European Journal of Vascular & Endovascular Surgery
Volume 34, Issue 6 , Pages 666-672, December 2007

Endovascular Treatment (EVT) of Acute Traumatic Lesions of the Descending Thoracic Aorta – 7 Years' Experience

Department of Thoracic and Vascular Surgery, University Hospital Ulm, Germany

Accepted 29 June 2007. published online 23 August 2007.

Article Outline

Objectives

To present a single centers' 7-year experience in the endovascular treatment of acute traumatic lesions of the descending thoracic aorta (ATL of the DTA).

Materials & methods

Between March 1999 and December 2006, 34 consecutive acute traumatic lesions of the descending aorta (23 men, mean age 44 years) were treated endovascularly. Stentgrafts used were TAG Excluder, Zenith TX2 and Talent. In 23 patients the Left Subclavian Artery (LSA) was covered. Mean procedural duration was 20 to 75 minutes.

Results

Exclusion of the rupture site was achieved in all cases with no conversion to open surgery. Overall 30-day mortality was 8.8%. Two patients died on post operative day (pod) 1 and one on pod 22 from cranial injuries. No death or neurological deficit related to the endovascular treatment was reported. Four type I endoleaks required treatment either by balloon reexpansion (n=2) or by additional stentgraft implantation (n=2). In two patients the stentgraft collapsed totally several days postoperatively. Two patients required secondary surgical procedures (iliac access complication and revascularisation of the left subclavian artery n=1). The average follow-up was 43.8 months (1–93 months). No stentgraft related abnormality has been subsequently documented.

Conclusions

The endovascular treatment of ATL of the DTA may offer the best means of therapy in a polytrauma patient.

Keywords: Endovascular, Traumatic, Thoracic aorta, Stentgraft, Transection

 

Back to Article Outline

Introduction 

Acute traumatic lesions of the descending thoracic aorta (ATL of the DTA) are usually accompanied by a set of severe concomitant injuries.1, 2 The optimal time to repair such aortic injuries is debated.3, 4, 5, 6 Even more debatable is whether to use classical open surgical treatment or the newly evolved endovascular treatment.7, 8, 9, 10, 11, 12, 13 Immediate open surgical repair carries a high risk of mortality in unstable patients.14, 15, 16, 17, 18

A low risk procedure to seal the rupture site, even if used as bridging technique, is an attractive option. From August 1995 to March 1999 we applied EndoVascularTreatment (EVT) to 216 patients with different aortic pathologies.19, 20 Relying on the experience gained, we decided to treat all incoming cases of ATL of the DTA by the EVT approach. We applied EVT in all the patients with documented ATL of the DTA in that period, with no exception. Centers with minimal EVAR experience, though, should follow carefully the limitations of use and contraindications for each particular stent graft, as these are provided by the manufacturer of each graft.

Since March 1999 we have performed EVT in 34 consecutive cases of ATL of the DTA, 30 of them presently alive. To our knowledge this represents the largest series of its' kind reported in the English literature. In two cases a phenomenon which we call Acute Stent Compression Syndrome (ASCS) was encountered. Scattered reports have also addressed the same phenomenon in the literature21, 22, 23 and awareness seems to be the best mean of both prevention and treatment.

Back to Article Outline

Materials and Methods 

From March 1999 to December 2006 36 consecutive patients with ATL of the DTA were admitted to our department. 34 were treated endovascularly, while the other 2 died in the Emergency Room before any thoracic intervention could be initiated. The review of the files involved was approved by the Ethical Committee of our Institution. The patients were victims of road traffic or occupational accidents and of falls from a great height and were either primarily admitted or after referral from another hospital. All patients were operated on within 16 hours (mean 8.6 +/− 7.5h) of the initial injury.

Patients' age varied from 12–79 years, 23 were men and the concomitant injuries were: Head and brain 80%, abdomen 73%, pelvis 33% and extremities 67%.

Spiral Computed Tomography (CT) scans showed circular (complete transection) or semicircular descending thoracic aortic injuries in 11 and 23 patients, respectively. Uni- or bilateral hemothorax were present in all cases.

