Endovascular treatment of acute aortic syndrome

Rachel E. Clough, MB, BS, BSc, MRCS,a Kevin Mani, MD, PhD,a,b Oliver T. Lyons, MB, BS, BSc, MRCS,a Rachel E. Bell, MB, BS, MS, FRCS,a Hany A. Zayed, MD, MSc, FRCS,a Matthew Waltham, MA, PhD, FRCS,a Tom W. Carrell, MA, MChir, FRCS,a and Peter R. Taylor, MA, MChir, FRCS,a London, United Kingdom; and Uppsala, Sweden

Background: The term acute aortic syndrome (AAS) encompasses a range of conditions that have a risk of imminent aortic rupture and where delays in treatment result in increased mortality. Endovascular treatment offers an attractive alternative to open but little is known about the durability of the repair and the factors that predict mortality. Methods: Prospective data were collected for a cohort of 110 consecutive patients with endovascular treatment for AAS. Patient and procedural characteristics were related to short- and midterm outcome using multivariate logistic regression analysis. Results: There were 75 men and 35 women with a median age of 68 (range 57-76) years. The pathologies treated were acute (35), symptomatic (32), infected aneurysm (18), transection (12), chronic dissection (9), penetrating ulcer (3), and intramural hematoma (1). Thirty-day mortality was 12.7% and this was associated with (odds ratio [OR], 5.25), use of general anesthetic (OR, 5.23), long procedure duration (OR, 2.03), and increasing age (OR, 1.07). The causes of death were aortic rupture (4), myocardial infarction (4), stroke (3), and multisystem organ failure (3). The stroke and paraplegia rates were 7.3% and 6.4%, respectively. The 1-year survival was 81% and the 5-year survival 63%. Secondary procedures were required in 13 (11.8%) patients. Factors associated with death at 1 year were presence of an aortic fistula (OR, 9.78), perioperative stroke (OR, 5.87), and use of general anesthetic (OR, 3.76); and at 5 years were aortic fistula (OR, 12.31) and increasing age (OR, 1.06). Conclusions: Acute aortic syndrome carries significant early and late mortality. Emergency endovascular repair offers a minimally invasive treatment option associated with acceptable short and midterm results. Continued surveillance is important as secondary procedures and aortic-related deaths continue to occur throughout the follow-up period. (J Vasc Surg 2011;54:1580-7.)

