Endovascular treatment for thoracoabdominal aneurysms:
outcomes and results§
Michael P. Siegenthalera,*, Ernst Weigangc, Kerstin Brehmb, Wulf Euringerb,
Tobias Baumannb, Markus Uhlb, Sujatha Raghua, Friedhelm Beyersdorfb
aDivision of Cardiac Surgery, the Heart Lung and Esophageal Surgery Institute, UPMC Presbyterian, Suite C-700,
200 Lothrop St., Pittsburgh, PA 15213, United States
bCenter for Cardiovascular Disease, University of Freiburg, Freiburg, Germany
cClinic for Cardiothoracic and Vascular Surgery, University Hospital Mainz, Mainz, Germany
Received 18 September 2007; received in revised form 16 June 2008; accepted 1 July 2008
Objective: Endovascular treatment of thoracoabdominal aortic aneurysms (TAAA) in combination with selective open surgical revasculariza-
tion may be an alternative to conventional surgical repair. We analyzed our patient outcomes after elective and emergent endovascular TAAA
repair. Methods: Mortality and outcome data from 21 consecutive patients treated with endovascular TAAA repair between 2000 and 2006 were
reviewed. An integrated neuroprotective approach was used on all patients. Mortality risk estimates for open surgery (OS) were calculated using
the published risk assessment models and compared to our outcomes. Results: Of the 21 patients, 9 had acute presentation: acute pain (9),
rupture (6), and malperfusion (1). The celiac axis was overstented in 15. Nine hybrid open surgical procedures were performed: visceral/renal
arteries (5), infrarenal aorta (3) and complete arch revascularization (1). Eleven patients had previous aortic surgery. Thirty-day mortality rate
was 4.8% (1/21, predicted OS value 8.3%), 1-, 2- and 3-year survival was 80%. One hospital death occurred due to ischemic colitis after inferior
neurologic complications included one minor left pontine stroke with complete resolution, postoperative confusion (1) and saphenous nerve
injury (1). No new late endoleaks occurred after initialcomplete aneurysmexclusion.Five patients underwent early(<30 days) and four patients
underwent late endovascular reinterventions for persistent endoleak. An additional reintervention included percutaneous stenting of a superior
mesenteric artery stenosis. Actual freedom from late reintervention was 81%, and 76% at 1-, 2 and 3-year follow-up. Late major adverse events
selective visceral and renal revascularization is associated with low mortality and can only be effectively performed by a surgeon. High-risk
patients and those with acute presentation appear to benefit most from this therapy. Early results up to three years of this therapy are
encouraging, but further follow up to validate long-term results is required.
# 2008 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.
Keywords: Thoracoabdominal aneurysm; Endovascular repair; Visceral revascularization
Many patients with aneurysms of the distal arch and
descending thoracic aorta are amenable to the emerging
endovascular treatment techniques, but the role of endovas-
Open surgical management of thoracoabdominal aneur-
ysms (TAAA) involves complex operative procedures and
carries an operative mortality of 10—20% in real life series
involving registry data [1—3]. A few centers of excellence
were able to report better data [4,5].
Endovascular stent grafting for thoracoabdominal disease
has been used mainly in populations with significant
comorbidities . Theaortic segment that has to bereplaced
inopen surgerydoes notstrictlycorrespond withthesegment
that has to be covered with an endovascular stent graft. A
stent graft can be landed distally at the visceral vessels or
even covering the celiac axis in a large aorta using a 46 mm
stent graft without endovascular or surgical revasculariza-
tion of the visceral or renal arteries. In conventional open
cases, an aorta of that size needs replacement. If complete
branched or fenestrated endografts or a hybrid open surgical
revascularization is required. These approaches are both
European Journal of Cardio-thoracic Surgery 34 (2008) 810—819
§Presented at the 21st Annual Meeting of the European Association for
Cardio-thoracic Surgery, Geneva, Switzerland, September 16—19, 2007.
* Corresponding author. Address: The Heart Lung and Esophageal Surgery
Institute, UPMC Presbyterian, Suite C-700, 200 Lothrop Street, Pittsburgh, PA
15213, United States. Tel.: +1 412 648 6648; fax: +1 412 802 8020.
E-mail address: firstname.lastname@example.org (M.P. Siegenthaler).