All patients admitted without appropriate imaging from a referral hospital had a CT-angiography performed in our department (Brilliance 64, Philips Medical Systems, Hamburg, Germany). Emergency whole body scanning was performed utilizing trauma adapted protocols developed in our radiology department. The protocol allows the generation of both vascular and parenchymal contrast in a single helical scan acquisition. This is achieved with a low flow pre-injection of contrast (70ml @ 2ml/s, lomeprol 400, Bracco, Milan, Italy) and subsequent high flow injection of the same agent (50ml @ 3.5ml/s) and a final saline flush (40ml @ 3ml/s). The low flow phase guarantees sufficient parenchymal contrast equivalent to a separate portal venous phase. The ensuing high-flow phase generates vascular contrast as available from early enhancing arterial phase scans. The scan is initiated with an automated ROI threshold technique (Bolus Pro, Philips Medical Systems, Hamburg, Germany) utilizing a region of interest placed in the descending thoracic aorta. Breathhold command (cessation of ventilation for intubated patients) is triggered according to Hounsfield density measurements in the ROI. A threshold of 150 HU is usually applied.

Mean aortic diameters were found in the range from 12 to 30mm, measured at the distal aortic arch between the left common carotid (LCC) and left subclavian artery (LSA) ostium. In 22 patients the lesion was found exactly opposite the LSA origin, in 11 patients it was found 10–20mm distal to the LSA ostium and in one patient the lesion was 50mm distal to the LSA ostium. Stentgrafts used, were 23 TAG Excluder (26–34mm diameter) plus one tapered 16.5 to 14mm iliac limb extension (W.L. Gore & Associates, Putzbrunn, Germany), 6 Zenith TX2 (22–34mm diameter, Cook Incorporated, Bloomington, IN, USA) and 5 Talent (20–24mm diameter, Medtronic World Medical, Sunrise, FI, USA). Three parameters determined graft selection. Availability, graft diameter required and most of all personal experience. In our center we have a pool of grafts of different sizes from two of the aforementioned companies. As mentioned above, the smallest available endoprosthesis diameter of the Gore TAG graft was 26mm. We came across cases were the aortic diameter was as small as 18mm. Even after adding a 10% oversizing to that measurement we still need a graft with an endoprosthesis diameter of 20mm, namely a graft constructed by Medtronic. Some authors believe the TAG is suited to angulated arches as it appears to be more conformable. In addition they propose that the TAG is low profile and can be inserted through smaller diameter access vessels and that the TAG is available in different diameters compared with the Talent. We have reported in 200324 the Gore Excluder to be uniquely flexible and equipped with a simple deployment mechanism. The Medtronic graft also has a very precise and delicate deployment mechanism.

In 22 patients the proximal covered edge of the stentgraft was positioned close to the distal margin of the LCC ostium with complete coverage of the LSA ostium. Free LCC perfusion was verified in all patients by completion angiography. In all other patients stentgraft deployment was close to LSA ostium, hence orthograde LSA perfusion was preserved.

All operations were performed under general anaesthesia in the surgical suite under fluoroscopy control by a C-arm image intensifier (OEC 9600 Vascular, GE Medical Systems Germany). Transesophagial echo (TEE) was additionally applied, whenever available, as an adjunct in the estimation of the correct placement of the stent in the true lumen of the vessel. 5 000IU of Heparin (Leo Pharmaceuticals) were administered once, before guide wire insertion. Stentgrafts were introduced over a super stiff Amplatz guide wire (Boston Scientific, Ratingen, Germany) through a 20, 22 or 24 F sheath. Stentgraft deployment was performed under respiratory arrest but without pharmacologically induced hypotension or cardiac arrest. After the stentgrafts were successfully deployed, a completion angiography was performed to confirm the exact position of the graft and the total exclusion and coverage of the traumatic aortic lesion. The mean duration for the entire endovascular procedure was 20 to 75 minutes. Some patients required simultaneous operations for concomitant lesions (e.g. laparotomy for spleen/liver rupture, stabilization of third grade open fractures etc.) and thus spent up to 320 minutes on the operating table. The imaging time varied between 3 and 20 minutes and the average contrast medium used was 140ml, range 60–220. The estimated blood loss was less than 500ml.

