Early Outcomes of Thoracic Endovascular Stent-Graft Repair for Acute Complicated Type B Dissection Using the Gore TAG Endoprosthesis
Article Outline
We assessed the technical success and early outcome of thoracic endovascular aortic repair (TEVAR) for complicated acute type B thoracic aortic dissection treated at a single institution using a commercially available device. All patients with symptomatic complicated acute type B thoracic aortic dissection treated with TEVAR since Food and Drug Administration approval of the Gore (Flagstaff, AZ) TAG® endoprosthesis were identified from a prospectively maintained vascular registry. Clinical indications, operative technique, perioperative complications, follow-up imaging, and mortality were analyzed. Between March 2005 and November 2007, 127 TEVARs using the TAG endoprosthesis were performed, of which 15 (11.8%) were for complicated acute type B thoracic aortic dissection. Indications for repair were malperfusion (53%), persistent pain (27%), and primary aortic failure (33%). Technical feasibility and success with deployment proximal to the entry tear was 93.3%, requiring at least partial coverage of the left subclavian artery in seven (46.7%). Adjunctive procedures required at the time of TEVAR included renal stent (n = 2), iliac stent (n = 3), and access-artery open repair (n = 2). Twelve patients (80%) had immediate resolution of the malperfusion deficit. Major perioperative complications included paraplegia (13.3%), renal failure requiring hemodialysis (13.3%), and stroke (6.7%). Perioperative mortality was 13.3%, occurring in one patient presenting with rupture and one with profound heart failure on admission. For complicated acute type B thoracic aortic dissection, TEVAR using commercially available stent grafts showed high technical success, excellent results at resolving malperfusion, and acceptably low complications and perioperative mortality.
Introduction
Debate continues over the proper management of acute type B aortic dissection. Traditional surgical replacement of the descending thoracic aorta for acute dissection carries a mortality rate of 25-46%.1 As a result, thoracic stent grafting is rapidly becoming the alternative option for complicated type B dissections. Multiple reports have documented the ability of stent grafts to exclude the entry flap in the majority of type B dissections.2, 3, 4, 5, 6 The major perceived advantages of stent-graft placement are stabilization of the intimal flap, preservation of the true lumen, positive remodeling of the aorta through thrombosis of the false lumen, and ultimately freedom from dissection-related complications during follow-up. While the durability of this modality is not yet known, early evidence suggests that stent-graft therapy may be equivocal to best medical therapy in regard to acute mortality.4 The purpose of this study was to evaluate the outcome of stent grafting for complicated acute type B dissection at a single institution since Federal Drug Administration (FDA) approval of a commercially available thoracic endoprosthesis.
Methods
Patients
A prospectively maintained vascular registry was reviewed to identify patients treated at our institution with the Gore (Flagstaff, AZ) TAG® endoprosthesis for thoracic aortic pathology, after FDA approval, from March 2005 through November 2007. Institutional review board approval was obtained to retrospectively study patients treated with the Gore TAG device for acute type B aortic dissection. Included were those presenting with acute type B thoracic aortic dissections (<14 days since presentation), all of whom had an entry flap at or distal to the left subclavian artery confirmed by computed tomography angiogram (CTA) and/or intravascular ultrasound (IVUS). All patients had complicated type B dissections defined by end organ or extremity malperfusion, persistent pain, acute heart failure, spinal cord paraplegia, renal failure, mesenteric ischemia, or refractory hypertension. Data collected included age, survival status, graft components utilized, access arteries, length of aorta covered, subclavian artery coverage, blood loss, length of stay, complications, follow-up outcomes, and status of false lumen and largest aortic diameter on follow-up imaging.
Technique
The structural design of the Gore TAG endoprosthesis is well described.7 Thoracic endograft procedures were performed in a vascular endosuite utilizing ceiling-mounted fluoroscopy (Figure 1). General anesthesia was utilized in all cases, and no prophylactic lumbar drains were placed. All stent grafts were deployed through the common femoral artery approach—via unilateral and/or bilateral femoral access; no conduits were necessary. IVUS was used to confirm entry into the true lumen, localization of the entry flap, and sizing measurements in all patients (Fig. 1). Once true lumen wire access was confirmed in the aortic arch, arch angiography was performed. If imaging indicated a need for left subclavian artery coverage (Fig. 2), selective vertebral angiography was performed to evaluate the adequacy of collateral vertebral artery flow and to confirm the absence of internal mammary grafts. Patients were then systemically heparinized with a 100 UI/kg intravenous bolus. Stent grafts were delivered under fluoroscopic guidance to cover the proximal entry flap with intent to exclude false lumen flow. Additional distal components were deployed by the surgeon when the dissection flap did not appear stabilized adequately on live fluoroscopy to improve distal perfusion. Completion angiography was performed to document the status of false lumen and to evaluate distal perfusion. When indicated, ancillary procedures including proximal extension of thoracic stent graft, renal artery stenting, and iliac artery stenting were performed to correct malperfusion. Direct arteriotomy repair was undertaken in all cutdowns, and closure devices were used for percutaneous access. All patients were transferred to the surgical intensive care unit for postoperative care.
