients at intermediate risk of mortality, as those presenting with refractory pain R15 and/or hypertension, early endovascular intervention could improve in-hospital and long-term R16 results. R17 The predictors of future aortic enlargement in uncomplicated ABAD patients who had initial R18 medical treatment is actually one the most debated issue in aortic dissections. Our observations R19 showed novel data in the literature and, obviously, have to be validated with further investigations. R20 However, our research highlighted as both imaging signs and role of some medicaments in R21 descending aortic diseases are still unexplored variables as predictors of aortic growth, and need R22 to be addressed. R23 Until now, routine laboratory tests have played only a minor role in the assessment of acute R24 and chronic diseases of the aorta. However, a number of new diagnostic screening tools have R25 been developed, in order to evaluate the use of biochemical markers in the peripheral blood as R26 initial diagnostic screening.26, 27 Ideally, a biochemical diagnostic test would be a chemical marker R27 reflective of the pathogenic activity, readily available, inexpensive, and non-invasive. A biological R28 test could be helpful in the initial patient triage at the emergency department of referral centers, R29 helping physicians determine whether aortic dissection should be considered as a possible cause R30 of thoracic pain. Other biomarkers might assist in routine office screening of well patients, and R31 for long term follow-up evaluation of disease progression in patients with aneurysm, offering the 12 R32 potential to predict adverse aortic events, such as enlargement, rupture, and dissection. R33 R34 R35 R36 R37 R38 R39

153 Chapter 12

R1 REfERENCES R2 R3 (1) Coady MA, Rizzo JA, Hammond GL, Kopf GS, Elefteriades JA. Surgical intervention criteria R4 for thoracic aortic aneurysms: a study of growth rates and complications. Ann Thorac Surg 1999 R5 Jun;67(6):1922-6. R6 (2) Elefteriades JA. Natural history of thoracic aortic aneurysms: indications for surgery, and surgical R7 versus nonsurgical risks. Ann Thorac Surg 2002 Nov;74(5):S1877-S1880. R8 (3) Davies RR, Goldstein LJ, Coady MA, Tittle SL, Rizzo JA, Kopf GS, et al. Yearly rupture or R9 dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg R10 2002 Jan;73(1):17-27. R11 (4) Svensson LG, Kouchoukos NT, Miller DC, Bavaria JE, Coselli JS, Curi MA, et al. Expert R12 consensus document on the treatment of descending thoracic aortic disease using endovascular R13 stent-grafts. Ann Thorac Surg 2008 Jan;85(1 Suppl):S1-41. R14 (5) Lobato AC, Puech-Leao P. Predictive factors for rupture of thoracoabdominal aortic aneurysm. J R15 Vasc Surg 1998 Mar;27(3):446-53. R16 (6) Bachet J, Goudot B, Dreyfus GD, Brodaty D, Dubois C, Delentdecker P, et al. Surgery for R17 acute type A aortic dissection: the Hopital Foch experience (1977-1998). Ann Thorac Surg 1999 R18 Jun;67(6):2006-9. R19 (7) David TE, Armstrong S, Ivanov J, Barnard S. Surgery for acute type A aortic dissection. Ann R20 Thorac Surg 1999 Jun;67(6):1999-2001. R21 (8) Ehrlich MP, Ergin MA, McCullough JN, Lansman SL, Galla JD, Bodian CA, et al. Results of R22 immediate surgical treatment of all acute type A dissections. Circulation 2000 Nov 7;102(19 R23 Suppl 3):III248-III252. R24 (9) Narayan P, Rogers CA, Davies I, Angelini GD, Bryan AJ. Type A aortic dissection: has surgical R25 outcome improved with time? J Thorac Cardiovasc Surg 2008 Nov;136(5):1172-7. R26 (10) Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection: R27 increasing prevalence and improved outcomes reported in a nationwide population-based study R28 of more than 14,000 cases from 1987 to 2002. Circulation 2006 Dec 12;114(24):2611-8. R29 (11) Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Mehta RH, et al. Contemporary R30 results of surgery in acute type A aortic dissection: The International Registry of Acute Aortic R31 Dissection experience. J Thorac Cardiovasc Surg 2005 Jan;129(1):112-22. R32 (12) Fattori R, Tsai TT, Myrmel T, Evangelista A, Cooper JV, Trimarchi S, et al. Complicated acute R33 type B dissection: is surgery still the best option?: a report from the International Registry of R34 Acute Aortic Dissection. JACC Cardiovasc Interv 2008 Aug;1(4):395-402. R35 (13) Kische S, Ehrlich MP, Nienaber CA, Rousseau H, Heijmen R, Piquet P, et al. Endovascular R36 treatment of acute and chronic aortic dissection: midterm results from the Talent Thoracic R37 Retrospective Registry. J Thorac Cardiovasc Surg 2009 Jul;138(1):115-24. R38 R39

154 Summary and General discussion

(14) Farber A, Wagner WH, Cossman DV, Cohen JL, Walsh DB, Fillinger MF, et al. Isolated R1 dissection of the abdominal aorta: clinical presentation and therapeutic options. J Vasc Surg R2 2002 Aug;36(2):205-10. R3 (15) Farber A, Lauterbach SR, Wagner WH, Cossman DV, Long B, Cohen JL, et al. Spontaneous R4 infrarenal abdominal aortic dissection presenting as claudication: case report and review of the R5 literature. Ann Vasc Surg 2004 Jan;18(1):4-10. R6 (16) Hirst AE, Jr., JOHNS VJ, Jr., KIME SW, Jr. Dissecting aneurysm of the aorta: a review of 505 R7 cases. Medicine (Baltimore) 1958 Sep;37(3):217-79. R8 (17) Sueyoshi E, Sakamoto I, Hayashi K, Yamaguchi T, Imada T. Growth rate of aortic diameter in R9 patients with type B aortic dissection during the chronic phase. Circulation 2004 Sep 14;110(11 R10 Suppl 1):II256-II261. R11 (18) Tsai TT, Evangelista A, Nienaber CA, Myrmel T, Meinhardt G, Cooper JV, et al. Partial R12 thrombosis of the false lumen in patients with acute type B aortic dissection. N Engl J Med 2007 R13 Jul 26;357(4):349-59. R14 (19) Dake MD, Kato N, Mitchell RS, Semba CP, Razavi MK, Shimono T, et al. Endovascular R15 stent-graft placement for the treatment of acute aortic dissection. N Engl J Med 1999 May R16 20;340(20):1546-52. R17 (20) Nienaber CA, Fattori R, Lund G, Dieckmann C, Wolf W, von KY, et al. Nonsurgical R18 reconstruction of thoracic aortic dissection by stent-graft placement. N Engl J Med 1999 May R19 20;340(20):1539-45. R20 (21) Ahimastos AA, Aggarwal A, D’Orsa KM, Formosa MF, White AJ, Savarirayan R, et al. Effect of R21 perindopril on large artery stiffness and aortic root diameter in patients with Marfan syndrome: R22 a randomized controlled trial. JAMA 2007 Oct 3;298(13):1539-47. R23 (22) Matt P, Schoenhoff F, Habashi J, Holm T, Van EC, Loch D, et al. Circulating transforming R24 growth factor-beta in Marfan syndrome. Circulation 2009 Aug 11;120(6):526-32. R25 (23) Suzuki T, Distante A, Zizza A, Trimarchi S, Villani M, Salerno Uriarte JA, et al. Diagnosis of acute R26 aortic dissection by D-dimer: the International Registry of Acute Aortic Dissection Substudy on R27 Biomarkers (IRAD-Bio) experience. Circulation 2009 May 26;119(20):2702-7. R28 (24) Jean-Martin Charcot. De l’Expectation en medecine. These agregation medecine Paris. Paris: R29 Germer-Bailliere; 1857. R30 (25) Hata M, Sezai A, Niino T, Yoda M, Unosawa S, Furukawa N, et al. Should emergency surgical R31 intervention be performed for an octogenarian with type A acute aortic dissection? J Thorac 12 R32 Cardiovasc Surg 2008 May;135(5):1042-6. R33 (26) Hovsepian DM, Ziporin SJ, Sakurai MK, Lee JK, Curci JA, Thompson RW. Elevated plasma R34 levels of matrix metalloproteinase-9 in patients with abdominal aortic aneurysms: a circulating R35 marker of degenerative aneurysm disease. J Vasc Interv Radiol 2000 Nov;11(10):1345-52. R36 (27) Suzuki T, Katoh H, Watanabe M, Kurabayashi M, Hiramori K, Hori S, et al. Novel biochemical R37 diagnostic method for aortic dissection. Results of a prospective study using an immunoassay of R38 smooth muscle myosin heavy chain. Circulation 1996 Mar 15;93(6):1244-9. R39