As soon as patients' condition allowed, preferably on the same day of the procedure or the day after, a control angio-CT evaluated the exclusion of the lesion and the position of the stentgraft. Additional postoperative angio-CTs were conducted on the 3rd, 6th and 12th month postoperatively and yearly thereafter. Observation time is up to 7 years.

Back to Article Outline

Results 

Complete exclusion of the rupture site was achieved primarily in all cases. No conversion to open surgery was required. No patient developed temporary or permanent neurological deficit related to the endovascular treatment, peri- or postoperatively.

Overall 30-day mortality was 8.8%. No aortic rupture related death was documented. Two patients died on pod 1 and one on pod 22 from cranial injuries. Autopsy showed an intact stent-graft repair of the thoracic aortic injury. Another patient died on pod 50, following ASCS and will be discussed separately.

Four type I endoleaks in the early post-operative period required treatment either by balloon reexpansion (n=2) or by additional stentgraft implantation (n=2). Two patients developed Acute Stentgraft Compression Syndrome (ASCS) and were treated accordingly. Another two patients required secondary surgical procedures (ilio-femoral bypass for iatrogenic external iliac artery trauma n=1 and revascularisation of the left subclavian artery n=1).

Follow up of all patients included angio-CT and simple chest X-Ray performed at 3, 6, 12 months postoperatively and yearly thereafter. The average follow-up time was 43.8 months (1–93 months). 24 patients were followed for ≥2 years, 14 for ≥5 years and of 8 for ≥7 years. (Fig. 1) No stentgraft migration, wire fractures, in stent stenoses or relevant shape changes, post traumatic aneurysms or any other stentgraft related abnormalities have been documented.

  • View full-size image.
  • Fig. 1 

    Representative images of stentgraft behaviour after 2, 5 and 7 years. Top series: stentgraft after 2 years from implantation, Middle series: after 5 years, and Bottom series: after 7 years. Note that stent configuration depends on complete (middle, bottom) or incomplete (top) aortic rupture/transection. In complete transection (bottom) oversized stentgrafts may open to full size at the level of transection with no unfavourable sequel observed so far.

Acute stent compression syndrome (ASCS) 

In two of our patients a safe coaxially centred position of the stentgraft within the aortic arch lumen could not be achieved. In both cases the stents were not passed all the way up to the vertex of the aortic arch but rather stopped at the origin of the LSA. Despite repeated balloon manoeuvres neither a smooth alignment nor improvement of the position could be achieved (Fig. 2). In the first patient, a 24 year old severely polytraumatized female, the aortic diameter was estimated from the CT to be approximately 20mm and a 26mm Zenith TX2 stentgraft was selected. Due to her extremely bad general condition further manipulations to improve the positioning of the positioned stentgraft were abandoned, after the on table completion angiography documented complete sealing of the aortic rupture site, no endoleaks and retrograde flow to the LSA. The control CT on the first postoperative day revealed an unchanged stent positioning with lack of graft attachment to the inner curvature of the aortic arch, thus forming a triangular space between the stent graft and the aortic wall. Despite the extremely severe concomitant injuries, the patient recovered within the next 10 days and was released from the ICU to the trauma ward. A second control CT 6 weeks later (day 41 postop.) documented unchanged stentgraft position and the patient was asymptomatic with no signs of malperfusion or aortic stenosis. On day 43 however she developed an acute abdomen and spinal syndrome. An emergency angio-CT was performed, confirming complete stent compression and total aortic occlusion (Fig. 3). 60 minutes later the aortic flow was restored by proximal overstenting with a 24mm Optimed Stent. She survived subsequent small bowel resection and right hemicolectomy but finally died 7 days later from diffuse intracerebral haemorrhage.

  • View full-size image.
  • Fig. 2 

    Discoaxial non-centered malposition of a 26mm Xenith-X2 in a 20mm aorta. The stentgraft at the inner curve is disattached and protrudes into the lumen. No improvement of alignment could be achieved by repeated balloon expansions of the stentgraft.

