European Journal of Cardio-thoracic Surgery 26 (2004) 614–620 www.elsevier.com/locate/ejcts

Endovascular surgery for failed open aortic aneurysm repairq Ludwig K. von Segesser*,1, Bettina Marty, Piergiorgio Tozzi, Christoph Huber, Ivan Bruschweiler, Augusto Gallino, Daniel Hayoz, Patrick Ruchat Department of Cardio-vascular Surgery, Centre Hospitalier Universitaire Vaudois, CHUV, Rue du Bugnon 46, CH-1011 Lausanne, Switzerland Received 28 December 2003; received in revised form 10 April 2004; accepted 21 April 2004

Abstract Objective: Determine the usefulness of endovascular surgery for repair of aortic lesions late after open surgical repair. Patients and methods: A retrospective analysis of our databank (Patient Analysis and Tracking System, Dendrite, UK) for 2000– 2002 showed 286 descending thoracic and/or abdominal aortic aneurysms: 60/286 (21%) descending thoracic, and 255/286 abdominal (89%). Endovascular surgery was planned in 98 patients (17/60 (28%) for thoracic lesions, and 81/255 (32%) for abdominal lesions). 13/98 patients (13%) underwent endovascular surgery late after failed open aortic repair: 4/13 at the level of distal aortic arch (3/4 for false aneurysms postcoarctation repair), 4/13 at the level of the descending thoracic aorta (3/4 for false aneurysms proximal to the previous graft), and 5/13 at the level of the infrarenal abdominal aorta (4/5 for false aneurysms proximal to the previous graft). Endovascular surgery included per procedural target site identification (previous graft) with intravascular ultrasound (IVUS) under fluoroscopic control (no angiographies), controlled hypotension (partial inflow occlusion with a right atrial balloon introduced through a femoral vein) for unloading of covered endoprostheses in the thoracic aorta, as well as in situ introducer sheath dilatation in case of complex access to the aorta. Results: There were no hospital deaths and no parapareses or paraplegias in this small series of patients who underwent endovascular surgery for aneurismal lesions occurring late after open repair. An endoleak type I was documented in 2/13 patients (15%) requiring a proximal extension in 1 patient. For the second patient with a minor endoleak, a control examination is planned at 6 months of follow-up. Conclusion: Endovascular surgery is an elegant approach for repair of recurring aortic lesions late after open aortic surgery. IVUS is a precious instrument for per procedural identification of the previous implants. However, long-term follow-up is mandatory after endovascular surgery. q 2004 Elsevier B.V. All rights reserved. Keywords: EVAR, Endovascular aneurysm repair; Thoracic aortic aneurysm; Abdominal aortic aneurysm; Complication; Surgery

1. Introduction The traditional approach for repair of descending thoracic aortic aneurysms and thoraco-abdominal aortic aneurysms is open surgery [1 –3]. Endovascular repair with covered stent-grafts has more recently been introduced into clinical routine [4,5]. The latter technique is not suitable for all types of descending thoracic and thoraco-abdominal aortic aneurysms, and for the time being, aneurysms involving major aortic branch-vessels are usually repaired with open surgical techniques. In addition, endovascular q Presented at the joint 17th Annual Meeting of the European Association for Cardio-thoracic Surgery and the 11th Annual Meeting of the European Society of Thoracic Surgeons, Vienna, Austria, October 12 –15, 2003. * Corresponding author. Tel.: þ41-21-314-2280; fax: þ 41-21-314-2278. E-mail address: [email protected] (L.K. von Segesser). 1 www.cardiovasc.net.

1010-7940/$ - see front matter q 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.ejcts.2004.04.045

aneurysm repair carries the burden of early and late endoleaks, which in turn may require open correction too [6]. However, recurrence of aneurismal lesions in the area previously repaired and/or at remote aortic sites are not uncommon after open surgical repair neither [7,8]. The present analysis was performed to determine the usefulness of endovascular surgery for repair of aortic lesions late after open surgical aortic repair.