Fig. 1
Sequence demonstrates successful exclusion of false lumen with preservation of arch vessels after proximal extension of stent graft. Also note improved flow through true lumen after stabilization of dissection with stent graft.
Fig. 2
A Preimplant arch angiogram with robust right vertebral artery and no internal mammary to coronary graft. B Magnified view demonstrating subclavian artery coverage with maintenance of antegrade left carotid flow despite struts extending into orifice.
Routine spinal drainage and mean arterial pressure (MAP) management are not performed at our institution. All patients were transferred to an intensive care setting for frequent neurovascular checks and monitoring. In conjunction with the Department of Neurosurgery, our policy is for expectant management of spinal cord ischemia. Success of expectant management with delayed spinal drainage has been reported in several series of open and endovascular thoracic aortic reconstructions.8, 9, 10, 11 Any patient with a change in exam has vasoactive agents initiated for MAP >110, immediate neurosurgical evaluation, immediate spinal magnetic resonance imaging, and spinal drainage at the discretion of the neurosurgery attending.
Statistical Analysis
This is a retrospective, noncomparative analysis. All statistics were descriptive on an intent-to-treat basis. Discrete variables are presented as percentages, continuous as counts, means, and standard deviations. Social security records were reviewed to evaluate patients lost to follow-up and generate life-table analysis.
Results
One hundred twenty-seven endograft repairs were performed for thoracic aortic pathology including aneurysm, dissection, and penetrating ulcer. A total of 26 patients had dissection as an indication for intervention; of these, 15 (11.8%) were for acute type B aortic dissection. The mean age at time of treatment was 61 ± 11.5 years (range 45-87), with a gender distribution of 12 males and three females. Indication for intervention included malperfusion in 53%, persistent pain in 27%, and aortic failure in 33%, with two patients having multiple indications (Table I). All patients were hypertensive on presentation, with 14/15 (93.3%) requiring intensive care unit monitoring and continuous IV infusion of antihypertensive agents.
Table I. Indications for repair
| Indicationa | n |
|---|---|
| Malperfusion (53%) | 8 |
| Lower extremity ischemia | 5 |
| Paraplegia | 2 |
| Refractory hypertension and encephalopathy | 1 |
| Renal artery malperfusion | 2 |
| Pain (27%) | 4 |
| Persistent chest/abdominal pain | 3 |
| Flap progression | 1 |
| Aortic failure (33%) | 5 |
| Contained rupture | 2 |
| Rapid expansion | 2 |
| Acute true lumen collapse and heart failure | 1 |
aTwo patients had multiple indications. |
Technical success was 93.3% as defined by exclusion of proximal entry flap on completion angiogram. Table II summarizes implant characteristics. Successful deployment required at least partial coverage of the left subclavian artery in 7/15 (46%) patients. Graft sizes ranged 31-40 mm, and the average length of aorta covered was 23.2 ± 5.9 cm (range 15-37). Delivery was successful via the femoral approach in all 15 patients, requiring bilateral femoral cutdown in 20%, unilateral cutdown alone in 20%, and unilateral cutdown with contralateral percutaneous access in 60%. Access-related complication occurred in only one patient (6.7%), requiring iliofemoral interposition grafting. Estimated blood loss was 318.8 ± 533.8 cc (range 75-2,000). Average length of stay was 8.9 ± 9.9 days (range 2-19). Based on angiographic appearance and patient symptoms, 12/15 (80%) patients had immediate resolution of the malperfusion deficit which prompted intervention. One symptomatic failure was a known type I endoleak on completion angiography despite complete coverage up to the left carotid artery, and two were due to inability to return flow to renal arteries resulting in persistent hypertension.