  • View full-size image.
  • Fig. 3 

    Compression of stentgraft by the systolic jet and infolding of its' inner-curvature. Large thrombus mass within the stentgraft leads to total aortic flow obstruction.

A second case of ASCS occurred to a 57 polytrauma victim. He was treated with a 28×150mm TAG Excluder for a 24mm aorta. Ten days later he developed signs of acute abdomen. Only one hour latter aortic flow was restored by balloon dilation of the proximal stentgraft segment. Continuous monitoring with TEE however revealed stentgraft recompression within the next 40 seconds. Therefore a second stentgraft (31/100mm Gore TAG Excluder) was introduced and deployed identically to the proximal position of the first graft, aiming to double the radial expanding force of the proximal end of the former stent (Fig. 4). This patient also underwent small bowel surgery and right hemicolectomy but survived. The yearly follow up CTs thereafter have revealed no stent dislocation or stent recompression up today.

Back to Article Outline

Discussion 

The percentages of on scene, in hospital, perioperative and postoperative mortality of ATL are known to most thoracic surgeons from numerous previous publications.14, 15 Many centers25, 26 worldwide have reported their experience on EVT of ATL of the DTA.

EVT of ATL of the DTA was technically successful in all of our cases. Four type I endoleaks presented at least 24hours postoperatively and were easily addressed. Only one death (1/34=2,9%) could be indirectly related to the procedure and that occurred 50 days after the initial operation. The cumulative survival rate is presented in Fig. 5. Because ASCS, as we have defined it, is potentially lethal unless treated immediately, all efforts have to be made to safely abolish it.

EVT of ATT has its own specific problems different from those already known from treating complex aneurysms and dissections, mainly due to the younger age index of the trauma victims population.

The main problems include graft size, graft bending and pulsatile movement of the aortic arch. The tight curvature of the aortic arch of young patients requires a very flexible graft. The graft tends to be pushed away from the inner curvature of the aortic arch due to the tendancy of the graft to return to its straight form. In some cases the graft may be unable to follow the sharp curving of the aortic arch. In those cases the graft configures itself in a non concentric non coaxial fashion, leaving a small portion of its inner wall protruding into the aortic lumen. This results in the graft having no contact to the aortic wall and creating a triangular space between itself and the aortic wall. The angle between the inner curvature of the arch and the stentgraft may be as much 35–45°(Fig. 2).

At end-systole both left and right ventricles are maximally contracted and decreased in size, thus pulling down the ascending aorta and bending the aortic arch even more, forcing it to assume an even sharper configuration. Stentgraft attachment force is lowest at the inner curvature of the aortic arch. During systole the bursting flow of the stroke volume may hit this weak point thus forcing the proximal end of the stentgraft which is attached to the inner curvature to infold towards the centre of the graft, causing a partial or complete occlusion of the stent graft.

During systole, the juvenile thoracic aorta dilates up to 15–20% in order to accommodate the ejected stroke volume. A stentgraft anchored to the thoracic aorta has to follow the pulsatile movement of the aortic wall and even in end-systole, when the thoracic aortic wall (and the stentgraft with it) is dilated to its' maximum, it should still possess some reserve to expand. Otherwise, during end-systole, there will not be sufficient radial force left to maintain wall attachment and anchoring. Migration or endoleak should then be expected.

On the other hand, a markedly oversized stentgraft in a sharply curved juvenile aorta could wrinkle, thus leading to type I endoleak or even worse infolding and total stentgraft collapse and obstruction. Therefore the optimal stentgraft for each patient has to be selected not only on the basis of aortic diameter and lesions' length but based on patients' age and anatomy as well. Today, we tend to oversize the diameter of the stentgraft used by about 10%, in contrast to the former common practice of 20%. The endovascular repair of the 12-year old child we performed, undoubtedly tests the limits of the EVT method, as the growth potential of the boy may render the stentgraft insufficient in the future.27 The decision to use an aortic stent graft in this special situation was based exclusively on the severity of the accompanying injuries and the increased risk of open surgery. After 5 years follow-up no problems have been encountered.