2. Patients and methods A retrospective analysis of our databank (Patient Analysis and Tracking System, Dendrite, UK) for 2000– 2002 showed 286 descending thoracic and/or abdominal aortic aneurysms: 60/286 (21%) descending thoracic or thoraco-abdominal aortic aneurysms, and 255/286 abdominal aortic aneurysm (89%) accounting for a total of 315

L.K. von Segesser et al. / European Journal of Cardio-thoracic Surgery 26 (2004) 614–620

615

aneurysmal aortic lesions in 285 patients. Endovascular surgery was planned in 98 patients (17/60 (28%) for thoracic aortic lesions, and 81/255 (32%) for abdominal lesions), accounting for a total 98 endovascular surgical repairs out of 315 aneurysmal aortic lesions (31%). A subgroup including 13/98 patients (13%) was identified who underwent endovascular surgery late after failed open aortic repair. Endovascular surgery included per procedural target site identification (previous graft) with intravascular ultrasound (IVUS) under fluoroscopic control (no angiographies) as previously reported in detail (4: includes the ordering information for introducer sheaths, guide wires, catheters that accept the various IVUS probes). This technique not only allows for per-procedural measurement of aortic diameter (Fig. 1), but also for neck quality assessment, identification of intra-aortic thrombus, recognition of synthetic graft material, localization of anastomotic leaks (Fig. 2), and quality assessment after leak closure with covered stent-grafts (Fig. 3). The various types of aneurysms and graft configurations treated are shown in Fig. 4. The IVUS probe is easy to recognize by fluoroscopy, and therefore road-mapping allows for reproducible identification of vascular lesions, branch vessels, and other land marks by positioning radio-opaque markers at the appropriate cross-sections between the plane generated by the IVUS probe and a well positioned guide wire. Diluted contrast medium is only used for filling the balloons for dilatation of access vessels, provoking controlled hypotension (vide infra) or expansion/modelling of stent-grafts. Various brands of self-expanding covered stent-grafts based on nitinol z-stents with a coverage made either from polyester or PTFE were used. We actually keep a complete set of covered stent-grafts in house. Due to the fact, that no manufacturer covers the entire spectrum of

covered stent-grafts required, this set has been assembled from various manufacturers and changed over time due to the availability of the different types of implants. Selection of covered stent-grafts was mainly based on perprocedural measurement of the landing zone diameters (IVUS cross-sections at the appropriate level) and road mapping under fluoroscopy (IVUS pull-back between land marks). The dimensions of the synthetic grafts previously implanted where also considered if available. Controlled hypotension (approximatively 40– 50 mmHg mean arterial pressure) was achieved by partial inflow occlusion with a right atrial balloon introduced through a femoral vein for unloading of covered endoprostheses in the thoracic aorta [4,9]. In cases with complex aortic

Fig. 1. Intravascular ultrasound (IVUS) displaying a cross-section of the descending thoracic aorta with a synthetic graft (G), the IVUS probe (I), a guidwire with its shadow (S) and an anastomotic leak (L), distances between dots represent 4.0 mm/division.

Fig. 3. IVUS displaying a cross-section of an aneurismal descending thoracic aorta with a false aneurysm after endo-vascular repair with a covered stent-graft inserted through the common femoral artery (same patient as Fig. 1): wall of aneurysm (A), synthetic graft (G), covered endoprosthesis (E) closing the anastomotic leak: successful repair.

Fig. 2. IVUS displaying a cross-section of an aneurismal descending thoracic aorta with a false aneurysm due to an anastomotic leak after open repair: wall of aneurysm (A), synthetic graft (G), IVUS probe (I), anastomotic leak (L), thrombus (T).

616

L.K. von Segesser et al. / European Journal of Cardio-thoracic Surgery 26 (2004) 614–620

Fig. 4. Horizontal. (A) Recurrence of true aneurysm, (B) Development of anastomotic false aneurysm, (C) Failure of synthetic graft, (D) Aortic coarctation repaired by patch with development of false aneurysm, (E) Enlargement of native aortic patch carrying branch vessel. Vertical. (1) Original lesion, (2) Original open repair, (3) Actual lesion, (4) Endovascular repair.

access due to artherosclerotic femoral and iliac vessels, excessive kinking or inadequate size of the latter, in situ introducer sheath dilatation was used as previously reported [10].

For per-procedural quality assessment IVUS is used in conjunction with trans-oesophageal echocardiography (TEE), and if necessary, intra cardiac echo-cardiography (ICE: Navigate, Accuson). Both, TEE and ICE have colour

L.K. von Segesser et al. / European Journal of Cardio-thoracic Surgery 26 (2004) 614–620

617

duplex facilities and allow for flow studies. Colour duplex allows for perprocedural identification of significant endoleaks and consecutive endovascular repair during the same session. A CT-study prior to discharge is routine, and is followed by an additional assessment at 6 months, and yearly controls afterwards.

