Evolving Approaches to Thoracoabdominal Aortic Aneurysm Repair: Open Surgical and Endovascular Treatment Using Fenestrated/Branched Endografts

Article Citation:

Jesse Manunga and Timothy M. Sullivan (2017) Evolving Approaches to Thoracoabdominal Aortic Aneurysm Repair: Open Surgical and Endovascular Treatment Using Fenestrated/Branched Endografts. Journal of the Minneapolis Heart Institute Foundation: January 2017, Vol. 1, No. 1, pp. 59-64.

Case Report

Jesse Manunga, MD and Timothy M. Sullivan, MD

Department of Vascular, Endovascular Surgery, Minneapolis Heart Institute at Abbott Northwestern Hospital, Minneapolis, MN

Address for correspondence:
Jesse Manunga, MD
Minneapolis Heart Institute at Abbott Northwestern Hospital
920 East 28th Street, Suite 300
Minneapolis, Minnesota 55407
Tel: 612-863-6800

Email: Jesse.Manunga@allina.com


Open repair remains the standard of care for the treatment of patients with thoracoabdominal aortic aneurysms. However, due to the magnitude of the operation and a typically older patient population with extensive comorbidities, less invasive methods of repair have received increasing acceptance—despite lack of long-term data on durability when compared with open surgical repair. One of the most dreaded complications of both open and endovascular thoracoabdominal aortic aneurysms repair (for patient and surgeon alike) remains the development of paraplegia, a complication that can be minimized with the techniques described herein.

Keywords: thoracoabdominal aneurysm, open repair, endovascular repair, fenestrated endografts, branched endografts


Until recently, patients with aneurysms of the thoracic and thoracoabdominal aorta had only one treatment option: open surgical repair. For those patients who could not tolerate operation because of medical comorbidities, continued aneurysm enlargement and eventual rupture was a constant, yet unpredictable threat to their lives. Several studies have documented improved survival rates in those patients treated surgically.1,2 Despite advances in surgical reconstruction and organ protection, the mortality rate for elective repair of thoracoabdominal aortic aneurysm (TAAA) ranges from 4% to 21%; advanced age, renal failure, and postoperative paraplegia are the most important risk factors predicting mortality at 30 days. In addition, for those patients age ≥79 years with an emergency presentation, history of diabetes or congestive heart failure, 30-day mortality is 50%. For aneurysms isolated to the distal thoracic aorta, the risk of paraplegia is 0% to 4% and is dependent on the extent of aorta replaced.3 A substantial number of patients surviving the operation have prolonged, complicated courses secondary to renal, cardiac, and pulmonary dysfunction. Perhaps the most devastating complication of these complex procedures is paraplegia.4

A myriad of techniques have been developed to protect the spinal cord during open surgical repair of the thoracic and thoracoabdominal aorta, including “clamp and sew,” distal aortic and visceral perfusion, complete cardiopulmonary bypass, profound hypothermia and circulatory arrest, direct spinal cord cooling, cerebral spinal fluid (CSF) drainage, and the use of pharmacologic adjuncts; some of these principles may be useful in preventing paraplegia at the time of endovascular repair. When the thoracic aorta is cross-clamped, spinal arterial perfusion pressure decreases while CSF pressure increases, resulting in decreased perfusion pressure.

In an important study of 1,004 patients by Safi et al,5 an immediate postoperative neurologic deficit occurred in 6.8% of patients operated without the adjuncts of CSF drainage and distal aortic perfusion, while only 2.4% of those operated with adjuncts suffered this devastating complication. These authors also stressed the importance of reimplantation of intercostal arteries during open repair, especially in the vulnerable area between T9 and T12, which frequently gives rise to the anterior spinal artery. Relative hypertension in the immediate post-operative period (maintaining mean arterial pressure between 90 and 100 mm Hg) is also advocated. Other risk factors for paraplegia in their series included the extent of aneurysmal disease, advanced age, emergency presentation, preoperative renal dysfunction, active smoking, and cerebrovascular disease. Additionally, delayed paraplegia has been noted as late as 2 weeks following surgery, and has been successfully treated by placement of a spinal drain.6,7

The argument for endovascular TAAA repair comes from the excellent outcomes of patients treated for infrarenal abdominal aortic aneurysm (AAA).8–10 Based on these findings, endovascular TAAA repair using either fenestration or directional branches is increasingly gaining acceptance in the medical community. The Achilles heel of this repair remains the long-term patency of target vessels and the devastating complication of spinal cord ischemia. For this reason, many surgeons rely on adjunct maneuvers such as a spinal drain, perioperative blood pressure manipulations, and intraoperative neuromonitoring to reduce the risk of paraplegia.