The concept of grouping a range of clinical conditions aortic rupture will allow early identification of patients with which share a common life-threatening risk into a single chest due to an aortic cause and expedite early imple- syndrome came from cardiology, where the term acute mentation of definitive treatment.4,5 Aneurysm, dissection, coronary syndrome (ACS) allowed a number of different transection, penetrating ulcer, and intramural hematoma clinical presentations of threatened coronary occlu- can be grouped together, not because they share the same sion to be considered together, to promote early identifi- pathology, but because in the acute setting these aortic cation and treatment and to improve survival.1-3 conditions require early identification, transfer to an appro- The concept of an acute aortic syndrome (AAS) to priate care facility, rapid institution of medical therapy and encompass those conditions where there is risk of imminent urgent aortic repair with high quality anesthesia, and a cardiovascular intensive care unit for postprocedure care.6,7 The reason for grouping these conditions into one entity is From the Department of , NIHR Comprehensive Biomed- ical Research Centre of Guy’s and St. Thomas’ NHS Foundation Trust so that specific algorithms can be developed in centers and King’s College London, King’s Health Partners, Londona; and the specializing in aortic intervention with the aim of improv- Institution of Surgical Sciences, Department of Vascular Surgery, Uppsala ing outcome.8-10 b University, Uppsala. Traditional management of AAS includes open surgical Supported by a Clinical Training Fellowship from the National Institute for Health Research (R.E.C.), Swedish Heart and Lung Foundation and the replacement of the thoracic and is associated with a 11,12 Uppsala University Hospital (K.M.), and by a Clinical Research Training high perioperative mortality and morbidity. Endovas- Fellowship from the Medical Research Council (O.T.L.). cular repair of aortic pathology is increasingly being used as Competition of interest: none. an alternative to open surgery.13-15 The published litera- Presented at the meetings of the European Society for Vascular Surgery ture contains few articles on the endovascular treatment of (Amsterdam, September 2010) and the Vascular Society of Great Britain and Ireland (Brighton, UK, November 2010). patients presenting with acute conditions of the thoracic Reprint requests: Rachel E. Clough, Department of Vascular Surgery, aorta. We report the midterm follow-up of our experience NIHR Comprehensive Biomedical Research Centre of Guy’s and St. of the endovascular management of AAS and investigate Thomas’ NHS Foundation Trust and King’s College London, and King’s factors that may predict mortality. Health Partners, Westminster Bridge Road, London, SE1 7EH, UK (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships METHODS to disclose per the JVS policy that requires reviewers to decline review of any The cohort consisted of 110 consecutive patients with manuscript for which they may have a competition of interest. acute thoracic aortic pathology (pathology affecting the 0741-5214/$36.00 Copyright © 2011 by the Society for Vascular Surgery. aortic arch and/or descending thoracic aorta) who under- doi:10.1016/j.jvs.2011.07.034 went endovascular treatment at a single university teaching 1580 JOURNAL OF VASCULAR SURGERY Volume 54, Number 6 Clough et al 1581 hospital between August 1997 and September 2009. A Table I. Patient demographics and procedure details team of five vascular surgeons and five interventional radi- ologists provided a 24-hour emergency service. Variable The study was designed as a retrospective analysis of a Age prospectively maintained database, case notes, and imaging (median [IQR]) 68 (57-76) studies. Patient and procedural characteristics were collated Gender on a standardized pro forma. (male:female) 75:35 Preoperative imaging. All patients underwent bolus- Aortic pathology Acute dissection 35 tracked computed tomography (CT) angiography. Images Symptomatic aneurysm 32 were interpreted by both experienced radiologists and sur- Infected aneurysm 18 geons to determine suitability for endovascular repair. Du- Transection 12 plex sonography was used to assess the ascending aorta to Chronic dissection 9 exclude type A dissection and to examine the status of the Penetrating ulcer 3 Intramural hematoma 1 carotid and vertebral if deliberate occlusion of the Device type arch vessels was planned. Magnetic resonance imaging was Gore TAG 66 not used due to its lack of availability in the emergency Gore Excluder 15 setting. Cook TX2 7 Medtronic Talent 9 Initial medical treatment. Patients presenting with Medtronic AneuRx 4 , penetrating ulcer, and intramural hema- Medtronic Valiant 1 toma were treated initially with hypotensive medication Endofit 6 and analgesia on a high dependency, intensive care, or a Gore cTAG 2 coronary care unit. was titrated to ensure IQR, Interquartile range. adequate urine output. Initial treatment was with ␤-blockers and ␣-blockers then angiotensin converting enzyme (ACE)- inhibitors, angiotensin-2 receptor blockers, calcium chan- clinic at 4 months. CT angiography and clinical review were nel blockers, and diuretics. However, patients presenting performed annually thereafter. Data on follow-up were with circulatory shock were taken immediately to the en- obtained from hospital electronic records and community dovascular suite after a diagnostic CT scan. hospitals and community care practitioners. Operative technique. Patients who were conscious Statistical analysis. SPSS Statistics 17.0 software gave written consent. All operative procedures were per- (SPSS, Inc, Chicago Ill) was used for statistical analysis. formed in a hybrid operating theater with fluoroscopic and Quantitative variables were expressed as median and inter- angiographic equipment available. Vascular access was quartile range (IQR). Instability was defined as systolic achieved via femoral arteriotomy or prosthetic conduit. A blood pressure Ͻ90 mm Hg. Renal disease was defined as catheter was inserted in the contralateral groin so that glomerular filtration rate (GFR) Ͻ90 mL/min/1.73m2 or continuous angiography could be performed during device a serum creatinine Ͼ120 mmol/L. The relationship be- deployment. The majority of procedures were performed tween death at 30 days, 1 year, and 5 years, and patient and under loco-regional anesthesia without routine preopera- procedural characteristics was compared, and correspond- tive placement of a cerebrospinal fluid (CSF) drain. Pro- ing odds ratios were calculated using univariate logistic phylactic spinal drain insertion was considered if the proce- regression analysis. Variables with a P value Ͻ.2 were dure was performed under general anesthetic; there was entered into a multivariate logistic regression model using long segment coverage (the descending thoracic aorta from forced entry analysis. To enable evaluation of all cases, proximal to the left subclavian to the coeliac axis), planned missing values were replaced with mean values for the occlusion of the left subclavian artery, extensive disease variable. No other variables were imputed. A P value of (proximal to the left subclavian to the coeliac axis with Յ.05 was considered significant. A Kaplan-Meier survival involvement of infra-renal aorta and/or internal iliacs), plot was produced to show the overall survival of the planned occlusion of the internal iliac arteries, or previous cohort. infra-renal aortic repair. Adjuvant procedures. The innominate and left com- RESULTS mon carotid arteries were always revascularized if coverage Patients and operating characteristics. The pathol- of the origin was planned. The left subclavian artery was ogy, demographics, and procedural details are shown in assessed in each patient taking into account the status of the Table I. The male: female ratio was 75:35, and the median vertebral arteries, the length of aorta to be covered, the age was 68 years (IQR, 57-76). Complicated acute type B presence of a previous infra-renal repair or aneurysm, and dissection was the most common pathology with 19 pa- the status of the internal iliac arteries. Revascularization of tients having a leaking aorta, 11 with end organ ischemia, arch vessels was performed immediately before endovascu- and five with continuing pain. Aortic fistulae included lar repair under the same anesthetic. aorto-bronchial, aorto-enteric, and aorto-cutaneous cases. Postoperative follow-up. CT angiography was per- Thirty-five percent of patients (38/110) had evidence formed at 3 months, and the patient was reviewed in the of aortic rupture on preoperative imaging, 26.4% (29/110) JOURNAL OF VASCULAR SURGERY 1582 Clough et al December 2011