155

Chapter 13

Dutch Summary – Nederlandse Samenvatting Chapter 13

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

158 Summary in Dutch – Nederlandse Samenvatting

Acute aortadissectie is een levensbedreigende aandoening, die gepaard gaat met een R1 hoge morbiditeit en mortaliteit. De laatste jaren hebben nieuwe diagnostische- en R2 behandelingsmethoden de resultaten verbeterd bij deze patienten. Daarnaast hebben we nieuwe R3 inzichten verkregen ten aanzien van de preventie van acute aortadissectie, hetgeen tevens kan R4 dienen als basis voor verder wetenschappelijk onderzoek. Onze studies hebben aangetoond dat de R5 aortadiameter geen onafhankelijke voorspeller is van acute aortadissectie, aangezien deze frequent R6 voorkomt bij patiënten met een normale aortadiameter, zonder predisponerende factoren zoals R7 bindweefselaandoeningen. Het beloop bij patienten met een acute dissectie die een open en/ R8 of endovasculaire behandeling behoeven, zijn sterk gerelateerd aan de preoperatieve klinische R9 conditie van de patient. Patiënten met een acute type B aortadissectie (ABAD) die zich presenteren R10 met therapieresistente pijn en hypertensie, maar zonder hemodynamische en/of ischemische R11 complicaties, vormen een groep met een intermediair risicoprofiel voor ziekenhuismortaliteit. R12 Een dergelijke risico-inschatting is bruikbaar bij het voorspellen van de uitkomst en kan de R13 chirurg helpen bij het adviseren van de patiënt en diens families ten aanzien van een operatie. R14 Uit follow-up van patiënten die gedurende de acute setting medicamenteus werden behandeld, R15 blijkt dat een aortadiameter van minder dan 4 cm bij presentatie is gerelateerd aan een R16 verhoogde aorta-expansie, terwijl het vrouwelijk geslacht, intramuraal hematoom en het gebruik R17 van calciumantagonisten minder aorta-expansie vertonen. In patiënten met gecompliceerde R18 ABAD, die middels een open chirurgische fenestratie van de aorta werden behandeld, hebben R19 we aangetoond dat het behandelde aortasegment geen groei vertoont gedurende follow-up. R20 Biomarkers bij aortadissectie vormen een belangrijk aandachtsgebied voor het wetenschappelijk R21 onderzoek, maar tot op heden zijn er geen standaard bloedtestsen beschikbaar voor de diagnose R22 en follow-up van deze patiënten. Hoewel niet specifiek, blijkt D-dimeer nuttig te zijn als een R23 serologisch diagnosticum voor aortadissectie gedurende de acute fase. R24 R25 De rol van de aortadiameter in acute dissectie R26 De rol van de initiële diameter van de aorta werd onderzocht in Hoofdstuk 2 met behulp van R27 het Internationale Register van Acute Aorta Dissectie (IRAD) database. In onze analyse van 613 R28 dissectie patiënten, bleek 81.6% van de patiënten met ABAD een aorta -descendens diameter R29 te hebben kleiner dan 5.5 cm en de gemiddelde aorta diameter waarbij seen dissectie optrad R30 was 4.1 cm. De huidige richtlijnen adviseren een electieve interventie voor de thoracale aorta R31 descenden bij een diameter groter dan 5.5 cm of 6.0 cm, ter preventie van levensbedreigende R32 complicaties zoals acute dissectie en aorta ruptuur. 1-5 Op basis van dit onderzoek blijkt electieve R33 chirurgie niet bij te dragen aan de preventie van ABAD. Tevens blijkt ook dat een aortaruptuur R34 en aortadissectie vermoedelijk verschillende triggers hebben en de huidige kennis ontoereikend 13 R35 is voor een correcte risco-analyse en preventie. Bovendien bleek uit hetzelfde cohort dat ongeveer R36 een vijfde van patiënten met acute type B aorta zich presenteerde met een aorta diameter van R37 minder dan 3.5cm (Hoofdstuk 3). Deze patiënten zijn vaker van het vrouwelijk geslacht en R38 R39

159 Chapter 13

R1 jonger, in vergelijking met patiënten met een grotere diameter. In beide onderzoeken, werd R2 in de IRAD ABAD patiënten geen verhoogde prevalentie van Marfan gezien, hetgeen een R3 belangrijke risicofactor is voor aortadissectie. Mogelijk kan dit verklaard worden door het feit R4 dat patiënten met het Marfan syndroom zich typisch presenteren met een vergroting en dissectie R5 van de aortawortel in plaats van de descenderende thoracala aorta. Beide studies hebben duidelijk R6 aangetoond dat de initiële aorta diameter geen waardevolle parameter is voor de identificatie van R7 patiënten met een verhoogd risico op aortadissectie. R8 R9 Risicostratificatie voor de sterfte in acute aorta dissectie R10 De stijging van de incidentie van patiënten met een aortadissectie is onder andere gecorreleerd R11 met de toegenomen levensverwachting van de bevolking. In het ziekenhuis bedraagt de mortaliteit R12 van acute type A aortadissectie (AAD) op dit moment 15-30% en is stabiel gebleven in de loop R13 der jaren. 6-11 Aangezien een toegenomen leeftijd een onafhankelijke voorspeller blijkt te zijn voor R14 mortaliteit in het ziekenhuis na cardiovasculaire interventies, hebben we in hoofdstuk 4 de invloed R15 van leeftijd op de uitkomst van AAD patiënten onderzocht. We hebben gekeken bij welke leeftijd R16 de risico’s van de aorta reparatie opwegen tegen het risico van een conservatieve behandeling R17 van AAD. Analyse van 936 AAD patiënten toonde aan dat een leeftijd boven de 70 jaar een R18 onafhankelijke voorspeller is voor sterfte ( 38.2% vs 26.0%, P<0.0001, odds ratio 1.73). Hoewel R19 het percentage chirurgische ingrepen afneemt met toename van de leeftijd, bleek het geassocieerd R20 te zijn met een significant lager sterftecijfer dan medicamenteuze behandeling tot de leeftijd van R21 80 jaar. Tachtigplussers met AAD kunnen baat heben van open chirurgische reconstuctie van R22 de aorta ascendens en derhalve dient een operatie overwogen te worden in alle patiënten met R23 AAD. Daarnaast analyseerden we de uitkomsten van 464 patiënten met gecompliceerde ABAD, R24 die waren geincludeerd in IRAD, waarbij patienten werden gestratificeerd naar leeftijd en type R25 behandeling ( medicamenteuze, chirurgische, of endovasculaire behandeling) (Hoofdstuk 5). R26 De gemiddelde leeftijd van patiënten die chirurgisch werden behandeld was 60.5 jaar, vergeleken R27 met 62.1 jaar in de endovasculaire groep, en 66.8 jaar in de medicamenteuze behandelgroep R28 (P<0.001). De sterftecijfers in het ziekenhuis voor gecompliceerde patiënten ≤ 70 jaar vs. > 70 R29 jaar, waren 7.7% vs. 35.7% voor de endovasculaire behandeling (P=0.012), 13.5% vs. 32.8% voor R30 chirurgische behandeling (P=0.001), en 14.6% vs. 26.6% voor de medicamenteuze behandeling R31 (P=0.036). Leeftijd >70 jaar (OR 6.6, 95% CL 1.9-22.8) bleek een voorspeller van mortaliteit. R32 Het aantal chirurgisch ingrepen neemt geleidelijk af terwijl het aantal van de medicamenteuze R33 behandelingen significant stijgt met de leeftijd, omdat het gebruikt kan worden als alternatief R34 voor een invasieve behandeling bij bejaarde patiënten. Het gebruik van endovasculaire methoden R35 voor gecompliceerde ABAD is toegenomen in de loop der jaren, maar ondanks deze ontwikkeling R36 blijft leeftijd van grote invloed op het sterftecijfer van dit cohort patiënten, ongeacht het type R37 behandeling. R38 R39

160 Summary in Dutch – Nederlandse Samenvatting

In Hoofdstuk 6 hebben we persisterende pijn en therapieresistente hypertensie onderzocht R1 als voorspellers voor mortaliteit in ABAD patiente. Patiënten met terugkerende en/of R2 therapieresistente pijn, zonder hemodynamische en/of ischemische complicaties, werden R3 vergeleken met ongecompliceerde ABAD. Over het algemeen was het sterftecijfer in het R4 ziekenhuis hoger bij patiënten met therapieresistente pijn en hypertensie, 17.4% versus 4.0% R5 (P=0.0003), en dit verschil was nog evidenter in medicamenteus behandelde patiënten.(35.6 R6 vs. 1.5%; P=0.0003). Patiënten met dergelijke complicaties hebben derhalve een intermediair R7 risicoprofiel voor mortaliteit. Aangezien de endovasculaire behandeling van ABAD patienten R8 zich steeds meer ontwikkelt als behandeling van eerste keus, zowel in de acute als chronische R9 setting, toont dit onderoek aan dat deze trend gunstig zou kunnen kan zijn voor patiënten met R10 een intermediair risicoprofiel, die zich presenteren met recidiverende pijn en therapieresistente R11 hypertensie. 12, 13 R12 Patiënten die zich presenteren met geïsoleerde acute abdominale aortadissectie (IAAD) vormen R13 een aparte klinische subgroep. Vanwege de lage incidentie, hebben we getracht om het natuurlijke R14 ziekteverloop te bestuderen door te kijken naar de klinische kenmerken, therapeutische R15 benaderingen, risicofactoren, resultaten in het ziekenhuis, en de lange termijn resulaten van R16 deze patienten. 14-16 In Hoofdstuk 7 worden 18 IAAD patiënten beschreven, een totaal van R17 1.3% van alle geregistreerde dissectie patiënten in de IRAD database. IAAD presenteert zich op R18 een gemiddelde leeftijd van 67.7 ± 13.3 jaar en voornamelijk in mannen (67%). De klinische R19 symptomen bij presentatie betroffen abdominale pijn, mesenteriale ischemie of infarct, perifere R20 ischemie en hypotensie. Twee patiënten presenteerden zich met een iatrogene IAAD. De meeste R21 patiënten werden succesvol medicamenteus behandeld met een ziekenhuis-mortaliteit van 5.6%. R22 In patiënten die een operatie of een endovasculaire procedure ondergingen was de mortaliteit R23 0%. Algehele overleving bij deze patiënten was 93.3% ± 12.6% na 1 jaar en 73.3% ± 27.2% na R24 5 jaar. Alle patiënten die stierven gedurende de follow-up periode waren initieel medicamenteus R25 behandeld. Op basis van deze resultaten lijkt chirurgische of endovasculaire behandeling R26 geindiceerd in IAAD patienten, gecombineerd met langdurige controle. R27 R28 Evolutie van de aorta descendens dissectie ( of type b dissectie) R29 In medicamenteus behandelde ABAD patiënten zijn de lange termijn resultaten sterk gerelateerd R30 aan de toename van de aorta diameter, hetgeen kan leiden tot een aneurysma van de aorta R31 en uiteindelijke een ruptuur. 17, 18 Met behulp van de IRAD database hebben we een studie R32 uitgevoerd om prognostische parameters te identificeren voor deze categorie patiënten, die wordt R33 beschreven in Hoofdstuk 8. Een totaal van 191 ABAD medicamenteus behandelde patiënten R34 met aorta diameter metingen beschikbaar bij opname, en na 6, 12 en/of 24 maanden na de 13 R35 acute fase werden geincludeerd. Een aorta diameter van kleiner dan 4 cm bij presentatie van R36 ABAD bleek geassocieerd te zijn met een toegenomen groei van de aorta tijdens de follow- R37 up (RC 6.3, 95%CL 4.0 tot 8.6). Het vrouwelijke geslacht (RC -3.8, 95%CL -6.1 tot -1.4), R38 R39