1010-7940/$ — see front matter # 2008 European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved.
feasible but technically challenging. The branched and
fenestrated endograft treatment is getting a lot of attention
and early results are encouraging . This treatment is
currently only available in elective cases, as the current
grafts have to be custom made. In theory, even if a hybrid
revascularization is required in addition to endovascular
therapy, the significant morbidity of a large conventional
open thoracoabdominal procedure, which includes aortic
cross-clamping and mechanical pump support, can be
reduced to the morbidity of a large abdominal operation
without aortic clamping and the endovascular manipulation
of a stent graft with visceral organ ischemia times of 10—
15 min .
We report our series of 21 patients who underwent
stage hybrid open endovascular procedure. We compare the
observed mortality with the predicted mortality risk scores
derived from published risk assessment models for surgical
replacement of the thoracoabdominal aorta [4,8—10].
2. Materials and methods
2.1. Study design
We performed a retrospective review of the mortality and
adverse events from a series of 21 patients who underwent
endovascular repair of TAAA between 2000 and 2006. In this
period, 92 patients with thoracic aortic pathology were
treated with stent graft interventions by our group of
cardiovascular surgeons. The ethics committee of the
University of Freiburg approved this retrospective data
collection. All patients undergoing endovascular aortic stent
grafting were identified. Outcome data were retrospectively
collected from hospital records. Follow-up data were
obtained from outpatient clinic notes and from phone calls
to either the patient’s primary care physician or to the
patient himself. Twenty of the 21 patients of this series had
their TAAA repair performed by or under the supervision of
the first author.
Of the 21 TAAA patients, mortality and paraplegia risk was
assessed using the published risk scores illustrated in Table 1
[4,8—10]. These scores were derived from a very large open
surgical experience and allow the calculation of a predicted
mortality of an entire cohort as well as individual patient’s
risk assessment for either all adverse events, for paraplegia
or for the likelihood of death. As the urgency of patients’
presentation could be argued in all cases but in frank
ruptures, wecalculated theriskscoresthatinclude allurgent
and elective cases as well as the scores exclusively for
elective patients and applied them to our entire cohort. All
clinical outcomes were recorded and compared to these
2.2. Definition of thoracoabdominal aneurysms
The definition of TAAAwas based on the definition used by
a large series of a center of excellence , using the
Crawford extent of the aneurysms. The definition for the
Crawford extent by this group involved the recording of the
segments of the thoracoabdominal aorta that was surgically
replaced. We used the same definition and classified the
Crawford extent of the TAAA based on the aortic segment
that would need replacement in an open surgical procedure.
We identified 21 patients with TAAA who were treated
with thoracic stent grafts, in which, for an open surgical
procedure, visceral aortic replacement would have been
2.3. Patient selection
During the study period, all patients with TAAAs who
needed treatment were considered candidates for an
endovascular and possibly hybrid approach. Only young
patients with connective tissue disease or those with
anatomy unsuitable for endovascular repair were treated
with open surgery during the study by the same surgeons who
perform the endovascular procedures. Unsuitable anatomy
for stent grafting included the absence of suitable landing
zones for stent graft therapy due to length of the landing
zone or their size. Proximal landing zones had to be ?15 mm
in length and distal landing zones ?10 mm. Vascular access
did not influence our decision to use stent grafts or open
surgery, as in cases with poor vascular access either the iliac
artery or the aorta was used for stent graft implantation.
During thestudy period,24patientsunderwent elective open
surgical TAAA repair. The result of 20 patients in this cohort
has been previously published by our group .
2.4. Perioperative management and spinal cord-
Preoperative evaluation for patients was similar to
patients undergoing open surgical repair. Cardiac catheter-
M.P. Siegenthaler et al./European Journal of Cardio-thoracic Surgery 34 (2008) 810—819
Risk score models
Predicted group mortality Expected mortality = E1 + (E1[(renal + rupture)/n]) E1 = (symptoms ? 0.01) +
(rupture ? 0.1) + (renal ? 0.07) + (C2 ? 0.06)
Expected # of deficits = E1 + (E1 ? 0.3[C2/(C1 + C2 + C3 + C4)]) E1 = (C1 ? 0.01) +
(C2 ? 0.06) + (C3 ? 0.01) + (C4 ? 0.003) + (acute ? 0.2)
Risk = odds/(1 + odds); odds = (C2 ? 0.7219) + (renal ? 1.1730) + (age ? 0,0535) +
(symptoms ? 0.9051)
Risk = odds/(1 + odds); odds = (exp[(diabetes ? 1.0327) + (C2 ? 0.9932) ? 3.7704]
Risk = odds/(1 + odds); odds = [(C2 ? 0.8945) + (renal ? 1.2612) + (age ? 0.0272) +
(symptoms ? 0.5504) ? 4.6597]
Predicted group paraplegia 
Predicted individual mortality 
Predicted individual paraplegia 
Predicted adverse events after elective surgery 
Risk scores — see refs. [4,8,9].