Table II. Adjuvant procedures Table III. In-hospital events

Adjuvant procedure Event

Debranching Stroke Innominate 1 Anterior circulation 6 Carotid-carotid 9 Posterior circulation 2 Carotid-subclavian 1 Paraplegia Conduit Reversed with CSF drain 4 Iliac 4 Permanent 3 Aortic 1 Death Procedures related to operative complications Aortic rupture 4 Iliac artery injury Myocardial infarction 4 Femoro-femoral cross-over 2 Stroke 3 Femoral artery injury Multisystem organ failure 3 Patch angioplasty 5 Other CSF, Cerebrospinal fluid. Left subclavian artery stent 1 Table IV. Secondary procedures

Indication Procedure were hemodynamically unstable, and 23.6% (26/110) had hemothorax. Endoleak (type I) Explant 3 Endovascular repair was performed within 24 hours of Proximal extension cuff 1 Distal extension cuff 2 presentation to the tertiary care center in 85.5% of patients Proximal and distal extension cuff 1 (94/110). The technical success rate was 98.2% (108/ Aortic banding 2 110). The two failures were due to inability to seal the Infection Further stent graft 3 proximal landing zone in patients with symptomatic aneu- Paraplegia Cerebrospinal fluid drain 1 rysm and a tortuous aorta. The proximal landing zone was Ishimaru zone 0 in one patient, zone 1 in nine patients, and zone 2 in 54 patients. patients died following myocardial infarction (one of these Adjuvant procedures were performed in 16 patients (Table presented with acute dissection). One case of myocardial II). Eleven patients had debranching procedures of the infarction was secondary to deliberate coverage of the aortic arch to provide an adequate proximal landing zone. origin of the left subclavian artery in a patient with aortic One patient required an aortic conduit for access as the rupture, without knowledge that this gave rise to a left length of the introducer sheath was designed for use in the internal mammary coronary bypass graft. A left subclavian infra-renal aorta. Procedures related to complications were artery chimney stent was deployed to re-establish antegrade performed in eight patients (Table II). The external iliac flow via a brachial artery puncture. artery was inadvertently ruptured in two patients and Overall 15 patients suffered neurological complications treated with oversew and femoro-femoral cross-over graft- (Table III). Eight patients (7.3%) had a stroke (3 died), 6 ing. Five patients required patch angioplasty of the com- related to the anterior circulation and 2 related to the mon femoral artery for arterial damage caused by the intro- posterior circulation. Of the patients with anterior circula- ducer sheath. The median procedure duration was 90 tion stroke, two had as their primary aortic pa- (IQR, 66-120) minutes and median blood loss was 200 thology; of those with a posterior circulation stroke, the (IQR, 150-400) mL. stent graft covered the origin of the left subclavian artery in Early outcomes. The overall in-hospital mortality was both cases without revascularization. 12.7% (14/110) (Table III). Deaths classified according to Seven patients developed paraplegia (6.4%); the proce- pathology were acute dissection 2/35 (mortality 5.7%), dures had been performed under loco-regional anesthesia symptomatic aneurysm 6/32 (18.8%), mycotic aneurysm and therefore the symptoms of spinal cord ischemia were 3/18 (16.7%), transection 1/12 (8.4%), chronic dissection detected early in their onset. Insertion of a cerebrospinal 1/9 (11.1%), and intramural hematoma 1/1 (100%). fluid drain fully reversed the neurological deficit in four The median time interval between endovascular inter- patients. vention and 30-day death was 7 days (IQR, 3-20). Four Secondary procedures. Thirteen patients (11.8%) re- deaths were related to aortic rupture; one in a patient with quired secondary procedures (Table IV), the majority for a mycotic , one in a patient with an athero- management of type I endoleak: 3 patients required con- sclerotic aneurysm where it was not possible to secure a version to open repair; 1 for infolding of the device, 1 for proximal seal, and two in patients with aortic dissection. In proximal extension of the aortic dissection, and 1 early in the latter two cases, the procedures were uncomplicated the series when a large enough device to provide an ade- but on the third postoperative day the patients died of quate seal was not commercially available; 4 patients re- rupture of the infra-renal aorta, confirmed at post-mortem quired extension cuffs, 1 proximal, 2 distal, and 1 patient examination. Three patients died with a stroke and four required both; 2 patients required open aortic banding due JOURNAL OF VASCULAR SURGERY Volume 54, Number 6 Clough et al 1583