161 Chapter 13

R1 intramuraal hematoom (RC -3.8, 95%CL -6.5 tot -1.1), en het gebruik van calciumantagonisten R2 (RC -3.8, 95%CL -6.2 tot -1.3) vertoonde daarentegen minder groei. De verminderde groei R3 van de aorta bij patiënten behandeld met calciumantagonisten suggereert dat het gebruik van R4 calciumantagonisten gunstig kan zijn voor alle ABAD patiënten. R5 Sinds de introductie van de endovasculaire behandeling van acute aortadissectie wordt, deze R6 vanwege het minimaal invasieve karakter, in toenemende mate gebruikt. 19, 20 Echter niet alle R7 patiënten komen in aanmerking voor endovasculaire interventie. In Hoofdstuk 9 bestudeerden we R8 de lange termijn resultaten van open chirurgische fenestratie bij patiënten die niet in aanmerking R9 kwamen voor een percutane behandeling. In totaal werden 18 patiënten behandeld middels een R10 chirurgische fenestratie, waarvan 10 patiënten een suprarenale fenestratie ondergingen en 8 een R11 infrarenale fenestratie. In het ziekenhuis betrof de mortaliteit 22% en gedurende de follow-up R12 werden er geen chirurgische complicaties waargenomen. Opvallend was dat het gefenestreerde R13 aortasegment geen aneurysmatisch groei vertoonde gedurende de follow-up. Deze resultaten R14 bewijzen dat in het geval van contra-indicaties of het falen van endovasculaire behandelingen, R15 chirurgische fenestratie een geschikte behandelingsoptie is in gecompliceerde ABAD. R16 R17 biomarkers bij aorta dissectie R18 Voor de identificatie, risicostratificatie en controle van patienten met aortadissectie spelen R19 biomarkers mogelijk een belangrijke rol. Eerdere studies in Marfan patiënten toonde aan dat R20 aorta remodelling is geassocieerd met een stijging van het circulerende transforming growth factor R21 (TGF)- beta. 21, 22 In Hoofdstuk 10 evalueerden we de TGF –beta niveau in 28 niet-Marfan R22 patiënten met een acute aorta dissectie. TGF –beta levels waren ongeveer vijfvoudig verhoogd R23 bij acute dissectie patiënten in vergelijking met normale controles, en tweemaal zo hoog in type R24 A ten opzichte van type B patiënten. Deze bevinding benadrukken de mogelijke toepasbaarheid R25 van TGF –beta niveau als een biomarker. Het blijft echter onduidelijk of circulerend TGF –beta R26 kan worden gebruikt ter controle van het aorta remodelling proces en wat de invloed is van R27 een farmacologische behandeling op deze biomarker. Aanvullende studies zijn noodzakelijk om R28 de toepasbaarheid van TGF –beta als biomarker gedurende diagnostiek en follow-up verder te R29 doorgronden. R30 In Hoofdstuk 11, hebben we het gebruik van routine laboratoriumtests onderzocht voor de R31 analyse van acute en chronische aortadissecties van de thoracale aorta. Verschillende bloedtesten R32 spelen een rol in het voorspellen van aortadissectie en aortagroei: D-Dimeer, Smooth Muscle R33 Myosin Heavy Chain Protein, Calponin, Serum Markers van Elastine en Collageen Metabolisme, R34 Cytokines and Gene Expression Signature in Peripheral Blood, Acute Phase Reactant, R35 Lipoprotein, Activatoren en Inhibitoren van Plasminogeen, Fibrinogeen, Hemostatic Markers, R36 Endotheline en Hepatocyte Growth Factor, Homocysteine, TGF-beta en Serum Hart-Type Fatty R37 Acid-Binding Protein. Hoewel deze serologische biomarkers een mogelijke voorspeller zijn het R38 voor het ontwikkelen van het ziektebeeld, kunnen ze, met uitzondering van D-Dimeer, niet R39

162 Summary in Dutch – Nederlandse Samenvatting

gebruikt worden voor het diagnosticeren of de follow-up van aortadissecties. Indien het bepaald R1 wordt binnen 24 uur na het optreden van de eerste symptomen en met een de cutt-off niveau R2 van <500ng/ml lijkt D-dimeer bruikbaar om aortadissectie in patienten uit te sluiten, net zoals R3 dat geldt voor het uitsluiten van longembolieën. Daarnaast kan de diagnose aortadissectie R4 mogelijk worden bevestigd bij een D-Dimeer > 1500 ng/ml binnen 6 uur na het optreden van de R5 eerste symptomen. 23 Hoewel gestandariseerde bloedtesten voor het identificeren van patienten R6 met een verhoogd risico op aorta aneurysma en aortadissectie nog niet beschikbaar zijn, zijn de R7 ontwikkelingen van de huidige kennis veelbelovend. R8 R9 Toekomstperspectieven R10 Zoals Charcot reeds in de 19e eeuw zei, “Ziekte is heel oud en daar is niets aan veranderd. R11 Het zijn wij die veranderen, wanneer we leren te herkennen wat voorheen niet waarneembaar R12 was”. 24 Met dit proefschrift hebben we geprobeerd om een beter inzicht te verkrijgen in acute R13 aorta dissectie om zo een betere behandeling voor deze patiënten te bewerkstelligen en voor het R14 ontwikkelen van nieuwe onderzoekshypotheses. R15 Hoewel ABAD blijkt voor te komen in patienten met een aanzienlijk kleinere aorta diameter R16 dan de huidige indicatiestelling voor electieve thoracale aorta reconstructie, is verlaging van de R17 cut-off niet de oplossing van het probleem. Het is zeer aannemelijk dat er in een populatie R18 velen zijn met een diameter van de aorta descendens groter dan 4 cm, maar slechts een klein R19 deel zal uiteindelijk een dissectie ontwikkelen. De medische, emotionele en financiële risico’s R20 gerelateerd aan medisch ingrijpen in deze populatie zijn aanzienlijk en complex. Totdat R21 genetische, preteoom en/of radiologische profielen beschikbaar zijn om een hoog risico patient R22 te identificeren, is een conservatieve behandeling gerechtvaardigd. Ter preventie van ABAD R23 is aanvullend wetenschappelijk onderzoek, gericht op het het natuurlijke ziektebeloop en de R24 genetische aanleg voor ziekten van de thoracale aorta, noodzakelijk. Op deze wijze kunnen R25 andere risicofactoren dan een vergrote aorta diameter worden gediagnosticeerd, hetgeen kan R26 resulteren in een betere indicatiestelling voor medicamenteuze behandeling en/of profylactische R27 thoracale aorta- reconstructie. R28 Nieuwe operatietechnieken hebben geleid tot een betere behandeling met verbeterde uitkomsten R29 in type A en B dissectie patiënten van oudere leeftijd, voornamelijk door verbeterde cerebrale R30 bescherming en het toegenomen gebruik van TEVAR. Echter voor patiënten die het overleven R31 is het van belang om de levensduur en de kwaliteit van het leven in acht te nemen. Onlangs R32 benadrukte Hata et al. de impact van postoperatieve complicaties bij tachtigplussers met AAD, R33 zoals een beroerte, depressie, dementie, en/of bedlegerigheid op hun familie, 25 Verschillende R34 families die de verantwoordelijkheid moesten dragen voor de patiënten die de operatie overleefden 13 R35 uitten hun bezwaren. Nader onderzoek naar dit aspect van de zorg is noodzakelijk. R36 InABAD patiënten met een intermediair risico op overlijden, gedefinieerd als patiënten die zich R37 presenteren met therapieresistente pijn en/of hypertensie, kan vroege endovasculaire interventie R38 mogelijk de korte en lange termijn resultaten verbeteren. R39

163 Chapter 13

R1 Voorspellers van aortagroei in patienten met ongecompliceerde, medicamenteus behandeling R2 ABAD is een van de meest besproken onderwerpen op het gebied van aortadissectie. Ons R3 onderzoek toonde enkele nieuwe bevindingen welke, uiteraard, moeten worden gevalideerd met R4 aanvullende onderzoeken. Desalniettemin benadrukte ons onderzoek dat zowel radiologische R5 kenmerken als sommige geneesmiddelen invloed hebben op de groei van de aorta en verder R6 onderzocht dienen te worden. R7 Tot op heden hebben routine laboratoriumtesten slechts een bescheiden rol gespeeld bij de R8 evaluatie van acute en chronische aandoeningen van de aorta. Een aantal nieuwe diagnostische R9 screenings technieken zijn ontwikkeld, waarbij biochemische markers in het bloed gebruikt R10 worden als voor diagnostische screening. 26, 27 Idealiter, zou een biochemische diagnostische test R11 een chemische marker zijn die correspondeert met de pathogene activiteit, direct beschikbaar R12 is, goedkoop en niet-invasief. Een dergelijke diagnostische test zou bruikbaar zijn voor de triage R13 van patiënten op de spoedeisende hulp van een verwijzingscentra, zodat de arts eenvoudig kan R14 bepalen of een aorta dissectie een verklaring is voor thoracale pijn. Andere biomarkers zouden R15 gebruikt kunnen worden bij screening in de gezonde populatie of gedurende de follow-up van R16 patiënten met een aneurysma ter evaluatie van de progressie van de ziekte. Op deze wijze zouden R17 aortagerelateerde complicaties kunnen worden voorspeld, zoals aneurysmatische groei, ruptuur R18 en dissectie. R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