A safe concentric coaxial configuration of the stent graft is the best prophylaxis against stentgraft compression and type I endoleaks. A proximal landing zone of at least 2.5cm. should be secured and that means that the proximal 2.5cm. of the stentgraft should be in full contact with healthy aortic wall. This implies that the LSA ostium will be covered in approximatelly 70% of all cases. We have not encountered any serious complication from LSA coverage in more than 100 patients so far. Whether LSA transposition is advocated prophylactically or after the development of symptoms is open to debate. We prefer the second option and we have undertaken it in 1 case with clinically documented left arm hypoperfusion.28

In the future, the radial expanding force of the proximal end of the stentgraft should be increased by the implantation by the manufacturers of an extra circumferential spring just at the beginning of the proximal graft. In both our cases of ASCS stent compression was abolished by increasing radial force through proximal overstenting.

A precurved graft is currently under manufacture, for use in the aortic arch. A precurved graft would better align itself concentrically within the aortic lumen, would better follow the curvature of the aortic arch and would present minimal debending tendency. A reinforced, precurved stentgraft could become the procedure of choice in the future for the polytrauma victim.11, 29

Back to Article Outline

Acknowledgement 

We wish to thank PD Dr Reiner Muhe for the statistical processing of the data and Mr. Bernd Fetcher for retrieving many articles from the literature for us.

Back to Article Outline

Conflicts of interest 

The authors have no conflicts of interest to disclose.