3. Results The subgroup composed of 13/98 patients (13%) who underwent endovascular surgery late after failed open aortic repair (mean interval between open surgery and endovascular repair: 9 ^ 8 years) included patients with false aneurysm formation at the level of distal aortic arch in 4/13 (3/4 for false aneurysms post-coarctation repair (Fig. 4D), 1/4 after arch repair for type A aortic dissection) at the level of the descending thoracic aorta in 4/13 (3/4 for false aneurysms proximal to the previous graft, 1/4 for false aneurysm distal to the previous graft (Fig. 4B)), and at the level of the infrarenal abdominal aorta in 5/13 (4/5 for false aneurysms proximal to the previous graft, 1/4 for false aneurysm distal to the previous graft). 10/13 patients received straight covered stent-grafts for late repair after open surgery, whereas 2/13 patients received tapered aortomono-iliac covered stent-grafts, and 1/13 patients received a bifurcated aorto-bi-iliac covered stent-graft (see also Table 1). The four patients who received an endo-vascular stentgraft of the distal aortic arch had their left subclavian artery orifice covered during the procedure. In one patient with an aneurismal subclavian artery, the latter was transposed onto the left carotid artery prior to endovascular aneurysm repair. In another patient, the left subclavian artery was re-vascularized with a carotidosubclavian bypass post-endovascular aneurysm repair. No distal perfusion problems were documented at rest. However, one patient declares upper left extremity fatigue at work. Until now, this phenomenon did not justify direct subclavian artery re-vascularization. All covered stent-grafts were introduced through the femoral arteries and unloaded under fluoroscopic control (Fig. 5) after previous road-mapping with IVUS. No perprocedural angiographies were performed, and no distal perfusion problems occurred in these patients. There were no hospital deaths and no parapareses or paraplegias in this small series of patients who underwent endovascular surgery for aneurismal lesions occurring late after open repair. Mean follow-up is 1.7 ^ 0.8 years. An endoleak type I was documented in 2/13 patients (15%) requiring a proximal extension in 1 patient. For the second patient with a minor endoleak, a further control examination is planned in 6 months. Endovascular graft extension, coiling, and/or open re-repair may be required for repair if spontaneous closure does not occur.

Fig. 5. Per-procedural fluoroscopic view in a patient with false aneurysm (A) after coarctation patch-repair as shown in Fig. 4D. Blunt needles positioned after road-mapping with IVUS mark: orifice of aneurismal subclavian artery (S, excluded but bypassed previously), end of proximal neck (P), and beginning of distal neck (D). Two covered stent-grafts (Talent, Medtronic, Tolochenaz, Switzerland: 120/22/24 and 140/26/28) were implanted.

4. Discussion Endovascular surgery is an elegant approach for repair of recurring aortic lesions late after open aortic repair. Because of the sometimes quite important adhesions formed after a previous surgical procedure, re-operations of the aorta taking the open route are often more demanding than primary surgical procedures. As a matter of fact emergency and redo-operations were the only significant risk factors for mortality in study on 115 patients who underwent open descending thoracic and thoraco-abdominal aneurysm repair [11]. In addition, lack of collaterals due to previous surgical dissection has been suggested as one factor enhancing the risk of parapareses and paraplegias during redo descending thoracic and thoraco-abdominal aortic surgery. Hence, an endovascular approach, using per definition a different route of access is particularly interesting for re-interventions.

618

L.K. von Segesser et al. / European Journal of Cardio-thoracic Surgery 26 (2004) 614–620

Table 1 Subgroup of patients who underwent endovascular surgery late after failed open aortic repair Patient Nr

Age

Sex

Previous procedure

Localization of new lesion

Type of repair

Outcome

1

65

M

Proximal false aneurysm

Straight covered stent-graft

Success

2 3 4 5

66 59 60 39

M M M M

Open repair AAA with bifurcated graft Open repair descending TAA Open repair descending TAA Open repair descending TAA Coarctation patch repair

Distal false aneurysm Proximal false aneurysm Distal false aneurysm False aneurysm including origin of subclavian artery

success Success Success Success

6

31

F

Coarctation patch repair

False aneurysm

7

51

M

Open repair descending TAA

Proximal false aneurysm

8

77

M

Proximal false aneurysm

9

69

M

Distal false aneurysm

10 11

70 73

M M

Proximal false aneurysm Distal false aneurysm

Straight covered stent-graft Straight covered stent-graft

Success after distal embolectomy Success Success

12

75

F

Proximal false aneurysm

Bifurcated covered stent-graft

Success

13

65

F

Open repair AAA with bifurcated graft Open repair AAA with bifurcated graft Open repair descending TAA Open repair AAA with bifurcated graft Open repair AAA with bifurcated graft Open arch repair