Perhaps the most important decision a surgeon faces is whether to perform open surgical or endovascular repair and which techniques to utilize to maximize patient safety and reduce complications. The following presentations illustrate several options for treatment. Figure 1 demonstrates the classification of TAAAs, based on work done by Crawford et al. and adopted by the vascular community1011 This classification is based on the extent of aorta involved: extent I involves most of the thoracic aorta; extent II involves the proximal descending aorta to below the renal arteries; extent III involves the distal descending aorta and the abdominal aorta; and Extent IV involves most of the abdominal aorta and typically requires thoracoabdominal exposure for open repair.

Classification of TAAA. (Reprinted with permission)


Patient 1

A 61-year-old male presented with progressive enlargement of the thoracoabdominal aorta extending from the mid-distal thoracic aorta to the aortic bifurcation (Crawford extent III), now measuring 6 cm in maximum diameter (Figure 2). His past history was significant for having had open repair (with a synthetic graft) of a right popliteal artery aneurysm at another institution and an endovascular repair of a left popliteal artery aneurysm at our institution. Physical examination was normal, with the exception of a palpable, pulsatile nontender mass in the epigastrium. A former smoker, he was otherwise healthy, and had a normal cardiac stress test and pulmonary function studies. Because of his relatively young age and the uncertainties surrounding durability of endovascular repair, he elected to proceed with open surgical repair.

Aneurysm of the thoracoabdominal aorta (measuring 6 cm), involving the celiac, superior mesenteric and renal arteries.

The patient was taken to the operating room, and after induction of general anesthesia, a spinal drain was placed in order to minimize the risk of spinal cord ischemia. The aorta and its visceral branches were exposed via a thoracoretroperitoneal approach. In order to limit the duration and extent of visceral artery ischemia, a “quad” graft was constructed to allow individual branch grafts to each of the visceral arteries, a technique pioneered by Ballard et al.12 Utilizing a partial aortic cross-clamp, an end-to-side anastomosis was performed to the thoracic aorta above the aneurysm, and subsequently, individual grafts were placed in sequence to the celiac, superior mesenteric, and left renal arteries. Ischemic time for each visceral vessel was <10 minutes. Subsequently, after cross-clamping the thoracic aorta below the quad graft (thereby maintaining visceral perfusion) and gaining control of the iliac arteries, the aneurysm was opened, the right renal artery revascularized, and the aneurysmal aorta replaced with a “tube” graft.

The patient was observed in the intensive care unit (ICU) for 36 hours, and the spinal drain was removed. He suffered no neurologic complications, was transferred to the inpatient vascular unit, and was discharged home on postoperative day 5. He continues to do well 6 years post-aneurysm repair, without recurrence of aneurysm or graft stenosis (Figure 3).

Computed tomography angiography following open surgical repair of TAAA using “Ballard” technique. Note “quad” graft originating from the thoracic aorta with separate branches to the celiac, superior mesenteric, right and left renal arteries.

Patient 2

A 71-year-old female with an enlarging Crawford extent I TAAA was offered an open repair at another institution but sought a minimally invasive approach. Her aneurysm measured 6.8 cm in the greatest diameter. Cardiac risk factors (CRF) included coronary artery disease (CAD), hypertension (HTN), hyperlipidemia, alcoholic hepatitis, and alcohol dependence. Her physical examination was otherwise unremarkable. She underwent a staged repair of her TAAA. The first stage consisted of a left subclavian to left carotid artery transposition followed by percutaneous endovascular coverage of the thoracic aorta, landing the endografts just above the celiac axis. This was followed by the final stage of visceral artery incorporation using a fenestrated stent graft 6 weeks later. The patient was admitted to the ICU following both operations for hemodynamic monitoring and spinal drain management. During both hospital visits, her spinal drain was removed 24 hours after the completion of the operation. She spent 2 days in the ICU and 1 day on the general vascular floor prior to discharge home. The patient has been followed for over 15 months and post-procedure computed tomography (CT) angiography of the chest, abdomen, and pelvis demonstrated a decrease in the aneurysm sac size, patency of target vessels, and no endoleak (Figure 4A).