Table II. Key operative variables
| Patient | Procedure description | Subclavian coverage | Successful graft deployment | Resolution of malperfusion | Length of aorta covered (cm) | Length of time of implant (months) |
|---|---|---|---|---|---|---|
| 1 | TAG 34 mm x 10 cm proximal; 31 mm x 10 cm distal thoracic aorta; graft deployed with gap between components | N | Y | Y | 20 | 19.56 |
| 2 | TAG 31 mm x 15 cm (2 overlapping components) | N | Y | Y | 25 | 19.56 |
| 3 | TAG 31 mm x 15 cm proximal, 31 mm x 10 cm distal | N | Y | N | 25 | 17.13 |
| 4 | TAG 31 mm x 15 cm proximal, 34 mm x 20 cm distal | N | Y | Y | 25 | 16.34 |
| 5 | TAG 34 mm x 15 cm with 34 mm x 10 cm proximal extension | Y | Y | Y | 17 | 15.16 |
| 6 | TAG 37 mm x 20 cm | Y | Y | Y | 20 | 15.06 |
| 7 | TAG 31 mm x 15 cm with 34 mm x 20 cm proximal extension | N | Y | Y | 25 | 12.07 |
| 8 | TAG 31 mm x 15 cm, right renal stent 6 mm x 18 mm x2, left CIA stent 9 mm x 10 cm; left CFA 8 mm PTFE interposition graft | N | Y | Y | 15 | 11.61 |
| 9 | TAG 31 mm x 15 cm (2 overlapping components) | N | Y | Y | 25 | 11.47 |
| 10 | TAG 34 mm x 20 cm | Y | Y | Y | 20 | 1.81 |
| 11 | TAG 34 mm x 15 cm, left renal stent 7 mm x 18 cm | N | Y | Y | 15 | 6.44 |
| 12 | TAG 37 mm x 20 cm (2 overlapping components) | Y | Y | Y | 32 | 6.18 |
| 13 | TAG 34 mm x 20 cm distal, 34 mm x 10 cm proximal extension | Y | Y | N | 22 | 0.59 |
| 14 | TAG 40 mm x 20 cm distal, 40 mm x 20 cm middle, 40 mm x 15 cm proximal extension | Y | N | N | 37 | 2.56 |
| 15 | TAG 34 mm x 15 cm (2 overlapping components), kissing iliac stents 14 mm x 6 cm | Y | Y | Y | 25 | 0.07 |
| 7/15 Covered LSC | 14/15 | 12/15 | Mean 23.2 ± 5.9 | Mean 10.4 ± 6.9 |
Overall 30-day complications occurred in 7/15 (46.7%) patients (Table II). New-onset spinal cord ischemia developed in 2/15 (13.3%) patients, requiring subsequent placement of lumbar drain and elevation of MAP ≥110 mm Hg per our institution protocol. One patient who developed postoperative paraplegia died before resolution of symptoms; the other patient had improvement of paraplegia to ambulation with assistive device but continued bowel and bladder incontinence. Two patients presented with ischemia-related paraplegia preoperatively (pulseless extremities): one expired after a prolonged intensive care unit course and the other improved to normal motor function by 1 month postoperatively. One patient suffered a watershed cerebral infarct detected on magnetic resonance imaging due to mental status changes postoperatively with mild cognitive and no motor deficit. Other complications included deep venous thrombosis (n = 1), chronic renal insufficiency from contrast nephropathy (n = 1), and persistent hypoperfusion of the left kidney with refractory hypertension (n = 1). Two patients required dialysis postoperatively: one presented with renal dysfunction and required permanent dialysis and the other was liberated from dialysis by postoperative day 9.
Perioperative 30-day mortality was 13.3%. One patient who presented with a contained thoracic aortic rupture died on postoperative day 18 after a prolonged course including hemodialysis, ventilator dependence, and visceral ischemia. The second 30-day mortality occurred in a patient who presented with significant true lumen compromise leading to acute heart failure (Fig. 3). Despite a postprocedure angiogram demonstrating exclusion of the proximal entry flap and initial improvement in cardiac output, the patient died on postoperative day 2, likely as a result of retrograde dissection into the aortic root. One additional patient expired 2 months after her procedure of respiratory failure in a rehabilitation facility, where she was being treated for mild lower extremity paraparesis. The overall mortality in this cohort was 3/15 (20%) at an average of 11.0 ± 6.9 months postoperation (range 0.1-19.6). By life-table analysis, mortality was 79% at 2 years (standard error 10%).