Straight covered stent-graft Straight covered stent-graft Straight covered stent-graft Transposition of subclavian artery, straight covered stent-graft Covered stent-graft excluding sub-clavian artery Straight covered stent-graft endoleak repaired Tapered covered aorto-monoiliac stent-graft Straight covered stent-graft

Distal false aneurysm

Straight covered stengraft

Endoleak type 1 coiled

A PubMed search (National Library of Medicine) updated in April 2004 showed for the search term ‘Descending thoracic aortic aneurysm re-operation’ 86 hits, whereas ‘Redo thoracic aortic aneurysm surgery’ accounted for 21, and ‘Redo descending thoracic aortic surgery’ provided 8 references. Only one reference [12] was captured by the search term ‘Redo endovascular aneurysm repair for thoracic aortic aneurysms’. The latter deals, however, with the repair of an endoleak after endovascular aneurysm repair, a situation most of us have encountered. For the former series there are a number of papers which are relevant to the issues raised here. Kawaharada et al. [13] reported on open surgical treatment of thoracoabdominal aortic aneurysm after repairs of descending thoracic or infrarenal aortic aneurysms and found a mortality 5.5% for patients with previous aneurysm repair versus 13% for patients without. In contrats the rate of paraplegia was 14% for patients with with previous aneurysm repair versus 3.1% for patients without. Renal failure requiring hemodialysis accounted for 22% with previous aneurysm repair versus 19% for patients without. There can be no doubt, that the proportion of problems reported was rather high in comparison to the endovascular approach presented here. The main reason for re-intervention in the former group was progression of the aneurismal disease as we are also used to see in aneurysms of the ascending aorta [14]. The fact, that endoleaks after endovascular aneurysm repair may require open surgery [15] is well known and also called, (early or late) conversion. Less reported is the inversed sequence where endovascular repair is the procedure selected after failed open repair as reported

Left arm fatigue (right arm dominant) Endoleak type 1 repaired with proximal extension Success

here. With the search strategy applied to PubMed as reported above, two case reports came-up dealing with the endovascular repair of an intercostal patch dehiscence [16,17] as depicted schematically in Fig. 4E. The various types of problems that can occur after primary open aortic repair are shown schematically in Fig. 4: the first column shows the primary lesion, the second column depicts the results after primary open repair, the third column exposes the new problems, and the fourth column displays the result after endovascular repair. Typical lesions include true aneurysm formation (Fig. 4A) in an aortic zone close to the previous repair (progression of the original disease), anastomotic false aneurysm formation (Fig. 4B), failure of a synthetic graft (Fig. 4C: excessive graft dilation), false aneurysm formation after coarctation repair with a synthetic patch (Fig. 4D), and dilatation of an aortic patch carrying branch vessels (Fig. 4E: previously preserved for perfusion of supra-aortic vessels, intercostal, visceral or renal arteries). In contrast to redo procedures after endovascular aortic aneurysm repair, where the previous implant can be readily seen by fluoroscopy (similar to Fig. 5 after secondary endovascular repair), traditional vascular prostheses made from polyester, PTFE or even homografts are more difficult to be detected from outside and the same holds true for potential defects like anastomotic false aneurysms. However, vision from within, like the one provided by intravascular ultrasound (IVUS) has the potential to provide the necessary information for endovascular repair of failed open procedures [4]. In the case of a recurrent true aneurysm directly at the level of the suture line (Fig. 4A) or more remote can be recognized with IVUS due to the differences in

L.K. von Segesser et al. / European Journal of Cardio-thoracic Surgery 26 (2004) 614–620