Table V. Five out of seven (71%) of the late deaths that were aortic related occurred in patients who presented with mycotic pathology

Cause of late death Months post Presenting Aortic primary Patient pathology related Other operation

1 Mycotic Yes 2 2 Aneurysm Pneumonia 2 3 Mycotic Yes 2 4 Mycotic Yes 5 5 Acute dissection Unknown 10 6 Aneurysm Pneumonia 13 7 Aneurysm Stroke 17 8 Aneurysm Unknown 20 9 Aneurysm Unknown 20 10 Aneurysm Yes 35 11 Acute dissection Unknown 36 12 Acute dissection Unknown 38 13 Aneurysm Yes 39 14 Aneurysm Unknown 52 15 Aneurysm Lung cancer 52 16 Acute dissection Unknown 55 17 Acute dissection Unknown 62 18 Mycotic Yes 62 Fig 1. Kaplan-Meier survival analysis. 19 Mycotic Yes 64 20 Chronic dissection Unknown 83 21 Acute dissection Unknown 111 (2) 1 year: presence of an aortic fistula (OR, 9.78; 95% 22 Transection Unknown 132 CI, 1.51-63.50; P ϭ .017), perioperative stroke (OR, 5.87; 95% CI, 1.05-32.92; P ϭ .044), and use of general anes- thetic (OR, 3.76; 95% CI, 1.13-12.49; P ϭ .031). to the proximity of the landing zone to the coeliac artery (3) 5 years: presence of an aortic fistula (OR, 12.31; and left common carotid arteries, respectively. Three pa- 95% CI, 1.69-89.47; P ϭ .013) and increasing age (OR, tients who presented with evidence of aortic infection 1.06; 95% CI, 1.01-1.10; P ϭ .021). suffered recurrent infection and required further stent graft placement. One patient suffered a myocardial infarction DISCUSSION resulting in systemic hypotension 6 weeks following the Endovascular repair is an attractive alternative to con- procedure and presented with paraplegia that completely ventional open surgery for treatment of AAS. There are resolved with cerebrospinal fluid drainage. currently no randomized controlled trials comparing endo- Follow-up outcomes. The median follow-up of pa- vascular with open surgery for AAS and so the evidence for tients who were alive at 30 days was 36 months (IQR, efficacy has been derived from large multicenter registries 11.8-62.9). There were 22 late deaths that occurred from such as EUROSTAR, the UK Thoracic Registry, and month 2 to 132 (Table V). The majority of patients re- IRAD.16,17 Single center cohort studies have evaluated turned to their local area for follow-up and therefore the individual diseases within AAS and have shown better out- cause of death was known in only 50% of cases. Seven come compared to open surgery with regard to ruptured (31.8%) of the late deaths were aortic related, and five descending thoracic aneurysm, complicated type B aortic occurred in patients who had initially presented with an dissection, and aortic transection.18-23 infected aneurysm. Kaplan-Meier analysis demonstrated This study has shown that endoluminal repair of pa- a 1-year survival of 81% and the 5-year survival was 63% tients presenting with AAS is associated with a 30-day (Fig 1). mortality of 12.7%, and this compares favorably with the Predictors of mortality. The results of univariate lo- results from open surgery where mortality is in the region of gistic regression analysis are shown in Table VI. Significant 25% to 45%.11,12,24-26 The benefit of endovascular repair is independent predictors of death at 30 days, 1 year, and 5 maintained at midterm follow-up with survival figures of years demonstrated using multivariate logistic regression 81% and 63% at 1 and 5 years, respectively. Multivariate analysis are shown below. logistic regression analysis of patient and procedural factors (1) 30 days: hypotension (odds ratio [OR], 5.25; 95% related to the repair has demonstrated that hypotension, confidence interval [CI], 1.20-22.95; P ϭ .027), use of general anesthesia, a long procedure duration, and old age general anesthetic (OR, 5.23; 95% CI, 1.34-20.49; P ϭ are independent predictors of 30-day mortality. Hypoten- .017), long procedure duration (OR, 2.03; 95% CI, 1.05- sion and old age are known to be predictors of perioperative 3.91; P ϭ .035), and increasing age (OR, 1.07; 95% CI, mortality in patients undergoing aortic intervention, and 1.01-1.13; P ϭ .026). unstable patients are more likely to undergo general anes- JOURNAL OF VASCULAR SURGERY 1584 Clough et al December 2011