 Cowan Jr JA, Dimick JB, Henke PK, Huber TS, Stanley JC, Upchurch Jr GR.
Surgical treatment of intact thoracoabdominal aortic aneurysms in the
 LeMaire SA, Miller III CC, Conklin LD, Schmittling ZC, Coselli JS.
Estimating group mortality and paraplegia rates after thoracoa-
bdominal aortic aneurysm repair. Ann Thorac Surg 2003;75(2):
 Svensson LG, Crawford ES, Hess KR, Coselli JS, Safi HJ. Experience with
1509 patients undergoing thoracoabdominal aortic operations. J Vasc
Surg 1993;17(2):357—68. discussion 368—70.
 Lee WA, Brown MP, Martin TD, Seeger JM, Huber TS. Early results after
staged hybrid repair of thoracoabdominal aortic aneurysms. J Am Coll
 Torsello G, Can A, Umscheid T, Tessarek J. Hybrid thoracoabdominal
aneurysm repair with simultaneous antegrade visceral revascularization
and supra-aortic debranching from the ascending aorta. J Endovasc Ther
 Coselli JS, LeMaire SA, Miller III CC, Schmittling ZC, Koksoy C, Pagan J,
Curling PE. Mortality and paraplegia after thoracoabdominal aortic
aneurysm repair: a risk factor analysis. Ann Thorac Surg 2000;69(2):
 LeMaire SA, Miller III CC, Conklin LD, Schmittling ZC, Koksoy C, Coselli JS.
A new predictive model for adverse outcomes after elective thora-
coabdominal aortic aneurysm repair. Ann Thorac Surg 2001;71(4):
 Coselli JS, LeMaire SA, Conklin LD, Koksoy C, Schmittling ZC. Morbidity
and mortality after extent II thoracoabdominal aortic aneurysm repair.
Ann Thorac Surg 2002;73(4):1107—15. discussion 1115—6.
 Weigang E, Sircar R, von Samson P, Hartert M, Siegenthaler MP, Luehr M,
Richter H, Szabo G, Czerny M, Zentner J, Beyersdorf F. Efficacy and
frequency of cerebrospinal fluid drainage in operative management of
thoracoabdominal aortic aneurysms. Thorac Cardiovasc Surg 2007;55(2):
 Weigang E, Hartert M, Siegenthaler MP, Beckmann NA, Sircar R, Szabo G,
Etz CD, Luehr M, von Samson P, Beyersdorf F. Perioperative management
to improve neurologic outcome in thoracic or thoracoabdominal aortic
stent-grafting. Ann Thorac Surg 2006;82(5):1679—87.
 Weigang E, Hartert M, Siegenthaler MP, Pitzer-Hartert K, Luehr M, Sircar
R, von Samson P, Beyersdorf F. Neurophysiological monitoring during
thoracoabdominal aortic endovascular stent graft implantation. Eur J
Cardiothorac Surg 2006;29(3):392—6.
 Weigang E, Luehr M, Harloff A, Euringer W, Etz CD, Szabo G, Beyersdorf F,
Siegenthaler MP. Incidence of neurological complications following over-
stenting of the left subclavian artery. Eur J Cardiothorac Surg
 Grunkemeier GL, Anderson RP, Starr A. Actuarial and actual analysis
of surgical results: empirical validation. Ann Thorac Surg 2001;71(6):
 Roselli EE, Greenberg RK, Pfaff K, Francis C, Svensson LG, Lytle BW.
Endovascular treatment of thoracoabdominal aortic aneurysms. J Thorac
Cardiovasc Surg 2007;133(6):1474—82.
 Panneton JM, Hollier LH. Nondissecting thoracoabdominal aortic aneur-
ysms: part I. Ann Vasc Surg 1995;9(5):503—14.
 Etz CD, Luehr M, Kari FA, Bodian CA, Smego D, Plestis KA, Griepp RB.
Paraplegia after extensive thoracic and thoracoabdominal aortic aneur-
ysm repair: does critical spinal cord ischemia occur postoperatively? J
Thorac Cardiovasc Surg 2008;135(2):324—30.