Back to Article Outline

References 

  1. Nagy K, Fabian T, Rodman G, Fulda G, Rodriguez A, Mirvis S. Guidelines for the Diagnosis and Management of Blunt Aortic Injury: an EAST Practice Management Guidelines Work Group. J Trauma. 2000;48(6):1128–1143
  2. Fabian TC, Richardson JD, Croce MA, Smith JS, Rodman G, Kearney PA, et al. Prospective study of blunt aortic injury: multicenter trial of the american association for the surgery of trauma. J Trauma. 1997;42(3):374–3831997
  3. Kalmar P, Otto C, Steinkraus V. Blunt aortic injury/choosing appropriate time for surgery. Langenbecks Arch Chir. 1991;(Suppl.):563–570Kongressband
  4. Kalmar P, Puschel K, Stubbe HM, Gultekin E. Delayed surgical therapy of acute aortic rupture. Zentralbl Chir. 1996;121:750–755
  5. Pierangeli A, Turinetto B, Galli R, Caldarera J, Fattori R, Garelli G. Delayed treatment of isthmic aortic rupture. Cardiovasc Surg. 2000;8:280–283
  6. Rousseau H, Soula P, Perreault P, Bui B, d'othee BJ, Massbuan P, et al. Delayed treatment of traumatic rupture of the thoracic aorta with endoluminal covered stent. Circulation. 1999;99:498–504
  7. Stone DH, Brewster DC, Kwolek CJ, Lamuraglia GM, Conrad MF, Chung TK, et al. Stent-graft versus open-surgical repair of the thoracic aorta: mid-term results. J Vasc Surg. 2006 Dec;44(6):1188–1197
  8. Hoornweg LL, Dinkelman MK, Goslings JC, Reekers JA, Verhagen HJM, Verhoeven EL, et al. Endovascular management of traumatic ruptures of the thoracic aorta: a retrospective multicenter analysis of 28 cases in The Netherlands. J Vasc Surg. 2006;43:1096–1102
  9. Andrassy J, Weidenhagen R, Meimarakis G, Lauterjung L, Jauch KW, Kopp R. Stent versus open surgery for acute and chronic traumatic injury of the thoracic aorta: a single-center experience. J Trauma. 2006 Apr;60(4):765–771[discussion 711–712]
  10. Schumacher H, BockIer D, von Tengg-Kobligk H, Allenberg JR. Acute traumatic aortic tear: open versus stent-graft repair. Semm Vasc Surg. 2006 Mar;19(1):48–59[Review]
  11. Bahn R, Hoornweg LL. Traumatic aortic ruptures. J Cardiovasc Surg (Torino). 2005 Apr;46(2):101–105[Review]
  12. Ott MC, Stewart TC, Lawlor DK, Gray DK, Forbes TL. Management of blunt thoracic aortic injuries: endovascular stents versus open repair. J Trauma. 2004 Mar;56(3):565–570
  13. Doss M, Balzer J, Martens S, Wood JP, Wimmer-Greinecker G, Fieguth HG, et al. Surgical versus endovascular treatment of acute thoracic aortic rupture: a single-center experience. Ann Thorac Surg. 2003 Nov;76(5):1465–1469[discussion 1469–1470]
  14. von Oppell VO, Dunne TT, De Groot MK, Zilla P. Traumatic aortic rupture: twenty-year metaanalysis of mortality and risk of paraplegia. Ann Thorac Surg. 1994;58:585–593
  15. Orend K-H, Kirchdorfer B, Sunder-Plassmann L. Traumatic rupture of the descending thoracic aorta. Ann Ital Chir. 1996;67:21–23
  16. Pate JW. Traumatic rupture of the aorta: emergency operation. Ann Thorac Surg. 1985;39(6):531–537
  17. Camp PC, Shackford SR. Outcome after Blunt Traumatic Thoracic Aortic Laceration: identification of a high-risk cohort. J Trauma. 1997;43(3):413–422
  18. Parmley I, Mattingly TW, William CM, Manion WC, Jahnke E. Nonpenetrating traumatic injury of the aorta. Circulation. 1958;17:1086
  19. Sunder-Plassmann L, Orend KH. Stentgrafting of the Thoracic Aorta - Complications. J Cardiovasc Surg. 2005;46:121–130
  20. Orend KH, Pamler R, Kapfer X, Liewald F, Gorich J, Sunder-Plassmann L. Endovascular Repair of Traumatic Descending Aortic Transection. J Endovasc Ther. 2002;9:573–578
  21. Steinbauer M, Stehr A, Pfister K, Herold T, Zorger N, Topel I. Endovascular repair of proximal endograft collapse after treatment for thoracic aortic disease. J Vasc Surg. 2006;43:609–612
  22. Mestres G, Maeso I, Femandez V, Matas M. Symptomatic collapse of a thoracic aorta endoprosthesis. J Vasc Surg. 2006 Jun;43(6):1270–1273
  23. du MM, Reekers JA, Balm R, Ponsen KJ, de Mol BA, Legemate DA. Collapse of a stent-graft following treatment of a traumatic thoracic aortic rupture. J Endovasc Ther. 2005 Aug;12(4):503–507
  24. Sunder-Plassmann L, Sharrer-Pamler R, Liewald F, Kapfer X, Gorich J, Orend KH. Endovascular Exclusion of thoracic aortic aneurysms: mid-term results of elective treatment and in contained rupture. J Card Surg. 2003;18:367–374
  25. Tehrani HY, Peterson BO, Katariya K, Morasch MD, Stevens R, DiLuozzo G, et al. Endovascular repair of thoracic aortic tears. Ann Thorac Surg. 2006 Sep;82(3):873–877[discussion 877–878]
  26. Pratesi C, Dorigo W, Troisi N, Pratesi G, Santoro G, Stefano P, et al. Acute traumatic rupture of the descending thoracic aorta: endovascular treatment. Am J Surg. 2006 Sep;192(3):291–295
  27. Karmy-Jones R, Hoffer E, Meissner M, Bloch RD. Management of traumatic rupture of the thoracic aorta in pediatric patients. Ann Thorac Surg. 2003 May;75(5):1513–1517
  28. Peterson BO, Eskandari MK, Gleason TO, Morasch MD. Utility of left subclavian artery revascularization in association with endoluminal repair of acute and chronic thoracic aortic pathology. J Vasc Surg. 2006 Mar;43(3):433–439
  29. Thompson CS, Rodriguez JA, Ramaiah VG, DiMugno L, Shafique S, Olsen D, et al. Acute traumatic rupture of the thoracic aorta treated with endoluminal stent grafts. J Trauma. 2002 Jun;52(6):1173–1177

PII: S1078-5884(07)00459-5

doi:10.1016/j.ejvs.2007.06.022

European Journal of Vascular & Endovascular Surgery
Volume 34, Issue 6 , Pages 666-672, December 2007

Article Tools

* Image Usage & Resolution