164 Summary in Dutch – Nederlandse Samenvatting

REfERENCE lIST R1 R2 (1) Coady MA, Rizzo JA, Hammond GL, Kopf GS, Elefteriades JA. Surgical intervention criteria R3 for thoracic aortic aneurysms: a study of growth rates and complications. Ann Thorac Surg 1999 R4 Jun;67(6):1922-6. R5 (2) Elefteriades JA. Natural history of thoracic aortic aneurysms: indications for surgery, and surgical R6 versus nonsurgical risks. Ann Thorac Surg 2002 Nov;74(5):S1877-S1880. R7 (3) Davies RR, Goldstein LJ, Coady MA, Tittle SL, Rizzo JA, Kopf GS, et al. Yearly rupture or R8 dissection rates for thoracic aortic aneurysms: simple prediction based on size. Ann Thorac Surg R9 2002 Jan;73(1):17-27. R10 (4) Svensson LG, Kouchoukos NT, Miller DC, Bavaria JE, Coselli JS, Curi MA, et al. Expert R11 consensus document on the treatment of descending thoracic aortic disease using endovascular R12 stent-grafts. Ann Thorac Surg 2008 Jan;85(1 Suppl):S1-41. R13 (5) Lobato AC, Puech-Leao P. Predictive factors for rupture of thoracoabdominal aortic aneurysm. J R14 Vasc Surg 1998 Mar;27(3):446-53. R15 (6) Bachet J, Goudot B, Dreyfus GD, Brodaty D, Dubois C, Delentdecker P, et al. Surgery for R16 acute type A aortic dissection: the Hopital Foch experience (1977-1998). Ann Thorac Surg 1999 R17 Jun;67(6):2006-9. R18 (7) David TE, Armstrong S, Ivanov J, Barnard S. Surgery for acute type A aortic dissection. Ann R19 Thorac Surg 1999 Jun;67(6):1999-2001. R20 (8) Ehrlich MP, Ergin MA, McCullough JN, Lansman SL, Galla JD, Bodian CA, et al. Results of R21 immediate surgical treatment of all acute type A dissections. Circulation 2000 Nov 7;102(19 R22 Suppl 3):III248-III252. R23 (9) Narayan P, Rogers CA, Davies I, Angelini GD, Bryan AJ. Type A aortic dissection: has surgical R24 outcome improved with time? J Thorac Cardiovasc Surg 2008 Nov;136(5):1172-7. R25 (10) Olsson C, Thelin S, Stahle E, Ekbom A, Granath F. Thoracic aortic aneurysm and dissection: R26 increasing prevalence and improved outcomes reported in a nationwide population-based study R27 of more than 14,000 cases from 1987 to 2002. Circulation 2006 Dec 12;114(24):2611-8. R28 (11) Trimarchi S, Nienaber CA, Rampoldi V, Myrmel T, Suzuki T, Mehta RH, et al. Contemporary R29 results of surgery in acute type A aortic dissection: The International Registry of Acute Aortic R30 Dissection experience. J Thorac Cardiovasc Surg 2005 Jan;129(1):112-22. R31 (12) Fattori R, Tsai TT, Myrmel T, Evangelista A, Cooper JV, Trimarchi S, et al. Complicated acute R32 type B dissection: is surgery still the best option?: a report from the International Registry of R33 Acute Aortic Dissection. JACC Cardiovasc Interv 2008 Aug;1(4):395-402. R34 (13) Kische S, Ehrlich MP, Nienaber CA, Rousseau H, Heijmen R, Piquet P, et al. Endovascular 13 R35 treatment of acute and chronic aortic dissection: midterm results from the Talent Thoracic R36 Retrospective Registry. J Thorac Cardiovasc Surg 2009 Jul;138(1):115-24. R37 R38 R39

165 Chapter 13

R1 (14) Farber A, Wagner WH, Cossman DV, Cohen JL, Walsh DB, Fillinger MF, et al. Isolated R2 dissection of the abdominal aorta: clinical presentation and therapeutic options. J Vasc Surg R3 2002 Aug;36(2):205-10. R4 (15) Farber A, Lauterbach SR, Wagner WH, Cossman DV, Long B, Cohen JL, et al. Spontaneous R5 infrarenal abdominal aortic dissection presenting as claudication: case report and review of the R6 literature. Ann Vasc Surg 2004 Jan;18(1):4-10. R7 (16) Hirst AE, Jr., JOHNS VJ, Jr., KIME SW, Jr. Dissecting aneurysm of the aorta: a review of 505 R8 cases. Medicine (Baltimore) 1958 Sep;37(3):217-79. R9 (17) Sueyoshi E, Sakamoto I, Hayashi K, Yamaguchi T, Imada T. Growth rate of aortic diameter in R10 patients with type B aortic dissection during the chronic phase. Circulation 2004 Sep 14;110(11 R11 Suppl 1):II256-II261. R12 (18) Tsai TT, Evangelista A, Nienaber CA, Myrmel T, Meinhardt G, Cooper JV, et al. Partial R13 thrombosis of the false lumen in patients with acute type B aortic dissection. N Engl J Med 2007 R14 Jul 26;357(4):349-59. R15 (19) Dake MD, Kato N, Mitchell RS, Semba CP, Razavi MK, Shimono T, et al. Endovascular R16 stent-graft placement for the treatment of acute aortic dissection. N Engl J Med 1999 May R17 20;340(20):1546-52. R18 (20) Nienaber CA, Fattori R, Lund G, Dieckmann C, Wolf W, von KY, et al. Nonsurgical R19 reconstruction of thoracic aortic dissection by stent-graft placement. N Engl J Med 1999 May R20 20;340(20):1539-45. R21 (21) Ahimastos AA, Aggarwal A, D’Orsa KM, Formosa MF, White AJ, Savarirayan R, et al. Effect of R22 perindopril on large artery stiffness and aortic root diameter in patients with Marfan syndrome: R23 a randomized controlled trial. JAMA 2007 Oct 3;298(13):1539-47. R24 (22) Matt P, Schoenhoff F, Habashi J, Holm T, Van EC, Loch D, et al. Circulating transforming R25 growth factor-beta in Marfan syndrome. Circulation 2009 Aug 11;120(6):526-32. R26 (23) Suzuki T, Distante A, Zizza A, Trimarchi S, Villani M, Salerno Uriarte JA, et al. Diagnosis of acute R27 aortic dissection by D-dimer: the International Registry of Acute Aortic Dissection Substudy on R28 Biomarkers (IRAD-Bio) experience. Circulation 2009 May 26;119(20):2702-7. R29 (24) Jean-Martin Charcot. De l’Expectation en medecine. These agregation medecine Paris. Paris: R30 Germer-Bailliere; 1857. R31 (25) Hata M, Sezai A, Niino T, Yoda M, Unosawa S, Furukawa N, et al. Should emergency surgical R32 intervention be performed for an octogenarian with type A acute aortic dissection? J Thorac R33 Cardiovasc Surg 2008 May;135(5):1042-6. R34 (26) Hovsepian DM, Ziporin SJ, Sakurai MK, Lee JK, Curci JA, Thompson RW. Elevated plasma R35 levels of matrix metalloproteinase-9 in patients with abdominal aortic aneurysms: a circulating R36 marker of degenerative aneurysm disease. J Vasc Interv Radiol 2000 Nov;11(10):1345-52. R37 (27) Suzuki T, Katoh H, Watanabe M, Kurabayashi M, Hiramori K, Hori S, et al. Novel biochemical R38 diagnostic method for aortic dissection. Results of a prospective study using an immunoassay of R39 smooth muscle myosin heavy chain. Circulation 1996 Mar 15;93(6):1244-9.

166 Chapter 14

Review Committee Acknowledgements List of publications Curriculum Vitae Chapter 14

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

168 Review Committee

REvIEw COMMITTEE R1 R2 Prof. dr. W.P.T.M Mali R3 Department of Radiology, University Medical Center Utrecht R4 Utrecht, the Netherlands R5 R6 Prof. dr. H.J.M. Verhagen R7 Department of Vascular Surgery, Erasmus Medical Center R8 Rotterdam, The Netherlands R9 R10 Prof. dr. G. Pasterkamp R11 Experimental Cardiology Laboratory, University Medical Center Utrecht R12 Utrecht, The Netherlands R13 R14 Prof. dr. J.R. Lahpor R15 Department of Cardio-Thoracic Surgery, University Medical Center Utrecht R16 Utrecht, The Netherlands R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 14 R38 R39

169

Acknowledgments

ACkNOwlEDGMENTS R1 R2 Dear Professor Moll, first of all I desire to thank you for the opportunity to defend this PhD. R3 Without your support, this would not have been possible. It represents the fantastic evolution R4 of our research collaboration, started with the visits and work in Milan of some of your guys, R5 first Frederik, then Jasper and, now, Jip. Working with important Universities in the Nederland R6 and in USA, like University of Utrecht and Yale University, was exciting and honoring. In all R7 the meetings and discussions that we had about our common programs, I had some lessons. I’m R8 positive that we’ll have an even further fruitful relationship and a sincere friendship in the future. R9 R10 Caro Professor Rampoldi, la passione ed il continuo impegno circa lo studio ed il trattamento R11 delle malattie dell’aorta toracica, in primo luogo delle dissezioni acute, ci lega ormai da oltre R12 venti anni. In questo tempo, fondamentali sono state per me la sua guida chirurgica, il suo R13 approccio filosofico a molte cose, anche di non stretta pertinenza medica, la sua conoscenza ed R14 il modo di approfondire le proprie passioni basato su una ricercatezza straordinaria, come per R15 l’alimentazione. La mia formazione chirurgica, mentale e culturale si poggia largamente su questi R16 concetti e rappresenterà la base del mio futuro. Grazie, spero di essere all’altezza. R17 R18 Dear Professor Eagle, Dear Kim, this PhD book is largely based on data and papers published by R19 the International Registry of Acute Aortic Dissection (IRAD), where you are one of the Principal R20 Investigators. You have been also senior author in my most important IRAD papers, reviewer R21 of my manuscripts since 2004, continuous supporter of my career. I tried to learn as much as R22 possible from the superb lessons of my scientific mentor. Dr. Eagle, an endless thank you for the R23 many gifts that you gave me, like your so warm friendship and making me feel at home in Ann R24 Arbor and in your place with Darlene and Taylor. R25 R26 Dear Professor Upchurch, Dear Gib, I desire to thank for our friendship and scientific R27 collaboration, that started in 2006, when we met in Ann Arbor. Our relationship has become R28 even closer when you moved to Charlottesville, Virginia, as Professor of Vascular Surgery, and R29 you invited me as your first foreign Visiting Professor at the University of Virginia. We had R30 good time both in USA and Italy and hope to have more in the future, also with the Upchurch’s R31 “band”. R32 R33 Dear Professor Elefteriades, it was great and useful to spend some time with you at the Yale Aortic R34 Center in 2008, doing clinical research on aortic diseases. After that we met many times at Yale R35 or during conferences and papers have been published and/or are being finalized. Thank you very R36 much for such opportunities. I’ve considered a truly privilege the invitation to introduce you at R37 the Bicentennial Lecture for the Yale Surgical Society in June, 2011. 14 R38 R39