Fig. 3
IVUS image demonstrating severe compromise of true lumen in patient presenting with acute heart failure 2 hr after onset of type B dissection.
Average clinic and radiographic follow-up was 5.8 ± 4.4 months (range 0-12). At 1 month, 8/10 patients had thrombosis of the false lumen across the treated segment of the thoracic aorta by CTA. One patient without thrombosis of the false lumen had a known proximal type I endoleak at completion of operation with aortic size increase by 3 mm during this interval. The second patient developed a new type I endoleak and subsequently underwent successful placement of a proximal TAG component to cover the subclavian artery. CT scan after this reintervention demonstrated a thrombosed false lumen and stabilization in aortic diameter. Of the grafts followed at 6 months, 7/7 had thrombosis of false lumen and 6/7 had no change (within 2 mm) in maximum aortic size. At 12 months, 3/3 had thrombosis of false lumen and 2/3 no change in maximum aortic size. One patient continues to increase with 7 mm growth of maximum aortic size despite thrombosed false lumen. Overall, two patients (13.3%) have undergone reintervention, one for type I endoleak and one for subsequent aneurysmal degeneration of uncovered abdominal aorta (Table III).
Table III. Complications/reinterventions
| n | Resolution | |
|---|---|---|
| Complications (n = 8 in 7 patients, 46.7%) | ||
| New-onset paraplegia | 2 | 1 expired with deficit,a 1 persistent cauda equina syndrome |
| HD | 2 | 1 expired on HD,a 1 off HD by POD 9 |
| Cerebral infarct | 1 | Mild cognitive deficit, no motor compromise |
| Deep venous thrombosis | 1 | Resolved after 3 months' coumadin therapy |
| Chronic renal insufficiency | 1 | Off dialysis and stable Cr at 3 months |
| Refractory HTN | 1 | Decreased left renal perfusion on NM scan at 3 months |
| Reinterventions (n = 2, 13.3%) | ||
| Proximal/distal TAG | 1 | Resolved type I leak and thrombosed false lumen at 12 months, improved flow celiac/left renal |
| Debranching and endograft | 1 | Successful exclusion of type IV TAAA |
aSame patient. |
Discussion
Acute aortic dissection has long been a significant cause of aorta-related mortality, with operative mortality rates for descending thoracic aortic replacement as high as 45% in 1986 and without much progress evidenced by a mortality of 31.4% in the more contemporary International Registry of Acute Aortic Dissection (IRAD) in 2000.12, 13 With mortality rates significantly lower in patients undergoing medical treatment, nonoperative treatment has long been the mainstay of therapy for uncomplicated type B aortic dissection. However, this has resulted in a selection bias toward medical management as patients requiring surgery were those with complicated dissection, e.g., visceral or extremity ischemia, retrograde arch dissection, acute true lumen collapse, persistent pain, and refractory hypertension. Understandably, patients with more complicated dissections would be expected to have worse outcomes independent of the inherently difficult technical challenge of operating on an acutely dissected aorta.
Medical therapy, however, also has significant morbidity and mortality. Not only did the IRAD study demonstrate an in-hospital mortality of 10.7% for patients treated medically with acute type B dissection,13 but the long-term sequelae of dissection are not benign. Onitsuka et al.14 demonstrated a 13% rate of conversion to surgery in a group of 76 patients randomized to medical therapy and a 23% incidence of dissection-related events including rupture and sudden death in the medical cohort studied out to 10 years. Others have reported a rupture risk of 18% during follow-up of medically treated type B dissection15 and that 20% of all thoracoabdominal aneurysms have chronic dissection as the etiology.1 The overall 3-year mortality of patients with type B dissection in the IRAD database who survived initial hospitalization was 24.9%.16 In fact, in the interim analysis of the IRAD data, patients with the best 1- and 3-year survival are those who survive open surgical intervention during their initial presentation.17
Numerous factors have been evaluated for improved outcome in patients with type B dissection. Successful surgical correction in the acute phase likely leads to improved long-term survival via obliteration of the false lumen while restoring normal perfusion to the distal aortic segment and stabilizing aortic size. Akutsu et al.18 demonstrated that in medically managed type B dissections followed over 10 years, patent false lumen was a significant predictor of dissection-related death (hazard ratio [HR] = 5.6) and dissection-related event (HR = 7.6). Mean aortic size and aortic growth rate have been demonstrated as independent predictors of dissection-related events during follow-up of medically treated patients.14 These observations support the potential benefit of excluding the entry point of the aortic dissection with stent grafts. In a nonrandomized cohort, Dialetto et al.2 demonstrated a thrombosed false lumen rate of 75% in stent-grafted patients vs. 10.7% in medically treated patients. The same cohort showed aneurysmal degeneration in only 3.5% of stent-grafted patients vs. 28.5% in the medically treated group.