Table VI. Predictors of death at 30 days, 1 year, and 5 years using univariate logistic regression analysis

Results of univariate logistic regression analysis 30 days 1 year 5 years Variable entered into regression model Sig OR 95% CI Sig OR 95% CI Sig OR 95% CI

Age (years) 0.057 1.048 0.999-1.100 0.146 1.207 0.991-1.065 0.012a 1.044 1.010-1.079 Female sex 0.363 1.700 0.542-5.328 0.311 1.688 0.613-4.648 0.173 1.821 0.770-4.310 Coronary artery disease 0.784 1.212 0.306-4.795 0.717 0.781 0.205-2.972 0.007a 3.905 1.451-0.506 Lung disease 0.475 1.538 0.472-5.015 0.884 0.920 0.301-2.812 0.012a 3.100 1.277-7.525 Renal disease 0.368 2.167 0.403-11.658 0.663 1.445 0.276-7.566 0.693 1.339 0.313-5.723 Hypotension 0.130 2.446 0.769-7.778 0.536 1.405 0.479-4.120 0.639 1.248 0.495-3.148 Rupture 0.453 1.547 0.495-4.835 0.769 1.167 0.418-3.258 0.509 1.336 0.565-3.158 Aortic fistula 0.752 1.431 0.155-13.227 0.045a 5.625 1.042-30.373 0.048a 5.852 1.014-33.767 Leg ischemia 0.647 0.608 0.072-5.112 0.437 1.750 0.426-7.181 0.077 3.000 0.887-10.149 General anesthetic 0.034a 3.545 1.100-11.432 0.140 2.118 0.782-5.740 0.785 0.887 0.375-2.099 Operation duration 0.017a 1.784 1.108-2.873 0.005a 2.141 1.256-3.648 0.006a 2.083 1.231-3.525 (hours) Intraoperative blood loss Ͼ500 mL 0.271 2.050 0.572-7.353 0.094 2.604 0.848-1.998 0.061 2.603 0.955-7.096 LSCA coverage 0.886 1.085 0.354-3.326 0.357 1.599 0.589-4.336 0.386 1.444 0.625-3.317 LCCA coverage 0.374 1.778 0.501-6.313 0.158 2.221 0.734-6.720 0.042a 2.738 1.037-7.230 Perioperative stroke 0.042a 5.073 1.064-24.184 0.019a 5.933 1.337-26.326 0.035a 5.000 1.117-22.377 Spinal drain insertion for paraplegia 0.646 1.467 0.286-7.513 0.122 2.833 0.757-10.60 0.259 2.033 0.593-6.966 Endoleak 0.349 0.367 0.045-2.993 0.475 0.566 0.119-2.697 0.251 1.856 0.646-5.330 Reintervention bb b 0.316 0.342 0.042-2.784 0.475 1.538 0.472-5.015

CI, Confidence interval; OR, odds ratio. aP Յ .05. bNo deaths in this group.

thesia as there is insufficient time to establish an adequate which only two of the six patients with anterior circulation loco-regional block.27 A long procedure time indicates a stroke had atherosclerotic as their primary pa- more complex repair, associated with a higher rate of thology. Two patients had a posterior circulation infarction reintervention and further hospital admissions affecting with intentional occlusion of the left subclavian artery long-term outcome. At 1 year, perioperative stroke in- without revascularization. The Society for Vascular Surgery creases the chance of death by almost sixfold. However, the and the majority of reviews and cohort studies now support presence of an aortic fistula demonstrated the greatest revascularization prior to planned coverage.32-35 However, strength of association of all the variables analyzed at both in the acute setting of an unstable patient with aortic 1 and 5 years. Endovascular treatment of infected aortic rupture this is not always possible. pathology has previously been suggested as a temporizing The study extends over 12 years, and the follow-up measure, and this series suggests that in the presence of an protocol has developed over that time in accordance with aortic fistula, endovascular treatment should form a bridge changes in imaging technology and published clinical data. to definitive open repair.28-30 However, in this series most Initially the patients underwent CT imaging prior to dis- of the patients who presented with infected aneurysms and charge and at 3, 6, and 12 months, and then annually fistulae were deemed unfit for surgery. Those who were fit thereafter. Due to concerns over radiation levels, this was declined open surgery when it was offered. reduced to a discharge CT scan if there were any concerns The risk of neurologic complications may be increased regarding the repair, otherwise a CT scan at 3 months and in the emergency setting because the hypovolemic state of then annually thereafter. Tertiary referrals had their scans the patient may contribute to cerebral and spinal cord performed locally by the referring physician and these were hypoperfusion. In this series, the rate of stroke and perma- then sent for discussion at our multidisciplinary meeting. nent paraplegia were 7.3% and 2.7%, respectively, both of A reason for concern about the endovascular manage- which compare favorably with conventional surgical treat- ment of AAS is the occurrence of graft-related complica- ment.12,25 The stroke rate in particular is encouraging as tions during follow-up. Early endovascular devices were 58% of patients required placement of the stent graft in the too rigid to track and conform to the aortic arch and eight aortic arch to achieve an adequate proximal seal. Previous percent of the patients in this series had type I endoleak, series have documented an association between aneurysmal half of which required conversion to open repair. The pathology of the arch and embolization during catheter introduction of compliant devices and the development of and wire manipulation resulting in anterior circulation disease specific endografts may reduce this in stroke.31 However, this is not substantiated in this study in the future. JOURNAL OF VASCULAR SURGERY Volume 54, Number 6 Clough et al 1585