171 Chapter 14

R1 Dear Professor Sundt, some papers reported in this book came from ideas that we discussed and R2 developed. I desire to thank you for the bright and fruitful revisions that you did and for sharing R3 with me some very nice time and interesting scientific proposals. R4 R5 Dear Dr. Froehlich, Eva, Jeanna, Dan, Patsy, Bruce, Dean, Elise and Alice, my thank to you and R6 to MCORRP for all the support and the friendship. As I said: “the best place to work”. R7 R8 Dear Professor Tsai, dear Tom, you are one of my idols. Everything you do is great. Thank you so R9 much for all support and for the delicious Chinese food that I discovered with you. R10 R11 Dear IRAD investigators, Dear Iradians, thank you to all of you for supporting our exciting and R12 very fruitful registry. Will we never finish? I hope no. R13 R14 Dear Dr. Muhs, dear Bart, our collaboration has been very fruitful since the first time that we R15 met at Yale. I hope that it’ll continue for a long time. Having collaboration with people that like R16 red wine is important. R17 R18 Caro Carlo, il mio ringraziamento a te nasce innanzi tutto dalla grande amicizia “acquisita” che ci R19 lega. Lavorare con te, sia in sala operatoria che fuori, rappresenta una delle parti migliori del mio R20 lavoro, fonte continua di novità e sviluppo di interessi comuni. Grazie per i vari insegnamenti, R21 il continuo supporto e la pazienza che hai in ogni occasione, con la speranza che questo tempo R22 comune duri il più a lungo possibile. R23 R24 Cara Viviana, la tua “temporanea” assenza dovuta all’arrivo di Martina, ha ulteriormente R25 sottolineato l’ìmportanza della tua figura nella nostra attività ospedaliera quotidiana, ed R26 in particolare nella mia. Un grazie grosso per quanto hai finora fatto, supportandomi e R27 sopportandomi, cosa che, spero, rifarai tra poco. R28 R29 Caro Valerio, la tua presenza nella mia vita “milanese” è stata, in senso positivo, decisiva. Hai R30 introdotto il sottoscritto, detto amichevolmente “terun”, a tutte le tue amicizie ed annessi giochi, R31 svaghi, vacanze: hai corso un rischio incredibile. Un abbraccio di cuore ed un grazie ancora a chi R32 mi ha inoltre permesso di partire, negli anni scorsi, per i vari viaggi in USA, sobbarcandosi il duro R33 lavoro quotidiano dell’Ospedale. R34 R35 Cari Paolo, Chiara, Sara e Matteo, grazie per il vostro continuo aiuto e per la vostra “fede” nel R36 sottoscritto. R37 R38 Caro Professor Ruberti, grazie per avermi accolto con grande benevolenza in una grande scuola R39 chirurgica come la sua.

172 Acknowledgments

Caro Dr. Frigiola, desidero ringraziarla per il supporto, le opportunità e gli insegnamenti R1 cardiochirurgici che mi sempre dato, e che mi hanno permesso di sviluppare la mia attività R2 chirurgica “ibrida”, cardiovascolare. Lei è la persona più “active” che abbia mai conosciuto su R3 questo pianeta, ma come fa? R4 R5 Caro Dr. Menicanti, un grazie di cuore per l’aiuto diretto ed indiretto con cui mi ha sempre R6 sostenuto e per la stima che mi dimostra quotidianamente. Lavorare con la Cardiochirurgia del R7 Policlinico San Donato è stato, ed è ancora, per me, fondamentale. Un privilegio assoluto. R8 R9 Caro Prof. Inglese, il tempo trascorso con lei in sala di emodinamica durante il trattamento R10 endovascolare di malattie aortiche è stato per me molto utile e divertente. Grazie per avermi R11 introdotto nel mondo delle “guide” e dei “cateteri”, oggi parte sostanziale della mia attività R12 lavorativa. R13 R14 Cari Prof. Pelissero, Dr. Bedin, Dr.ssa Cuppone, Colleghi ed Infermieri del Policlinico San R15 Donato, un grazie per il vostro supporto, la vostra disponibilità e la vostra amicizia nei miei R16 confronti. R17 R18 Caro Prof. Sangiorgi, caro Max, sei stato il primo “cornerstone” della mia attività di ricerca R19 scientifica di base. Come passare dal buio della caverna alla luce abbagliante di Circulation. Un R20 abbraccio forte. R21 R22 Caro Prof. Bossone, caro Eduardo, il tuo arrivo a San Donato nel 1998 è stato determinante R23 nel favorire l’inizio della costruzione dei nostri ponti scientifici con gli USA, attraverso l’IRAD. R24 Tanti viaggi insieme, tanti hotels ( il tutto non in prima classe ….), tante discussioni notturne. R25 La condivisione di tutto questo ha costruito e cementato una forte amicizia, come tu dici, del R26 “Regno delle due Sicilie”. R27 R28 Dear Anna, Barbara, Ryan and Rob, thanks for the wonderful support. Hope that our R29 collaboration will continue for a long time. R30 R31 Dear Jasper, there are so many emotions thinking of you. Indeed, too short, so warm, forever. R32 R33 Dear Frederik, I desire to thank you for the fantastic job we did and are still doing in this R34 time. You are a superb partner of a very fruitful research company and a very good friend. But, R35 more importantly, you came in Milan in the same season during which Inter F.C. won its third R36 Champion League, after 45 years. It was the first in my life and happened during our common R37 work and your stay in the city. Coincidence? I don’t know, but, please, come more often. 14 R38 R39

173 Chapter 14

R1 Dear Jip, it has been planned that I’ll be your boss for at least two years and you have been in R2 Milan since August 2011. In these 7 months, Inter F.C. played probably the worst soccer in the R3 last 40 years. Coincidence? I don’t know, but prefer to talk on different stuffs. You are doing R4 such a fantastic job on aortic research and we are totally busy every day. Probably, we’ll be even R5 busier in the next year. I have to thank you for your support in making this book and other very R6 important issues that I completed in the same time. Jip, you are a very good friend and also a R7 deep, deep expert to suggest nice restaurants, good bars and “interesting” night clubs across the R8 world, included Milan. Here, you know much more than me, even when I was a single. You are R9 a true professional. Chapeau. R10 R11 Dear Professor Verhagen, Dear Hence, it is just a pleasure and honor to collaborate with you, R12 doing research on aortic diseases. Thank you for this and hope this will continue in the future. R13 R14 Dear Dr. van Herwaarden, Dear Joost, we met because the Jasper accident, and, when people R15 meet in such occasions, the friendship can be immediate and robust. I hope that we’ll collaborate R16 for a long time, also sharing good time. My best congrats for your so nice family, still increasing R17 in number. Ad majora. R18 R19 Dear Cobie and Susan, thank you so much for your basic support in making this book, the PhD R20 defence and its organization possible. R21 R22 Cari Dr. Briguglio e Rizzo, Cari Bruno ed Antongiacomo, grazie per il vostro sostegno all’inizio R23 dei miei corsi di studio universitari, perchè niente evolve e si sviluppa senza un inizio. R24 R25 Caro Dr. Saitta, caro Peppe, sei stato “una presenza” durante l’Università a Messina, per meglio R26 dire, “la presenza”. Credo che abbiamo fatto un buon lavoro insieme. Grazie per tutto questo, ed R27 anche per tutto ciò che ancora fai per i miei genitori. R28 R29 Care Loredana e Nunziatina, a voi un grazie di cuore per l’aiuto fondamentale che mi avete dato R30 al mio arrivo a Milano. Siete un anello fondamentale della mia storia lavorativa. R31 R32 Caro Nico, grazie per le mandorle pugliesi, le nocciole delle Langhe, i fagioli della Ciociaria, R33 le lenticchie di Ustica, la carne del Parti, ….. grazie per la vostra amicizia. Il concetto del R34 “programma triennale” è geniale. R35 R36 Cara Valentina, che UPAR ti assista. R37 R38 R39

174 Acknowledgments

Dear Sammy, thank you for accepting to be the “Godfather” of Filippo. Probably you thought R1 that it was easy, but he seems to have a very strong personality, too strong. So, good luck, to you R2 and, above all, to me. R3 R4 Cara Carmen, grazie anche a te, ovvero una sorella che mi supporta “at large”, in tutto. R5 R6 Caro papà e cara mamma, questo libro è il risultato di anni di lavoro intenso ed appassionante a cui R7 avete partecipato attivamente anche voi. In quanto “donatori” del mio corredo genetico, magari R8 in modo inconscio, anche voi siete interessati alle malattie dell’aorta toracica, comunque per R9 proprietà transitiva. Per cui, considerato che sono stato fisicamente, ma non sentimentalmente, R10 più lontano da S. Teresa e da voi, non lamentatevi, era inevitabile. Vostro figlio Sanni. R11 R12 Caro Filippo, stella rossa del papà. Spero che questo libro ti possa un giorno essere di esempio su R13 come lavorare con tanti amici, in giro per il mondo. R14 R15 Cara Checca, il mio grazie a te non può che essere continuo ed enorme. Come moglie che mi R16 “sopporta” malgrado i grossi impegni che entrambi i nostri lavori comportano, e come mamma, R17 sempre attenta, amorevole, stimolante. Un bacio tenero. R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 14 R38 R39