Stent-graft technology is quickly replacing open graft replacement as the treatment of choice for complicated type B dissection. Ideally, stent grafts should offer patients improved in-hospital mortality compared to traditional repair, while maintaining the survival benefit demonstrated by surgical correction.17 Much of the early published experience included custom stent grafts and use of abdominal graft components for the treatment of complicated type B dissection. The current study is limited to thoracic aortic dissection treated with the TAG endoprosthesis, representing contemporary outcomes of a commercially available device. Although FDA approved for thoracic aortic aneurysm, the TAG endoprosthesis has been used increasingly for treatment of aortic dissection.
Our 93.3% success in excluding the proximal entry tear is commensurate with other published reports of 86-95%,5, 6, 19, 20 although there was an increased need for intentional left subclavian artery coverage at 46%. As demonstrated by others, this is well tolerated in patients when adequate evaluation of the vertebral arteries and internal mammary bypass grafts is undertaken.21, 22 A low threshold for left subclavian coverage is attributable to our preference that complete exclusion of the entry flap is essential for adequate treatment of type B dissection. Fortunately, we had no adverse events related to upper extremity ischemia or vertebrobasilar symptoms.
The intent of operation in all patients was stabilization of the true lumen and correction of malperfusion. Improved flow through the true lumen was noted in each case, with stabilization of dynamic obstruction in visceral segments demonstrated on subtraction fluoroscopy. We were able to successfully treat cases of static obstruction in visceral and extremity segments found on completion angiography via endovascular stenting in all but one instance. Technically, this was aided by the use of femoral artery exposure in cases of dissection into the iliac system, which allows for easier identification of the true lumen. We did not need to utilize the brachial approach in our cohort.
The perioperative 30-day mortality rate of 13.3% in this study is similar to published reports ranging 3.2-10.7%.2, 4, 5, 6 This mortality rate corresponds to reported mortality of medically treated patients in historic controls and is better than the expected open surgical mortality for these patients. Effectively, endovascular techniques have shifted these patients from the higher mortality associated with complicated dissection treated surgically to a lower mortality commensurate with uncomplicated dissection.
In this study, significant complications of permanent dialysis, neurological deficit, and stroke occurred in 20%. A meta-analysis of stent grafts for type B dissection demonstrated a major complication rate of 11.2%.4 Nathanson et al.20 demonstrated a 38% overall complication rate when investigating this device under trial in patients with type B dissection. Although the reported spinal cord ischemia rate is higher in this study than other series, the sample size lacks sufficient numbers to reach significance. Three patients with documented postoperative paraplegia (including one with preoperative paraplegia) had coverage of the left subclavian artery. This has been suggested to exclude collateral circulation to the intercostal artery distributions and spinal arteries.23 The one patient with preoperative paraplegia who resolved postoperatively did not undergo coverage of the left subclavian. Length of aorta covered was variable in these patients (20-32 cm); thus, it is unclear if this was a contributing factor. Postoperative hypotension often contributes to delayed neurological deficit.8, 10, 11 The unstable nature of symptomatic acute type B dissections may create an increased risk for postoperative hemodynamic instability and neurological deficit. Ultimately, spinal cord ischemia is multifactorial and we have not advocated preemptive lumbar drainage or carotid-subclavian revascularization due to the urgent need to treat.
During follow-up, persistent exclusion of the entry flap has demonstrated consistent thrombosis of the thoracic false lumen in our patients, but reduction of the overall aortic size during the relatively short interval was not observed. Nathanson et al.20 also noted no reduction in aortic size despite absence of endoleak in 71% of patients, bringing into question the fate of the uncovered abdominal aorta as demonstrated by a 56% rate of increased abdominal false lumen diameter. In our cohort, reintervention was required in two patients for abdominal aortic pathology at less than 1 year after treatment for acute type B dissection. Postoperative scans showed persistent false lumen flow in the perivisceral aorta of all three patients evaluated at 12 months, and this area continues to be a concern with stent-graft treatment of the thoracic aorta only.