A. Postoperative CTA reconstruction of patient 2 (staged repair), showing exclusion of the aneurysm from the left carotid artery to the aortic bifurcation with widely patent target vessels. B.Postoperative CTA reconstruction of patient 3 (single stage with no spinal drain or neuromonitoring) showing patent target vessels and excluded aneurysm. Patient presented with a history of “complete stent graft explantation”. However, note the presence of the old endografts in the common iliac arteries. C. Postoperative CTA reconstruction of patient 4 (previous history of an elephant trunk, treated in a single stage with spinal drain and neuromonitoring), all target vessels are widely patent. The patient is status post right nephrectomy for renal cell carcinoma.

Patient 3

An 81-year-old male with a 6.9-cm Crawford extent IV TAAA was referred to us for a second opinion. The patient had undergone endovascular aneurysm repair for infrarenal AAA 9 years prior, but developed a type IA endoleak that was treated with stent graft explantation, all at outside institutions. A subsequent CT scan revealed continued growth of the aneurysm that now extended above the visceral vessels. His CRF included HTN, hyperlipidemia, CAD, chronic kidney disease (CKD) stage III, and advanced age. His exam was significant for a pulsatile abdominal mass and absence of pedal pulses with good femoral pulses. The patient underwent a single stage, endovascular repair with incorporation of all 4 visceral arteries (celiac axis, superior mesenteric artery, and both renal arteries) using a fenestrated stent graft. He was discharged home on postoperative day 2.Patient 4

An 81-year-old male with a history of prior ascending and arch aneurysm repair with an “elephant trunk” presented with an enlarging 8.3-cm Crawford extend III TAAA. His CRF included CAD, HTN, hyperlipidemia, CKD stage III, chronic obstructive pulmonary disease, obesity, and renal cell carcinoma. He was status post right nephrectomy. The patient appeared older than his stated age and had a pulsatile abdominal mass. His pulses were palpable at every level checked. Because of the large aneurysm size, he was treated in a single stage with a spinal drain; motor-evoked potentials (MEP) and somatosensory-evoked potentials (SSEP) were monitored. He was discharged home on postoperative day 3 with no complications.


Patients 3 and 4 have been followed up for 8 months now. Computed tomography angiography of the chest, abdomen, and pelvis obtained prior to discharge revealed no evidence of endoleak. Subsequent CT angiography revealed shrinking of aneurysm sac, with widely patent target vessels (Figure 4B,C).

The technical details of endovascular repair of complex aortic aneurysms using fenestrated stent grafts have been described.13 In our practice, all patients with extent I, II, and III TAAA routinely undergo spinal drain placement during endovascular repair. Patients with extent IV TAAA are only selectively drained. We routinely monitor MEP and SSEP for patients with Crawford extent I, II, and III TAAA undergoing an endovascular single stage repair.

For all repairs, we liberally consider preadmission, especially in patients with complex anatomy or multiple comorbidities. Furthermore, preadmission affords us the ability to place a spinal drain the day before the operation to avoid delays on the day of the surgery. Vasodilator antihypertensive medications are either discontinued or dose decreased a full week prior to repair. They are slowly introduced back a week following repair. Mean arterial pressure is maintained above 80 mm Hg throughout the operation and 48 to 72 hours postoperatively. Patients are liberally transfused, especially during the first 48 hours to maintain hemoglobin above 10 mg/dL and an international normalized ratio <1.5. The spinal drain is kept in place for 24 to 72 hours at a baseline pressure of 10 mm Hg. In the event of changes in neurologic status, the CSF pressure is decreased between 0 to 5 mm Hg.