175 Chapter 14

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

176 List of publications lIST Of PUblICATIONS R1 R2 Scientific Publications - Pubblicazioni Scintifiche R3 R4 1. F. Lemma, S. Lazzara, C. Lorenzini, A.G. Rizzo, M. Trovato, S. Trimarchi, F. D’Anna. R5 Utilità della citodiagnostica nelle lesione non palpabili della mammella. Minerva Chir R6 1991, 46(11): 599-603. R7 R8 2. F. Giordanengo, P. Pizzocari, V. Rampoldi, S. Trimarchi, M. De Monti, G.F. Giuffrida. R9 Gli pseudoaneurismi anastomotici non infetti a sede femorale. Minerva Chir 1992, 47. R10 R11 3. E.M. Bortolani, G. Ghilardi, P. Pizzocari, P. Coppini, F. Longhi, S. Trimarchi. Chirurgia R12 della carotide: morfologia della placca e correlazioni cliniche. Minerva Cardioangiologica R13 1992, 40: 369-374. R14 R15 4. G. Ghilardi, M. De Monti, F. Longhi, S. Trimarchi, E. Bortolani. Prevalenza del kinking R16 carotideo in una popolazione residente. Risultati parziali del programma OPI (Obiettivo R17 Prevenzione Ictus). Minerva Cardioangiologica 1993, 41: 129-32. R18 R19 5. G.F. Giuffrida,S. Trimarchi, S. Miani, A. Lovaria, P.L. Giorgetti, F. Giordanengo. R20 Chirurgia ricostruttiva dell’aorta sottorenale con veca cava inferiore sinistroposta. R21 Minerva Cardioangiologica 1993, 41. R22 R23 6. F. Giordanengo, S. Trimarchi, L. Franch, V. Rampoldi, S. Miani. Gli aneurismi R24 infiammatori dell’aorta sottorenale. Minerva Cardioangiologica 1994, 43. R25 R26 7. F. Giordanengo, P.L. Vandone, S.Trimarchi, N. Zaniboni, S. Miani. Ruptured aneurysm R27 of the internal iliac artery. Panminerva Med 1995, 37. R28 R29 8. V. Rampoldi, A. Arpesani, P.L. Giorgetti, A. Rignano, S. Trimarchi, D. Conti, U. R30 Ruberti. Ruptured aneurysms of the thoracoabdominal aorta: a cases series. Panminerva R31 Med 1995, 37(3): 123-8. R32 R33 9. P.L. Vandone, V. Tolva, S. Trimarchi, V. Rampoldi, G.F. Giuffrida, E.M. Bortolani. La R34 profilassi delle infezioni in chirurgia cardiovascolare. Minerva Chir 1998, 53: 397-403. R35 R36 10. V. Rampoldi, S. Trimarchi, V. Tolva, P. Righini. Proximal pseudo-aneurysm of ascending- R37 abdominal aortic bypass. European Journal of Cardio-Thoracic Surgery 2001,19:531-533. 14 R38 R39

177 Chapter 14

R1 11. G. Sangiorgi, R. Daverio, A. Mauriello, M. Bondio, S. Castelvecchio, M. Pontillo, S. R2 Trimarchi, V. Tolva, G. Nano, V. Rampoldi, L.G. Spagnoli, L. Inglese. Plasma levels R3 of metalloproteinases-3 and –9 as markers of successful abdominal aortic aneurysms R4 exclusion after endovascular graft treatment. Circulation 2001;104(suppl I):I-288-I-295. R5 R6 12. V. Rampoldi, P. Righini, S. Trimarchi, V. Tolva, L. Bonandrini. Single stage repair of R7 symptomatic type IV thoracoabdominal aortic and iliac aneurysm in Behcet’s disease. R8 J Cardiovascular Surgery 2001;42:691-4. R9 R10 13. V. Rampoldi, S. Trimarchi, V. Tolva, P. Righini. Acute type A aortic dissection and R11 coarctation of aortic isthmus. J Cardiovascular Surgery 2002;43(5):701-3. R12 R13 14. E. Bossone, V. Rampoldi, C. A. Nienaber, S. Trimarchi, A. Ballotta, J. Cooper, D. R14 Smith, K. A. Eagle, and R.H. Mehta. Usefulness of pulse deficit to predict in-hospital R15 complications and mortality in patients with acute type A aortic dissection. Am J Cardiol R16 2002;89:851-855. R17 R18 15. Suzuki T, Mehta RH, Nagai R, Sakamura Y, Sumiyoshi T, Bossone E, Trimarchi S, R19 Cooper JV, Smith DE, Isselbacher EM, Eagle KA, Nienaber CA. Clinical Profile and R20 Outcomes of Acute Type B Aortic Dissection in the Current Era. Lesson From the R21 International Registry of Acute Aortic Dissection (IRAD). Circulation 2003;108(suppl R22 II):II-312-II-317. R23 R24 16. V. Rampoldi, S. Trimarchi, P. Righini, V. Tolva and L. Inglese. Open Aortic Surgical R25 Repair for Left Hemi-Arch Stent- Graft Failure. Ann Thorac Surg 2004 Sep;78(3):1075- R26 8. R27 R28 17. S. Trimarchi, C.A. Nienaber, V. Rampoldi, T. Myrmel, T. Suzuki, R.H. Mehta, R29 E. Bossone, J.V. Cooper, D.E. Smith, L. Menicanti, A. Frigiola, J.K. Oh, M. Deeb, R30 E.M. Isselbacher, K. A. Eagle: Contemporary results of surgery in acute type A aortic R31 dissection: The International Registry of Acute Aortic Dissection experience. J Thorac R32 Cardiovasc Surg 2005;129:112-22. R33 R34 18. L. Bonavina, T. Lupattelli, D. Bona, S. Trimarchi, G. Nano, V. Rampoldi and L. R35 Inglese. Brochoesophageal fistola after endovascular repair of ruptured aneutysm of the R36 descending thoracic aorta. J Vasc Surg 2005 ;41 :712-4. R37 R38 R39

178 List of publications

19. Mukherjee D, Evangelista A, Nienaber CA, Sechtem U, Suzuki T, Trimarchi S., Llovet R1 A, Myrmel T, O’Gara PT, Fang J, Cooper JV, Oh JK, Jannuzzi JL, Hutchinson S, R2 Fattori R, Pape LA, Isselbacher EM, Eagle KA. Implications of Periaortic Hematoma in R3 Patients with Acute Aortic Dissection (from the International Registry of Acute Aortic R4 Dissection). Am J Cardiol 2005; Dec 15;96(12):1734-8. R5 R6 20. V. Rampoldi, S. Trimarchi. Aortic bulboplasty in acute type A aortic dissection. R7 J Cardiovasc Surg 2006;47:79-82. R8 R9 21. G. Sangiorgi, S. Trimarchi, A. Mauriello, P. Righini, E. Bossone, T. Suzuki, V. Rampoldi, R10 KA Eagle.Plasma levels of metalloproteinases-9 and -2 in the acute and subacute phases of R11 type A and type B aortic dissection. J Cardiovasc Med (Hagerstown). 2006 May;7(5):307- R12 315. R13 R14 22. G. Sangiorgi, A. Mauriello, E. Bonanno, C. Oxvig, CA Conover, M. Christiansen, R15 S. Trimarchi, V. Rampoldi, DR Holmes jr, RS Schwartz L.G. Spagnoli Pregnancy- R16 associated plasma protein-a is markedly expressed by monocyte-macrophage cells in R17 vulnerable and ruptured carotid atherosclerotic plaques: a link between inflammation R18 and cerebrovascular events. J Am Coll Cardiol. 2006 Jun 6;47(11):2201-11. R19 R20 23. Thomas T Tsai, Rossella Fattori, Santi Trimarchi, Christoph Nienaber, Truls Myrmel, R21 Stuart Hutchinson, Udo Sechtem, Jeanna V. Cooper, Jianming Fang, Linda A. Pape, R22 James Jannuzzi, Kim A. Eagle, Eric Isselbacher. Long-Term Survival in Patients with R23 Type A Aortic Dissection: Insights from the International Registry of Acute Aortic R24 Dissection (IRAD). Circulation 2006, Jul 4;114 (I Suppl):I350-6. R25 R26 24. Santi Trimarchi, Christoph A Nienaber, Vincenzo Rampoldi, Truls Myrmel, Toru R27 Suzuki, Valerio Tolva, Michael G Deeb, Gilbert Upchurch, Jeanna V. Cooper, Jianming R28 Fang, Eric M Isselbacher, Thoralf Kim A. Eagle. Role and Results of Surgery in Acute R29 Type B Aortic Dissection - Insights from the International Registry of Acute Aortic R30 Dissection (IRAD). Circulation 2006, Jul 4;114 (I Suppl):I357-64. R31 R32 25. Thomas T Tsai, Rossella Fattori, Santi Trimarchi, Eric Isselbacher, Truls Myrmel, R33 Arturo Evangelista, Stuart Hutchinson, Udo Sechtem, Jeanna V. Cooper, Dean E R34 Smith, Linda A. Pape, James Froelich, Arun Raghupathy, James Jannuzzi, Kim A. Eagle, R35 Christoph Nienaber,. Long-Term Survival in Patients with Type B Aortic Dissection: R36 Insights from the International Registry of Acute Aortic Dissection (IRAD). Circulation R37 2006,114:2226-2231 14 R38 R39