In short-term analysis, commercially available stents grafts can be deployed with a high rate of success in excluding the entry flap into acute type B dissections. Further, in the acute, complicated setting, stent grafts can lower perioperative mortality to levels commensurate with patients treated medically for uncomplicated dissection. Stent grafting should be the treatment of choice in patients with indication for operative repair. While positive remodeling of the thoracic segment is evident in early follow-up scanning, vigilance will be required to evaluate the outcome of uncovered segments of the aorta during follow-up. Evolution of treatment for both complicated and uncomplicated type B dissection will continue with the anticipated results of ongoing randomized trials comparing stent grafting to best medical therapy.24, 25
References
- . Acute aortic dissection. In: Rutherford RB editors. Rutherford: Vascular Surgery. Philadelphia: Elsevier Saunders; 2005;p. 1521
- Treatment of type B aortic dissection: endoluminal repair or conventional medical therapy? Eur. J Cardiothorac Surg. 2005;27:826–830
- Changes in false lumen after transluminal stent-graft placement in aortic dissections: six years' experience. Circulation. 2005;111:2951–2957
- Endovascular stent-graft placement in aortic dissection: a meta-analysis. Eur Heart J. 2006;27:489–498
- Endovascular repair of acute type B aortic dissection: early and mid-term results. J Vasc Surg. 2006;43:1090–1095
- Endovascular repair of acute type B aortic dissection: long-term follow-up of true and false lumen diameter changes. Ann Thorac Surg. 2007;83:1059–1066
- Midterm outcome in 158 consecutive Gore TAG thoracic endoprostheses: single center experience. Ann Thorac Surg. 2006;81:1570–1577
- Spinal cord ischemia after elective stent-graft repair of the thoracic aorta. J Vasc Surg. 2005;42:11–17
- Cerebrospinal fluid drainage to reverse paraplegia after endovascular thoracic aortic aneursym repair. J Endovasc Ther. 2000;7:132–135
- . Delayed onset of neurology deficit: significance and management. Semin Vasc Surg. 2000;13:340–344
- Risk of spinal cord ischemia after endograft repair of thoracic aortic aneurysms. J Vasc Surg. 2001;34:997–1003
- Early and late results following repair of dissections of the descending thoracic aorta. J Vasc Surg. 1986;3:226–237
- The International Registry of Acute Aortic Dissection (IRAD): new insights into an old disease. JAMA. 2000;283:897–903
- Long-term outcome and prognostic predictors of medically treated acute type B aortic dissections. Ann Thorac Surg. 2004;78:1268–1273
- Risk factors for rupture of chronic type B dissections. J Thorac Cardiovasc Surg. 1999;117:776–786
- Partial thrombosis of the false lumen in patients with acute type B aortic dissection. N Engl J Med. 2007;357:349–359
- Long-term survival in patients presenting with type B acute aortic dissection: insights from the International Registry of Acute Aortic Dissection. Circulation. 2006;114:2226–2231
- Effects of the patent false lumen on the long-term outcome of type B acute aortic dissection. Eur J Cardiothorac Surg. 2004;26:359–366
- Self-expandable aortic stent-grafts for treatment of descending aortic dissections. Ann Thorac Surg. 2002;73:1138–1142
- Endoluminal stent-graft stabilization for thoracic aortic dissection. J Endovasc Ther. 2005;12:354–359
- Initial experience with intentional stent-graft coverage of the subclavian artery during endovascular thoracic aortic repairs. J Endovasc Ther. 2002;9(Suppl. 2):II39–II43
- Technical strategies to expand stent-graft applicability in the aortic arch and proximal descending thoracic aorta. J Endovasc Ther. 2002;9(Suppl. 2):II32–II38
- Neurologic complications associated with endovascular repair of thoracic aortic pathology: incidence and risk factors. A study from the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) registry. J Vasc Surg. 2007;46:1103–1111
- Investigation of stent grafts in patients with type B aortic dissection: design of the instead trial—a prospective, multicenter, European randomized trial. Am Heart J. 2005;149(4):1512–1533
- Brunkwall J. ADSORB—rationale and study design of a prospective, randomized trial in acute, non-complicated aortic dissection. Invited lecture, 29th Charing Cross International Symposium, April 14-17, 2007.
PII: S0890-5096(08)00338-5
doi:10.1016/j.avsg.2008.08.035
© 2008 Annals of Vascular Surgery Inc. Published by Elsevier Inc All rights reserved.