 Maniar HS, Sundt III TM, Prasad SM, Chu CM, Camillo CJ, Moon MR, Rubin
BG, Sicard GA. Delayed paraplegia after thoracic and thoracoabdominal
aneurysm repair: a continuing risk. Ann Thorac Surg 2003;75(1):113—9.
 Resch TA, Greenberg RK, Lyden SP, Clair DG, Krajewski L, Kashyap VS,
O’Neill S, Svensson LG, Lytle B, Ouriel K. Combined staged procedures for
the treatment of thoracoabdominal aneurysms. J Endovasc Ther
 Safi HJ, Estrera AL, Azizzadeh A, Coogan S, Miller III CC. Progress and
future challenges in thoracoabdominal aortic aneurysm management.
World J Surg 2008;32(3):355—60.
Appendix A. Conference discussion
Dr R. Griepp (New York, NY): My question is, in how many of these patients
were you able to preserve both hypogastric or internal iliac arteries, because
of course we think those are very important to spinal cord perfusion?
this series and did not occlude any internal iliac arteries. I actually had one
case where we did an abdominal stent after a thoracic endograft. One of the
iliac limbs was too long and covered a hypogastric artery, and the patient
immediately lost his motor evoked potentials. So we used a very aggressive
therapy algorithm that Dr Weigang had implemented at our clinic: We lowered
the CSF pressure and raised the mean arterial blood pressure really high.
Subsequently the MEPs recovered within a short time. But I think it is very
crucial that the hypogastric arteries stay open.
Dr Griepp: I agree.
Dr H. Shennib (Phoenix, AZ): This is a really challenging group of patients,
and the issues always with treating thoracoabdominal aneurysms is the length
of the graft and the landing zones. You are always worried that if you extend
and you keeponcovering moreof theaorta yourisk spinal nerve injury,and the
secondissueis usually where doyou land thegraft and howdoyou preservethe
important visceral branch vessels.
Now, you mentioned that you have actually covered the celiacs in some of
those patients and got away with it, but you didn’t really tell us about the
landing zone in those patients. Were you sparing the SMAs, recognizing that
they are involved or not involved in the aneurysmal disease? How do you select
your patients and how do you select your landing zone in patients with more
than celiac artery occlusion? And how often do you resort to doing a
debranching procedure with those patients?
The second question I have has to do with the type of graft that you put in.
Now, I see that those are Medtronic, and I am not sure whether it is a Talent or a
Valiant graft, and the other one is a Gore graft. And I am assuming that because
Dr Siegenthaler: No, the Gore graft cannot be reversed but the Medtronic
graft can be reversed on the OR table.
need to reverse it, particularly with the Valiant graft. I just wanted to get your
opinion as to how do you choose your graft, and have you actually resorted at
any time of putting the graft in from above, from the subclavian or axillary?
Dr Siegenthaler: To the Valiant graft, you can order the graft reversed and
thecompanydeliversittoyou.But asI showedinthetalk, alargeproportionof
these patients came with ruptures or urgent presentations. In such situations,
it takes too long to order grafts. So what you can do with the Valiant stent
graft, youcan deploy it onthe OR back table and reverseit yourself. This works
very well and nicely. So if you have a stack of stents in different sizes for
emergencies ready and you have one of those cases where you can’t wait, this
can be a very nice option to choose the approach to reverse it. It cannot be
done with the Gore because of the different deployment system.
Intermsofthe stent graft landing zone, Ithink we broke pretty much every
rule in this series that you read about. In many cases of this series our landing
zones were significantly shorter than 2 cm. Since the results of open
conventional surgery in these often poly-morbid patients really aren’t that
great, we will try to do the case with a short landing zone, and we will explain
to all the patients that there is a significant risk that there might be an
endoleak and that there will be a possibility that ultimately they will have to
undergo a large operative procedure.
So this basically also answers to your next question: We did a debranching
procedure only when it was absolutely necessary. If we felt that there was a
possibility to get away leaving that usually slightly more narrow segment of a
thoracoabdominal aneurysm at the visceral arteries intact and seal it on both
sides ordo an infrarenal replacement in addition to a very large stent graft just
at the level of the celiac artery, we chose that approach in order to avoid a
And then in terms of your last question about antegrade stent deployment,
I only have used this for arch debranching cases. Otherwise with the Valiant
system, which has a very nice deployment system and is pretty stiff, you can
overcome even severe kinking at the thoracoabdominal junction and you can
get up here and land the stent graft where you want to have it.
M.P. Siegenthaler et al./European Journal of Cardio-thoracic Surgery 34 (2008) 810—819