179 Chapter 14

R1 26. Tolva V, Bertoni GB, Trimarchi S, Grassi V, Fusari M, Rampoldi V. Unreliability of R2 Depopulated Bovine Ureteric Xenograft for Infra Inguinal Bypass Surgery: Mid-term R3 Results from Two Vascular Centres. Eur J Vasc Endovasc Surg. 2007;33(2):214-6. R4 R5 27. Rampoldi V, Trimarchi S, Eagle KA, Nienaber CA, Oh J, Bossone E, Myrmel T, R6 Sangiorgi GM, De Vincentiis C, Cooper JV, Fang J, Smith DE, Tsai T, Raghupathy A, R7 Fattori R, Sechtem U, Deeb MG, Sundt T IIIrd, Isselbacher EM. Simple risk models to R8 predict surgical mortality in acute type A aortic dissection: the International Registry of R9 Acute Aortic Dissection score. Ann Thorac Surg. 2007 Jan;83(1):55-61. R10 R11 28. Bossone E, Di Benedetto G, Frigiola A, Carbone GL, Panza A, Cirri S, Ballotta A, R12 Messina S, Rega S, Citro S, Trimarchi S, Fang J, Righini P, Distante A, Eagle KA, Mehta R13 RH. Valve surgery in octagenarians: in-hospital and long-term outcomes. Can J Cardio R14 2007 Mar 1;23(3):223-7. R15 R16 29. Bossone E, Evangelista A, Isselbacher E, Trimarchi S, Hutchison S, Gilon D, Llovet A, R17 O’Gara P, Cooper JV, Fang J, Jannuzzi JL, Mehta RH, Distante A, Nienaber CA, Eagle R18 KA, Armstrong WF; International Registry of Acute Aortic Dissection Investigators. R19 Prognostic role of transesophageal echocardiography in acute type A aortic dissection. R20 Am Heart J. 2007 Jun;153(6):1013-20. R21 R22 30. Pape LA, Tsai TT, Isselbacher EM, Oh JK, O’Gara PT, Evangelista A, Fattori R, R23 Meinhardt G, Trimarchi S, Bossone E, Suzuki T, Cooper JV, Froehlich J, Nienaber R24 CA, Eagle KA. Aortic diameter > or = 5.5 cm is not a good predictor of type A aortic R25 dissections: observation from the International Registry of Acute Aortic Dissection R26 (IRAD). Circulation 2007 Sep 4;116(10):1120-7. R27 R28 31. Thomas T Tsai, Eric M Isselbacher,Santi Trimarchi, Eduardo Bossone, Linda Pape, R29 James L Jannuzzi, Arturo Evangelista, Jae Oh, Alfredo Llovet, Joshue Beckman, Jeanna R30 V. Cooper, Dean E Smith, James Froelich, Rossella Fattori, Kim A. Eagle, Christoph R31 A Nienaber,. Acute Type B Aortic Dissection: Does Aortic Arch Involvement Affect R32 Management and Outcomes? Insights from the International Registry of Acute Aortic R33 Dissection (IRAD). Circulation 2007 Sep 11;116(11 Suppl):I150-6 R34 R35 32. Santi Trimarchi, Thomas T Tsai, Kim A. Eagle, Eric M Isselbacher, Jim Froelich, R36 Jeanna V. Cooper, Vincenzo Rampoldi, Gilbert R Upchurch. Acute Abdominal Aortic R37 Dissection: Insights from the International Registry of Acute Aortic Dissection (IRAD). R38 J Vasc Surg 2007 Nov;46(5):913-919. R39

180 List of publications

33. Trimarchi S, Tolva VS, Grassi V, Frigiola A, Carminati M, Rampoldi V. Descending R1 thoracic and abdominal aortic coarctation in the young: Surgical treatment after R2 percutaneous approaches failure. J Vasc Surg. 2008 Apr;47(4):865-867. R3 R4 34. de Vincentiis C, Kunkl AB, Trimarchi S, Gagliardotto P, Frigiola A, Menicanti L, Di R5 Donato M. Aortic valve replacement in octogenarians: is biologic valve the unique R6 solution? Ann Thorac Surg. 2008 Apr;85(4):1296-301. R7 R8 35. Suzuki T, Distante A, Zizza A, Trimarchi S, Villani M, Salerno Uriarte JA, de Luca R9 Tupputi Schinosa L, Renzulli A, Sabino F, Nowak R, Birkhahn R, Hollander JE, R10 Counselman F, Bossone E, Eagle K; International Registry of Acute Aortic Dissection R11 Substudy on Biomarkers (IRAD-Bio) Investigators. Preliminary experience with the R12 smooth muscle troponin-like protein, calponin, as a novel biomarker for diagnosing R13 acute aortic dissection. Eur Heart J. 2008 Jun;29(11):1439-45 R14 R15 36. Bossone E, Trimarchi S, Esposito G, Aliberti S, Citro R, Allegra L, Blasi F.Chlamydia R16 pneumoniae and acute aortic syndrome: a call for a multi-institutional study. Monaldi R17 Arch Chest Dis. 2008 Jun;70(2):68-70. R18 R19 37. Raghupathy A, Nienaber CA, Harris KM, Myrmel T, Fattori R, Sechtem U, Oh J, R20 Trimarchi S, Cooper JV, Booher A, Eagle K, Isselbacher E, Bossone E; International R21 Registry of Acute Aortic Dissection (IRAD) Investigators.Geographic differences R22 in clinical presentation, treatment, and outcomes in type A acute aortic dissection R23 (from the International Registry of Acute Aortic Dissection). Am J Cardiol. 2008 Dec R24 1;102(11):1562-6. R25 R26 38. Complicated Acute Type B Dissection: Is Surgery Still the Best Option?: A Report From R27 the International Registry of Acute Aortic Dissection. Rossella Fattori, Thomas T. Tsai, R28 Truls Myrmel, Arturo Evangelista, Jeanna V. Cooper, Santi Trimarchi, Jin Li, Luigi R29 Lovato, Stephan Kische, Kim A. Eagle, Eric M. Isselbacher, and Christoph A. Nienaber. R30 J. Am. Coll. Cardiol. Intv. 2008;1;395-402. R31 R32 39. Acute aortic dissection: perspectives from the International Registry of Acute Aortic R33 Dissection (IRAD). Tsai TT, Trimarchi S, Nienaber CA. Eur J Vasc Endovasc Surg. 2009 R34 Feb;37(2):149-59. R35 R36 40. A better understanding of diagnosis and treatment of isolated abdominal aortic R37 dissection. S Trimarchi, V. Tolva, V. Grassi, V. Rampoldi and G. Upchurch. J Endovasc 14 R38 Ther 2009;16:81-3. R39

181 Chapter 14

R1 41. Retrograde ascending aortic dissection during or after thoracic aortic stent graft placement: R2 insight from the European registry on endovascular aortic repair complications. R3 Eggebrecht H, Thompson M, Rousseau H, Czerny M, Lönn L, Mehta RH, Erbel R; R4 European Registry on Endovascular Aortic Repair Complications. Circulation. 2009 Sep R5 15;120(11 Suppl):S276-81. R6 R7 42. pathobiologic link between risk factors profile and morphological markers of carotid R8 instability. Mauriello A, Sangiorgi GM, Virmani R, Trimarchi S, Holmes DR Jr, R9 Kolodgie FD, Piepgras DG, Piperno G, Liotti D, Narula J, Righini P, Ippoliti A, R10 Spagnoli LG. Atherosclerosis. 2010 Feb;208(2):572-80. Epub 2009 Aug 3 R11 R12 43. Acute management of aortobronchial and aortoesophageal fistulas using thoracic R13 endovascular aortic repair. Jonker FH, Heijmen R, Trimarchi S, Verhagen HJ, Moll FL, R14 Muhs BE. J Vasc Surg. 2009 Nov;50(5):999-1004. R15 R16 44. Diagnosis of acute aortic dissection by D-dimer: the International Registry of Acute R17 Aortic Dissection Substudy on Biomarkers (IRAD-Bio) experience. Suzuki T, Distante R18 A, Zizza A, Trimarchi S, Villani M, Salerno Uriarte JA, De Luca Tupputi Schinosa L, R19 Renzulli A, Sabino F, Nowak R, Birkhahn R, Hollander JE, Counselman F, Vijayendran R20 R, Bossone E, Eagle K; IRAD-Bio Investigators. Circulation. 2009 May 26;119(20):2702- R21 7. R22 R23 45. Role of age in acute type A aortic dissection outcome: Report from the International R24 Registry of Acute Aortic Dissection (IRAD).Trimarchi S, Eagle KA, Nienaber CA, R25 Rampoldi V, H W Jonker F, De Vincentiis C, Frigiola A, Menicanti L, Tsai T, Froehlich R26 J, Evangelista A, Montgomery D, Bossone E, Cooper JV, Li J, G Deeb M, Meinhardt G, R27 Sundt TM, Isselbacher EM; International Registry of Acute Aortic Dissection (IRAD) R28 Investigators. J Thorac Cardiovasc Surg. 2010 Oct;140(4):784-9. R29 R30 46. Meta-analysis of open versus endovascular repair for ruptured descending thoracic aortic R31 aneurysm. Jonker FH, Trimarchi S, Verhagen HJ, Moll FL, Sumpio BE, Muhs BE. J R32 Vasc Surg. 2010 Apr;51(4):1026-32, 1032.e1-1032.e2 R33 R34 47. Outcomes of endovascular repair of ruptured descending thoracic aortic aneurysms. R35 Jonker FH, Verhagen HJ, Lin PH, Heijmen RH, Trimarchi S, Lee WA, Moll FL, R36 Athamneh H, Muhs BE. Circulation. 2010 Jun 29;121(25):2718-23. Epub 2010 Jun 14. R37 R38 R39