diameter at the level of the previous implant as compared to the dilated aortic wall. In addition the different echo-densities of natural aortic wall, wall of a false aneurysm, thrombus, as well as synthetic graft wall all help for proper localization of a problem like a leak (Fig. 1) leading to a false aneurysm (Figs. 2 and 4B). Synthetic graft failure can be once again identified with IVUS due to the changing diameter (Fig. 4B) of the vascular implant and its density, and the same holds true for aneurysm formation after aortic repair with a synthetic patch (4D), as well as for aortic patch enlargement (4E) following preservation of collaterals during previous repair. Covered stent-grafts carry also their specific signatures if viewed with IVUS: endografts based on large Z-stents (Fig. 5) can be recognized due to the shadows thrown by their metallic parts which are similar to guide wires (Fig. 1), whereas the small amounts of air captured in PTFE covered stent-grafts in conjunction with thinner wires, also provide a specific image (Fig. 3). Of course per-procedural angiography also allows for detection of true and false aneurysms, which can be marked on the corresponding fluoroscopic screen and also allow for endovascular repair. The main advantage of IVUS in this context, is that its circumferential view allows for checking the entire circumference of a graft and easy leak detection at the site of interruption (Figs. 1 and 2). Hence, for measurement of the dehiscence, it is not necessary to have fluoroscopy in the right plane. The latter can be positioned for optimal exposure of the concerned aortic (arch) segment. Likewise, for post-implant quality assessment, IVUS allows for checking perfect coaptation between the new implant and the previous graft and/or original aortic wall, respectively. In addition there is a number of indirect IVUS signs that allow for some insight into the effectiveness of endovascular aneurysm repair. Increasing IVUS gain and contrast allows for detection of blood components (Fig. 1: see ‘gray’ content of vascular graft) and also its movement. The latter is reduced if the orifice feeding the false aneurysms is successfully sealed. Likewise, pulsations of sealed (false) aneurysm walls are reduced. Although, we have the additional possibility of enhanced (endo-) leak detection by use of an intravascular colourduplex probe (Navigate, Accuson), this tool is in general not necessary for assessment of the repaired descending thoracic aorta, which can be well studied by transoesophageal ultrasound (TEE). Hence, at the thoracic level, we usually rely on the disappearance of the TEE flow signal in the aneurismal sac for considering an endovascular aneurysm repair procedure successful. In conclusion, we consider, the endovascular route the preferred approach for repair of late failure after open aortic surgery. The technique used after previous open surgery is similar to primary endovascular procedures [4]. Knowledge about the size of previous implants can help during the selection process of the covered stent-grafts to be used for repair. IVUS is a precious instrument for per-procedural identification of the previous implants, assessment of

619

the failure zone, neck dimension and quality determination, as well as for quality control. Considering the fact, that a major part of an endoprosthesis used for correction of a previous open procedure is positioned within a pre-implanted synthetic graft, therefore resulting in a double layered configuration, we expect this type of procedure rather more durable than a primary endovascular repair. In about half of the procedures described, the endoprosthesis used covered a relatively small orifice feeding a false aneurysm. Considering the fact of lacking long-term data for this application, we, however, maintain the necessity of life-long follow-up of these repairs with regard to device integrity, occurrence of other aneurismal lesions, and, most importantly [18] efficient blood pressure control.

References [1] von Segesser LK, Burki H, Schneider K, Siebenmann R, Schmid ER, Turina M. Outcome and risk factors in surgery of descending thoracic aneurysms. Eur J Cardiothorac Surg 1988;2:100– 5. [2] von Segesser LK, Marty B, Mueller X, Ruchat P, Gersbach P, Stumpe F, Fischer A. Active cooling during open repair of thoraco-abdominal aortic aneurysms improves outcome. Eur J Thorac Cardiovasc Surg 2001;19:411 –5. [3] Cooley DA, Golino A, Frazier OH. Single-clamp technique for aneurysms of the descending thoracic aorta: report of 132 consecutive cases. Eur J Cardiothorac Surg 2000;18:162–7. [4] von Segesser LK, Marty B, Ruchat P, Bogen M, Gallino A. Routine use of intravascular ultrasound for endovascular aneurysm repair: angiography is not necessary. Eur J Vasc Endovasc Surg 2002;22: 316–8. [5] Grabenwoger M, Fleck T, Czerny M, Hutschala D, Ehrlich M, Schoder M, Lammer J, Wolner E. Endovascular stent-graft placement in patients with acute thoracic aortic syndromes. Eur J Cardiothorac Surg 2003;23:788–93. [6] Mitchell RS, Dake MD, Semba CP, Fogarty TJ, Zarins CK, Liddell RP, Miller DC. Endovascular stent-graft repair of thoracic aortic aneurysms. J Thorac Cardiovasc Surg 1996;111:1054–62. [7] Okita Y, Ando M, Minatoya A, Tagusari O, Kitamura S, Nakajjama N, Takamoto S. Early and long term results of surgery of the thoracic aorta in septuagenarians and octogenarians. Eur J Cardiothorac Surg 1999;16:317 –23. [8] Genoni M, Paul M, Tavakoli R, Ku¨nzli A, Lachat M, Graves K, Seifert B, Turina M. Predictors of ccomplications in acute type B aortic dissections. Eur J Cardiothorac Surg 2002;22:55–63. [9] Marty B, Morales CC, Tozzi P, Ruchat P, Chassot PG, von Segesser LK. Partial inflow occlusion facilitates accurate deployment of thoracic aortic endografts. J Endopvasc Ther 2004;11:175–9. [10] von Segesser LK, Marty B, Tozzi PG, Corno A. In situ introducer sheath dilatation for complex aortic access. Eur J Cardiothorac Surg 2002;22:316 –8. [11] Bachet J, Guilmet D, Rosier J, Cron C, Dreyfus G, Goudot B, Piqois A, Brodaty D, Dubois C, de Lentdecker P. Protectection of the spinal cord during surgery of thoraco-abdominal aortic aneurysms. Eur J Cardiothorac Surg 1996;10:817–25. [12] Lachat M, Pfamatter T, Turina M. Transfemoral endografting of thoracic aortic aneurysm under local anesthesia: a simple, safe, and fast track procedure. Vasa 1999;28:204–6. [13] Kawaharada N, Morishita K, Fukada J, Watanabe T, Abe T. Surgical treatment of thoracoabdominal aortic aneurysms after repairs of descending thoracic or infrarenal abdominal aortic aneurysms. Eur J Cardiothorac Surg 2001;20:520–6.