Fig 2. Aneurysm, dissection, penetrating ulcer, intramural hematoma, and transection are grouped together because in the acute setting these aortic conditions all require early identification, transfer to an appropriate care facility, rapid institution of medical therapy, and urgent aortic repair.

High rates of reintervention were also seen with regard have been used as a surrogate marker for quality and to aortic infection. Multivariate analysis demonstrated a correlate with lower mortality and morbidity after vascular strong association between aortic infection and mortality, procedures. The concept of an acute aortic center has many dying early in the series due to multisystem organ therefore been proposed to streamline treatment thereby failure. Those that survived the initial period continued to further improving results in this complex group of patients. have ongoing problems with infection and constituted 71% Limitations. The data presented in the manuscript of late aortic-related deaths. Vigilant surveillance was im- represent over 12 years of a single centers’ experience in the portant as aortic-related deaths continued to occur during management of patients presenting with AAS. The tertiary the follow-up period. nature of the referrals may introduce a bias with case Internationally, the management of aortic disease has selection, and the expertise of a specialist referral center may changed during this study and now endovascular therapy is not be applicable to less experienced centers. The follow-up an established treatment for disease of the thoracic aorta. of patients from tertiary centers was limited by patient Thoracic endovascular treatment was introduced in our compliance. The present study details, to our knowledge, unit in 1997. Prior to this, all thoracic aortic procedures the largest series to date of endovascular repair of AAS. were performed by open surgical repair. Initially a small However, a sample size of 110 may be too small to inves- number of patients who had symptomatic aneurysms and tigate patient and procedural factors that predict mortality severe comorbidity were referred for endovascular treat- and result in type II error during interpretation of the ment. The results were good, and now endovascular treat- results. We chose not to correct for multiple testing to ment and appropriate blood pressure management is estab- avoid type II error but accept that this may have resulted in lished in our center as the first line treatment for all aortic type I error. Finally, although our analyses have demon- pathology of the arch and descending thoracic aorta. strated significant independent predictors of mortality, Despite the diversity of the pathology within acute these findings may have been affected by unknown con- aortic syndrome, these patients are at risk of imminent founders or under-reporting of events. aortic rupture and early diagnosis and treatment improves survival. Acute care pathways introduced for myocardial CONCLUSION infarction and stroke have successfully improved outcome. Endovascular repair of acute aortic syndrome provides A similar algorithm for acute aortic syndrome should facil- a viable alternative to conventional open surgery and is itate rapid initiation of definitive treatment in an appropri- associated with encouraging results. However, it is associ- ately certified facility (Fig 2). Hospital and surgeon volume ated with a significant number of secondary procedures. JOURNAL OF VASCULAR SURGERY 1586 Clough et al December 2011