182 List of publications

48. Long-term outcomes of surgical aortic fenestration for complicated acute type B aortic R1 dissections. Trimarchi S, Jonker FH, Muhs BE, Grassi V, Righini P, Upchurch GR, R2 Rampoldi V. J Vasc Surg. 2010 Aug;52(2):261-6. Epub 2010 Jun 22. R3 R4 49. Retroperitoneal hematoma after percutaneous coronary intervention: prevalence, risk R5 factors, management, outcomes, and predictors of mortality: a report from the BMC2 R6 (Blue Cross Blue Shield of Michigan Cardiovascular Consortium) registry. Trimarchi R7 S, Smith DE, Share D, Jani SM, O’Donnell M, McNamara R, Riba A, Kline-Rogers E, R8 Gurm HS, Moscucci M; BMC2 Registry. JACC Cardiovasc Interv. 2010 Aug;3(8):845- R9 50. R10 R11 50. Endovascular management of type B aortic dissections. DiMusto PD, Williams DM, R12 Patel HJ, Trimarchi S, Eliason JL, Upchurch GR Jr. J Vasc Surg. 2010 Oct;52(4 R13 Suppl):26S-36S. R14 R15 51. Importance of Refractory Pain and Hypertension in Acute Type B Aortic Dissection: R16 Insights From the International Registry of Acute Aortic Dissection (IRAD). Trimarchi R17 S, Eagle KA, Nienaber CA, Pyeritz RE, Jonker FH, Suzuki T, O’Gara PT, Hutchinson R18 SJ, Rampoldi V, Grassi V, Bossone E, Muhs BE, Evangelista A, Tsai TT, Froehlich JB, R19 Cooper JV, Montgomery D, Meinhardt G, Myrmel T, Upchurch GR, Sundt TM, R20 Isselbacher EM; on behalf of the International Registry of Acute Aortic Dissection R21 (IRAD) Investigators. Circulation. 2010 Sep 28;122(13):1283-1289. R22 R23 52. Endovascular Repair of Ruptured Thoracic Aortic Aneurysms: Predictors of Procedure- R24 related Stroke. Jonker FH, Verhagen HJ, Heijmen RH, Lin PH, Trimarchi S, Lee WA, R25 Moll FL, Athamneh H, Muhs BE. Ann Vasc Surg. 2011 Jan;25(1):3-8. Epub 2010 Oct 6. R26 R27 53. In search of blood tests for thoracic aortic diseases. Trimarchi S, Sangiorgi G, Sang X, R28 Rampoldi V, Suzuki T, Eagle KA, Elefteriades JA. Ann Thorac Surg. 2010 Nov;90(5):1735- R29 42. Review R30 R31 54. Endovascular Treatment of Ruptured Thoracic Aortic Aneurysm in Patients Older than R32 75 Years. Jonker FH, Verhagen HJ, Heijmen RH, Lin PH, Trimarchi S, Lee WA, Moll R33 FL, Athamneh H, Muhs BE. Eur J Vasc Endovasc Surg. 2011 Jan;41(1):48-53. Epub R34 2010 Nov 24. R35 R36 R37 14 R38 R39

183 Chapter 14

R1 55. Ascending Thoracic Aorta Dimension and Outcomes in Acute Type B Dissection (from R2 the International Registry of Acute Aortic Dissection [IRAD]).Booher AM, Isselbacher R3 EM, Nienaber CA, Froehlich JB, Trimarchi S, Cooper JV, Demertzis S, Ramanath VS, R4 Januzzi JL, Harris KM, O’Gara PT, Sundt TM 3rd, Pyeritz RE, Eagle KA; International R5 Registry of Acute Aortic Dissection (IRAD) Investigators. Am J Cardiol. 2011 Jan R6 15;107(2):315-20. R7 R8 56. Open surgery versus endovascular repair of ruptured thoracic aortic aneurysms. Jonker R9 FH, Verhagen HJ, Lin PH, Heijmen RH, Trimarchi S, Lee WA, Moll FL, Atamneh H, R10 Rampoldi V, Muhs BE. J Vasc Surg. 2011 May;53(5):1210-6. Epub 2011 Feb 5. R11 R12 57. Descending aortic diameter of 5.5 cm or greater is not an accurate predictor of acute R13 type B aortic dissection. Trimarchi S, Jonker FH, Hutchison S, Isselbacher EM, Pape R14 LA, Patel HJ, Froehlich JB, Muhs BE, Rampoldi V, Grassi V, Evangelista A, Meinhardt R15 G, Beckman J, Myrmel T, Pyeritz RE, Hirsch AT, Sundt Iii TM, Nienaber CA, Eagle R16 KA. J Thorac Cardiovasc Surg. 2011 Sep;142(3):e101-7. Epub 2011 May 17. R17 R18 58. Do branched and fenestrated devices have a role in chronic type B aortic dissection? R19 Trimarchi S, Righini P, Grassi V, Lomazzi C, Segreti S, Rampoldi V, Verhoeven E. J R20 Cardiovasc Surg (Torino). 2011 Aug;52(4):529-38. R21 R22 59. Circulating transforming growth factor-Beta levels in acute aortic dissection.Suzuki T, R23 Trimarchi S, Sawaki D, Grassi V, Costa E, Rampoldi V, Nagai R, Eagle K. J Am Coll R24 Cardiol. 2011 Aug 9;58(7):775. R25 R26 60. Invited commentary. Trimarchi S, De Vincentiis C. Ann Thorac Surg. 2011 R27 Dec;92(6):2084. R28 R29 61. Evidence of a topographical link between unstable carotid plaques and luminal stenosis: R30 Can we better stratify asymptomatic patients with significant plaque burden? Mauriello R31 A, Sangiorgi G, Virmani R, Servadei F, Trimarchi S, Holmes DR Jr, Kolodgie F, Biondi R32 Zoccai G, Leuzzi C, Spagnoli LG. Int J Cardiol. 2011 Dec 20. R33 R34 62. Mechanisms of Symptomatic Spinal Cord Ischemia After TEVAR: Insights From the R35 European Registry of Endovascular Aortic Repair Complications (EuREC). Czerny M, R36 Eggebrecht H, Sodeck G, Verzini F, Cao P, Maritati G, Riambau V, Beyersdorf F, Rylski R37 B, Funovics M, Loewe C, Schmidli J, Tozzi P, Weigang E, Kuratani T, Livi U, Esposito R38 G, Trimarchi S, van den Berg JC, Fu W, Chiesa R, Melissano G, Bertoglio L, Lonn L, R39 Schuster I, Grimm M. J Endovasc Ther. 2012 Feb;19(1):37-43.

184 List of publications

63. Immediate conversion to CAS after neurological intolerance at cross-clamping test R1 during CEA: a preliminary experience. Guy Bianchi P, Tolva V, Dalainas I, Bertoni G, R2 Cireni L, Trimarchi S, Rampoldi V, Casana R. Int Angiol. 2012 Feb;31(1):22-7. R3 R4 64. Clinical presentation, management, and short-term outcome of patients with type R5 A acute dissection complicated by mesenteric malperfusion: Observations from the R6 International Registry of Acute Aortic Dissection. Di Eusanio M, Trimarchi S, Patel R7 HJ, Hutchison S, Suzuki T, Peterson MD, Di Bartolomeo R, Folesani G, Pyeritz RE, R8 Braverman AC, Montgomery DG, Isselbacher EM, Nienaber CA, Eagle KA, Fattori R. R9 J Thorac Cardiovasc Surg. 2012 Feb 14. [Epub ahead of print] R10 R11 Chapters of books R12 R13 1) D.G. Tealdi, V. Rampoldi, S. Trimarchi. Le dissecazioni aortiche di tipo B: trattamento R14 medico o chirugico? Gli aneurismi dell’aorta toracica e toraco-addominale. 1998, pag. 203- R15 213. Europa Scienze Umane Editrice. R16 R17 2) V. Rampoldi, S. Trimarchi, V. Tolva, A. Arpesani. Le Dissecazioni Aortiche. “Emergenze R18 Vascolari” Ruberti U, Scorza R; Ed. UTET 2003. R19 R20 3) Giuseppe M Sangiorgi, Alessandro Mauriello, Frank Kolodgie, Santi Trimarchi, R21 Giuseppe Biondi Zoccai, Renu Virmani, Luigi Giusto Spagnoli. Pathobiology of the R22 asymptomatic atherosclerotic carotid plaque. 2006 R23 R24 4) S. Trimarchi, Truls Myrmel, Antonio Panza, Giuseppe Di Benedetto, V. Rampoldi. R25 Vascular Surgical Options. Aortic Dissection and Related Syndromes, pag 333-346; Ed. R26 Springer 2007. R27 R28 5) Truls Myrmel, S. Trimarchi, V. Rampoldi. Cardiovascular surgery in the initial treatment R29 of aortic dissection and acute related syndrome. Aortic Dissection and Related Syndromes, R30 pag 167-190; Ed. Springer 2007. R31 R32 R33 R34 R35 R36 R37 14 R38 R39

185 Chapter 14

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13 R14 R15 R16 R17 R18 R19 R20 R21 R22 R23 R24 R25 R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 R38 R39

186 Curriculum Vitae

Curriculum Vitae R1 R2 Santi Trimarchi was born on April 14, 1965, in Santa Teresa di Riva (Messina), Italy. After R3 graduating from scientific secondary school Enrico Trimarchi in Santa Teresa, he obtained the R4 University Degree in Medicine, cum laude, at the University of Messina in 1990. In September R5 1990, he moved to Milano where began his residency in Vascular Surgery at the Institute of R6 General and Cardiovascular Surgery of the University of Milano, directed by Professor Ugo R7 Ruberti. He graduated as Vascular Surgeon on July 1995. From September 1995 to February R8 1997, he was fellow in Cardiac Surgery at Policlinico San Donato, San Donato Milanese, trained R9 by Drs. Alessandro Frigiola e Lorenzo Menicanti. Since February 1997, he has been Staff Member R10 at the II° section of Vascular Surgery of the same Institution, Head Dr. Vincenzo Rampoldi. R11 Under his supervision, Santi started to operate on a wide spectrum of vascular diseases, with R12 particular clinical and scientific interest on thoracic aortic dissections and aneurysms. In 2008 R13 he obtained a Master in Endovascular Technologies (MET) at the University of Milano Bicocca. R14 Based on the clinical interests on thoracic aortic diseases, he became co-investigator of the R15 International Registry of Aortic Dissection (IRAD) in 2000, co-director of the IRAD surgical/ R16 endovascular program in 2010 and investigator in other national and international studies. In R17 the last decade, Santi has been visiting research fellow at the University of Michigan (UMICH), R18 Ann Arbor, Michigan, USA, where also spent a sabbatical during the spring and summer seasons R19 in 2006, and at the Yale University, New Haven, Connecticut, USA in November-December R20 2008. He also visited other important Aortic Center in USA, like the Methodist Hospital in R21 Houston, Cleveland Clinic and Mayo Clinic, and currently collaborates with Universities of R22 Utrecht, Rotterdam, Tokyo and Osaka. Santi has been speaker in many national and international R23 meetings, including as Visiting Professor at the University of Virginia (UVA), Charlottesville, R24 USA. Since March 2011, he is Director of the Thoracic Aortic Research Center at Policlinico R25 San Donato IRCCS. R26 R27 R28 R29 R30 R31 R32 R33 R34 R35 R36 R37 14 R38 R39

187