620

L.K. von Segesser et al. / European Journal of Cardio-thoracic Surgery 26 (2004) 614–620

[14] Dossche KM, Tan ME, Schepens MA, Morshuis WJ, Brutel de la Rivie`re A. Twenty-four years experience with reoperations after ascending aortic or aortic root replacement. Eur J Thorac Cardiothorac Surg 1999;16:607–12. [15] Iguchi A, Tsuru Y, Tabayashi K. Open surgical repair for perigraft leak after endovascular stent-graft emplacemenet of descending thoracic aneurysm. Jpn J Thorac Cardiovasc Surg 2001;49:618 –20. [16] Dorros G, Avula S, Fox P, Rhomberg B, Werner P. Endovscular covered stent repair of an intercostals artery patch dehiscence from a descending thoracic aortic aneurysm graft. J Endovsc Surg 1996;3: 229–305. [17] Hachiro Y, Harad H, Mawatari T, Baba T, Honma Y, Abe T. Endovascular stent-grafting for rupture of a thoracoabdominal aortic aneurysm after replacement of the entire aorta. Ann Thorac Surg 2004;77:1101. [18] Genoni M, Paul M, Jenni R, Graves K, Seifert B, Turina M. Chronic b-blocker therapy impoves outcome and reduces treatment costs in chronic type B aortic dissection. Eur J Cardiothorac Surg 2001;19: 606–10.

Appendix A. Conference discussion Dr M. Turina (Zurich, Switzerland): You are to be congratulated for addressing this very difficult problem, which is a major area of concern,

after the technique of the implantation of the intercostal arteries became accepted and aortic patches are being avoided. Did this experience with dilatation of the visceral artery patch and the dilatation of the intercostal arterypatches lead you to change the technique presently? Presently grafts are becoming available with not only 1 but up to 4 branches, which I preferentially use to implant not only several intercostal arteries, but all visceral arteries as well. Dr von Segesser: I have not changed the technique until now, but I think it’s a very important issue and probably this is one way to do it. But for the patient I have shown you here, I think that even with end-to-end anastomoses with one of these branches, she may come back one day. Dr Turina: You’re perfectly right. But one feels much better with a classical end-to-end anastomosis rather than with large patches which are sometimes, especially in the redos, very difficult to implant in an end-to-end technique. You will have to use an inclusion technique and you never know what you are really stitching. Dr von Segesser: I agree completely. Dr A. Scheule (Tuebingen, Germany): Do you use the IVUS also for placement of the endovascular prosthesis or just for identification of the endoleak? Dr von Segesser: No, no. We implant all endoprostheses with intravascular ultrasound. So the procedure is to have the patient in the operating theater with fluoroscopy; and instead of using angiography, using intravascular ultrasound. We map the different branches and the beginning and the end of the necks and of the aneurysm, and we mark this under fluoroscopy so that we know where we are.