The high number of aortic-related deaths during follow-up vascular aneurysm repair with the Zenith TX2 endovascular graft: 1-year emphasizes the importance of vigilant clinical and imaging results. J Vasc Surg 2008;47:247-57; discussion 257. surveillance. 15. Buth J, Harris PL, Hobo R, van Eps R, Cuypers P, Duijm L, et al. Neurologic complications associated with endovascular repair of tho- racic aortic pathology: Incidence and risk factors. A study from the AUTHOR CONTRIBUTIONS European Collaborators on stent/Graft Techniques for Aortic Aneu- Conception and design: RC, KM, RB, HZ, MW, TC, PT rysm Repair (Eurostar) registry. J Vasc Surg 2007;46:1103-10; discus- Analysis and interpretation: RC, KM, OL, TC, PT sion 1110-1. 16. Leurs LJ, Bell R, Degrieck Y, Thomas S, Hobo R, Lundbom J. Data collection: RC, OL, RB, HZ, MW, TC, PT Endovascular treatment of thoracic aortic diseases: combined experi- Writing the article: RC, KM, OL, TC, PT ence from the Eurostar and United Kingdom Thoracic endograft Critical revision of the article: RC, KM, OL, RB, HZ, MW, registries. J Vasc Surg 2004;40:670-9; discussion 679-80. TC, PT 17. Tsai TT, Trimarchi S, Nienaber CA. Acute aortic dissection: perspec- Final approval of the article: RC, KM, OL, RB, HZ, MW, tives from the International Registry of Acute Aortic Dissection (IRAD). Eur J Vasc Endovasc Surg 2009;37:149-59. TC, PT 18. Walsh SR, Tang TY, Sadat U, Naik J, Gaunt ME, Boyle JR, et al. Statistical analysis: KM, OL Endovascular stenting versus open surgery for thoracic aortic disease: Obtained funding: RC, KM, OL systematic review and meta-analysis of perioperative results. J Vasc Surg Overall responsibility: RC 2008;47:1094-8. 19. Fattori R, Tsai TT, Myrmel T, Evangelista A, Cooper JV, Trimarchi S, et al. Complicated acute type B dissection: is surgery still the best REFERENCES option? A report from the International Registry of Acute Aortic Dissection. JACC Cardiovasc Interv 2008;1:395-402. 1. Alpert JS, Thygesen K, Antman E, Bassand JP. Myocardial infarction 20. Xenos ES, Abedi NN, Davenport DL, Minion DJ, Hamdallah O, Sorial redefined–a consensus document of The Joint European Society of EE, et al. Meta-analysis of endovascular vs open repair for traumatic Cardiology/American College of Cardiology Committee for the redef- descending thoracic aortic rupture. J Vasc Surg 2008;48:1343-51. inition of myocardial infarction. J Am Coll Cardiol 2000;36:959-69. 21. Xenos ES, Minion DJ, Davenport DL, Hamdallah O, Abedi NN, Sorial 2. Amsterdam EA, Kirk JD, Bluemke DA, Diercks D, Farkouh ME, EE, et al. Endovascular versus open repair for descending thoracic aortic Garvey JL, et al. Testing of low-risk patients presenting to the emer- rupture: institutional experience and meta-analysis. Eur J Cardiothorac gency department with chest pain: a scientific statement from the Surg 2009;35:282-6. American Heart Association. Circulation 2010;122:1756-76. 22. Greenberg RK, Lu Q, Roselli EE, Svensson LG, Moon MC, Hernandez 3. O’Connor RE, Brady W, Brooks SC, Diercks D, Egan J, Ghaem- AV, et al. Contemporary analysis of descending thoracic and thoraco- maghami C, et al. Part 10: acute coronary syndromes: 2010 American abdominal aneurysm repair: a comparison of endovascular and open Heart Association Guidelines for Cardiopulmonary Resuscitation and techniques. Circulation 2008;118:808-17. Emergency Cardiovascular Care. Circulation 2010;122(Suppl 3):S787- 23. Nienaber CA, Rousseau H, Eggebrecht H, Kische S, Fattori R, 817. 4. Ahmad F, Cheshire N, Hamady M. Acute aortic syndrome: pathology Rehders TC, et al. Randomized comparison of strategies for type B and therapeutic strategies. Postgrad Med J 2006;82:305-12. aortic dissection: the Investigation of Stent Grafts in Aortic Dissec- 5. Ince H, Nienaber CA. The concept of interventional therapy in acute tion (INSTEAD) trial. Circulation 2009;120:2519-28. aortic syndrome. J Card Surg 2002;17:135-42. 24. Minatoya K, Ogino H, Matsuda H, Sasaki H, Yagihara T, Kitamura S. 6. Coady MA, Ikonomidis JS, Cheung AT, Matsumoto AH, Dake MD, Replacement of the descending aorta: recent outcomes of open surgery Chaikof EL, et al. Surgical management of descending thoracic aortic performed with partial cardiopulmonary bypass. J Thorac Cardiovasc disease: open and endovascular approaches: a scientific statement from Surg 2008;136:431-5. the American Heart Association. Circulation 2010;121:2780-804. 