In single-stage patients, neuromonitoring is used to guide adjunct maneuvers necessary to reducing risk of paraplegia. We define a significant change as a greater than 75% decrease in MEP or SSEP amplitude compared with baseline. In patients with decline in MEP and/or SSEP, MAP is increased to >90 mm Hg, CSF pressure decreased to 0 to 5 mm Hg, and perfusion to the lower extremity is restored as quick as possible by removing large femoral sheaths. These maneuvers usually result in return of MEP and SSEP to baseline. In the rare case of continued suppression of monitoring signals, one of the fenestrations (usually the celiac fenestration) is not bridged with a covered stent to allow for continued perfusion of the spinal cord. Alternatively, all fenestrations are lined up with bridging stent grafts, the bifurcated device is placed in the usual fashion but the gate is not cannulated, leaving a large endoleak.


Even with significant advances in surgical and anesthetic techniques, morbidity and mortality associated with open repair of TAAA remains high. This is particularly true for patients with advanced age and multiple comorbid conditions.14,15 Spinal cord ischemia resulting in paraplegia or paraparesis is one of the most feared complications in patients undergoing open and endovascular TAAA. For this reason, we liberally place a spinal drain in all patients undergoing such repairs. The decision to only selectively drain patients with Crawford extent IV TAAA is based on a history of prior aortic surgery and patency of the left subclavian, left vertebral, and hypogastric arteries. We are less likely to drain a patient with a robust anterior spinal artery and widely patent hypogastric arteries.

We prefer staged endovascular repair of all patients with extent I, II, or III TAAA to minimize risk of paraplegia. In this case, the first stage usually consists of coverage of the thoracic aorta to just above the celiac axes with or without left subclavian bypass, followed 6 to 8 weeks later by completion of repair with a fenestrated stent graft. In patients with large aneurysms, where the risk of rupture is prohibitively high, single-stage repair with MEP and SSEP monitoring is entertained. A decrease in MEP and SEEP amplitude of >75% triggers an increase in the MAP to >90 mm Hg, decrease in CSF pressure to 0 to 5 mm Hg, and a timely removal of the large femoral sheath to allow perfusion of the hypogastric arteries and lower extremities. Patients with extent IV and V TAAA are only selectively drained and endovascularly treated in a single stage.

Much has been written about branches versus fenestrations as the best device configuration for the endovascular repair of TAAA. Some centers have used branched devices exclusively while others have used fenestrations routinely.16–18 Others utilize devices with a combination of branches and fenestrations. Published series have shown that fenestrations have an extremely high patency in the renal arteries while longer branches tend to kink over time as the aneurysm sac shrinks. The best configuration is likely patient specific and contains a combination of branches and fenestrations depending of geometric trajectory of visceral arteries targeted. All 3 patients treated with fenestrated devices and discussed in this report were discharged home within 2 to 3 days. None of them required discharge to a rehabilitation center. All patients were followed 8 to 15 months, with CT showing a decreasing aneurysm sac and no need for any secondary intervention.


Open repair remains the standard of care for treatment of TAAA, and can be performed safely and with durable results in many patients. Over the last few years, however, data is emerging in support of endovascular repair for TAAA, especially in high-risk patients. Published reports continue to document excellent long-term target vessel patency. Spinal cord injury is still considered the “Achilles heel” of this type of repair. This complication can be significantly reduced with the placement of a spinal drain and strongly considering a staged approach to these procedures. When a single approach is indicated, the use of neuromonitoring can help trigger maneuvers necessary in reducing spinal cord injury.