25. Estrera AL, Rubenstein FS, Miller CC 3rd, Huynh TT, Letsou GV, Safi 7. Davies MG, Younes HK, Harris PW, Masud F, Croft BA, Reardon MJ, HJ. Descending : surgical approach and treat- et al. Outcomes before and after initiation of an acute aortic treatment ment using the adjuncts cerebrospinal fluid drainage and distal aortic center. J Vasc Surg 2011;52:1478-85. perfusion. Ann Thorac Surg 2001;72:481-6. 8. Holt PJ, Poloniecki JD, Gerrard D, Loftus IM, Thompson MM. 26. Schermerhorn ML, Giles KA, Hamdan AD, Dalhberg SE, Hagberg R, Meta-analysis and systematic review of the relationship between volume Pomposelli F. Population-based outcomes of open descending thoracic and outcome in abdominal aortic aneurysm surgery. Br J Surg 2007;94: aortic aneurysm repair. J Vasc Surg 2008;48:821-7. 395-403. 27. Suzuki T, Mehta RH, Ince H, Nagai R, Sakomura Y, Weber F, et al. 9. Holt PJ, Poloniecki JD, Hinchliffe RJ, Loftus IM, Thompson MM. Clinical profiles and outcomes of acute type B aortic dissection in the Model for the reconfiguration of specialized vascular services. Br J Surg current era: lessons from the International Registry of Aortic Dissection 2008;95:1469-74. (IRAD). Circulation 2003;108(Suppl 1):II312-7. 10. Lumsden AB, Reardon MJ. Once dissected always dissected! Can stent 28. Clough RE, Black SA, Lyons OT, Zayed HA, Bell RE, Carrell T, et al. grafts change the natural history of type B dissections? A report from the Is endovascular repair of mycotic aortic aneurysms a durable treatment International Registry of Acute Aortic Dissection. JACC Cardiovasc option? Eur J Vasc Endovasc Surg 2009;37:407-12. Interv 2008;1:403-4. 29. Burks JA Jr, Faries PL, Gravereaux EC, Hollier LH, Marin ML. 11. Girardi LN, Krieger KH, Altorki NK, Mack CA, Lee LY, Isom OW. Endovascular repair of bleeding aortoenteric fistulas: a 5-year experi- Ruptured descending and thoracoabdominal aortic aneurysms. Ann ence. J Vasc Surg 2001;34:1055-9. Thorac Surg 2002;74:1066-70. 30. Ting AC, Cheng SW, Ho P, Poon JT. Endovascular stent graft repair 12. Barbato JE, Kim JY, Zenati M, Abu-Hamad G, Rhee RY, Makaroun for infected thoracic aortic –a durable option? J Vasc MS, et al. Contemporary results of open repair of ruptured descending Surg 2006;44:701-5. thoracic and thoracoabdominal aortic aneurysms. J Vasc Surg 2007;45: 31. Reece TB, Gazoni LM, Cherry KJ, Peeler BB, Dake M, Matsumoto 667-76. AH, et al. Reevaluating the need for left subclavian artery revasculariza- 13. Makaroun MS, Dillavou ED, Wheatley GH, Cambria RP, Gore TAG tion with thoracic endovascular aortic repair. Ann Thorac Surg 2007; Investigators. Five-year results of endovascular treatment with the Gore 84:1201-5; discussion 1205. TAG device compared with open repair of thoracic aortic aneurysms. J 32. Holt PJ, Johnson C, Hinchliffe RJ, Morgan R, Jahingiri M, Loftus IM, Vasc Surg 2008;47:912-8. et al. Outcomes of the endovascular management of aortic arch aneu- 14. Matsumura JS, Cambria RP, Dake MD, Moore RD, Svensson LG, rysm: implications for management of the left subclavian artery. J Vasc Snyder S, et al. International controlled clinical trial of thoracic endo- Surg 2010;51:1329-38. JOURNAL OF VASCULAR SURGERY Volume 54, Number 6 Clough et al 1587

33. Rizvi AZ, Murad MH, Fairman RM, Erwin PJ, Montori VM. The effect 35. Matsumura JS, Lee WA, Mitchell RS, Farber MA, Murad MH, Lums- of left subclavian artery coverage on morbidity and mortality in patients den AB, et al. The Society for Vascular Surgery Practice Guidelines: undergoing endovascular thoracic aortic interventions: a systematic management of the left subclavian artery with thoracic endovascular review and meta-analysis. J Vasc Surg 2009;50:1159-69. aortic repair. J Vasc Surg 2009;50:1155-8. 34. Clough RE, Modarai B, Topple JA, Bell RE, Carrell TW, Zayed HA, et al. Predictors of stroke and paraplegia in thoracic aortic endovascular intervention. Eur J Vasc Endovasc Surg 2011;41:303-10. Submitted May 2, 2011; accepted Jul 1, 2011.