1.Bickerstaff LK, Pairolero PC, Hollier LH, et al. Thoracic aortic aneurysms: a population-based study. Surgery. 1982; 92:1103–1108. [Google Scholar]
2.Safi HJ, Miller CC 3rd, Huynh TT, et al. Distal aortic perfusion and cerebrospinal fluid drainage for thoracoabdominal and descending thoracic aortic repair: ten years of organ protection. Ann Surg. 2003;238:372–380; discussion, 380–381. [Google Scholar]
3.Coselli JS, LeMaire SA, Conklin LD, Adams GJ. Left heart bypass during descending thoracic aortic aneurysm repair does not reduce the incidence of paraplegia. Ann Thorac Surg. 2004;77:1298–1303; discussion 303. [Crossref] [Google Scholar]
4.Huynh TT, Miller CC 3rd, Estrera AL, Sheinbaum R, Allen SJ, Safi HJ. Determinants of hospital length of stay after thoracoabdominal aortic aneurysm repair. J Vasc Surg. 2002;35:648–653. [Crossref] [Google Scholar]
5.Safi HJ, Bartoli S, Hess KR, et al. Neurologic deficit in patients at high risk with thoracoabdominal aortic aneurysms: the role of cerebral spinal fluid drainage and distal aortic perfusion. J Vasc Surg. 1994;20:434–444; discussion, 442–443. [Crossref] [Google Scholar]
6.Safi HJ, Miller CC 3rd, Azizzadeh A, Iliopoulos DC. Observations on delayed neurologic deficit after thoracoabdominal aortic aneurysm repair. J Vasc Surg. 1997;26:616–622. [Crossref] [Google Scholar]
7.Rizvi AZ, Sullivan TM. Incidence, prevention and management in spinal cord protection during TEVAR. J Vasc Surg. 2010;52:86S–90S. [Crossref] [Google Scholar]
8.Sicard GA, Zwolak RM, Sidawy AN, et al. Endovascular abdominal aortic aneurysms repair: Long-term outcome measures in patients at high-risk for open surgery. J Vasc Surg. 2006;44:229–236. [Crossref] [Google Scholar]
9.Bush RL, Johnson ML, Hedayati N, Henderson WG, Lin PH, Lumsden AB. Performance of endovascular aortic aneurysm repair in high-risk patients: results from the Veterans Affairs National Surgical Quality Improvement Program. J Vasc Surg. 2007;45:227–233. [Crossref] [Google Scholar]
10.Crawford ES, Crawford JL, Safi HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients. J Vasc Surg1986;3:389–404. [Crossref] [Google Scholar]
10.Chaikof EL, Fillinger MF, Matsumura JS, et al. Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg. 2002;35:1061–1066. [Crossref] [Google Scholar]
11.Coselli JS, Bozinovski J, LeMaire SA. Open surgical repair of 2286 thoracoabdominal aortic aneurysms. Ann Thor Surg. 2007;83:S862–S864. [Crossref] [Google Scholar]
12.Ballard JL, Abou-Zamzam AM Jr, Teruya TH. Type III and type IV thoracoabdominal aortic aneurysm repair: results of a trifurcated/two-graft technique. J Vasc Surg. 2002;36:211–216. [Crossref] [Google Scholar]
13.Manunga J, Titus J. Regional anesthesia as the anesthetic of choice for high-risk surgical patients undergoing repair of juxtarenal aortic aneurysm with fenestrated stent grafts. J Vasc Surg. 2016;5214:1261–1267. [Google Scholar]
14.Timaran CH, Veith FJ, Rosero EB, et al. Endovascular aortic aneurysm repair in patients with the highest risk and in-hospital mortality in the United States. Arch Surg. 2007;142:520–524. [Crossref] [Google Scholar]
15.Coselli JS, LeMaire SA, Miller CC 3rd, et al. Mortality and paraplegia after thoracoabdominal aortic aneurysm repair: a risk factor analysis. Ann Thorac Surg. 2000;69:409–414. [Crossref] [Google Scholar]
16.Verhoeven EL, Tielliu IF, Bos WT, Zeebregts CJ. Present and future of branched stent grafts in thoraco-abdominal aortic aneurysm repair: a single-centre experience. Eur J Vasc Endovasc Surg. 2009;38:155–161. [Crossref] [Google Scholar]
17.Chuter TAM, Rapp JM, Hiramota JS, Schneider DB, Howel B, Reilly LM. Endovascular treatment of thoracoabdominal aortic aneurysms. J Vasc Surg. 2008;47:5–16. [Crossref] [Google Scholar]
18.Bicknell CD, Cheshire NJW, Riga CV, Bourke P, et al. Treatment of complex aneurysmal disease with fenestrated and branched stent grafts. Eur J Vasc Endovasc Surg. 2009;37:175–181. [Crossref] [Google Scholar]