Endoluminal stent-graft placement in patients with acute aortic dissection type B q

European Journal of Cardio-thoracic Surgery 21 (2002) 964 969 www.elsevier.com/locate/ejcts Endoluminal stent-graft placement in patients with acute aortic dissection type B q D. Hutschala a, T. Fleck a, M. Czerny a, M. Ehrlich a, M. Schoder b, J. Lammer b, E. Wolner a, M. Grabenwöger a, * a Department of Cardio-Thoracic Surgery, University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria b Department of Angiography and Interventional Radiology, University of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria Received 17 September 2001; received in revised form 17 January 2002; accepted 18 January 2002 Abstract Objectives: This study was performed to evaluate the feasibility, safety and effectiveness of endovascular stent-grafting in treating Stanford type B acute aortic dissection. We describe our first clinical experiences and initial results with stent-grafting across the primary entry tear in patients with acute aortic dissection type B. Methods: Between March 2000 and August 2001, nine patients with acute type B dissection were treated endoluminally by stent-graft implantation. There were seven male and two female patients with a mean age of 63 years (between 48 and 85 years). In all nine patients aortic dissection was diagnosed by multislice computed tomography (CT) angiography. All nine patients had a maximal aortic diameter of 5.5 cm or more and recurrent pain, one patient showed hemoptysis. This patient with signs of a contained rupture was treated under emergency condition, the eight remaining patients were in hemodynamic stabile condition at the time of intervention. The GORE w Excluder stent-graft system was used in eight patients (mean 1.8 stents/patient) and the TALENT w stentgraft system in one patient, which were introduced transfemorally. Results: The primary entry tear could be sealed successfully in all nine patients. Complete thrombosis of the false thoracic aortic lumen was obtained in two patients, in the remaining seven patients the false lumen was obliterated in the area of the thoracic aorta but perfused via re-entries in the abdominal region. No severe intraoperative complications occurred. One patient developed bilateral incomplete paraplegia with motor and sensory deficits affecting completely the right leg and partially the left leg, 14 h after intervention. A cerebrospinal fluid drainage was initiated by inserting a lumbar catheter. All nine patients, including the patient with the transient paraplegia, could be discharged from the hospital in excellent condition and without remaining neurologic deficits. Control CT scans showed a reduction of the false lumen from 2.34 ^ 0.58 to 0.7 ^ 0.44 cm and an increase of the true lumen from 1.56 ^ 0.5 to 4.10 ^ 0.6 cm in the thoracic aortic region. Mean ICU stay was 1.8 days, mean postoperative hospital stay was 7.6 days. Conclusions: Stent-grafting of acute type B dissections may represent a very effective and promising new method by closure of the primary entry tear, thereby minimizing the risk of rupture of the thoracic aorta and optimizing distal perfusion by decompression of the true lumen. q 2002 Elsevier Science B.V. All rights reserved. Keywords: Acute aortic dissection type B; Stent-graft placement; Cerebrospinal fluid drainage 1. Introduction q Presented at the joint 15th Annual Meeting of the European Association for the Cardio-thoracic Surgery and the 9th Annual Meeting of the European Society of Thoracic Surgeons, Lisbon, Portugal, September 16 19, 2001. * Corresponding author. Tel.: 143-1-40400-5620; fax: 143-1-40400-6972. E-mail address: martin.grabenwoeger@univie.ac.at (M. Grabenwöger). Diseases of the thoracic aorta are still a great therapeutic challenge. Especially acute aortic dissection is a life-threatening situation for the patient. For patients with acute Stanford type A dissection, surgical intervention is performed immediately after diagnosis to avert the high risk of death due to various complications, such as cardiac tamponade, aortic regurgitation and myocardial infarction. In contrast, the preferred treatment for the uncomplicated Stanford type B dissection is medical therapy, including the use of antihypertensive drugs. Surgical treatment is reserved for patients who are complicated by progression of dissection, impending rupture, refractory hypertension, contained pain or end-organ ischemia caused by compromise of the aortic branches. Although diagnostic imaging, pharmacological control, anesthesia technology and surgical techniques have improved, surgical aortic repair is still associated with high morbidity and mortality 1010-7940/02/$ - see front matter q 2002 Elsevier Science B.V. All rights reserved. PII: S1010-7940(02)00064-7

D. Hutschala et al. / European Journal of Cardio-thoracic Surgery 21 (2002) 964 969 965 rates especially in a population that is frequently elderly and debilitated [1 3]. Ever since Parodi et al. [4] described their first clinical experiences with the use of a stent-graft to treat an abdominal aortic aneurysm, endoluminal stent-graft placement has been emerging as a less invasive alternative to conventional surgery in patients with thoracic aortic aneurysms. Dake et al. [5,6] reported excellent initial clinical results with transcatheter stent-graft treatment of aortic type B dissection in patients at high risk. In the present study, we evaluated the feasibility and safety of endoluminal stent-graft treatment for acute type B dissection. 2. Material and methods 2.1. Patient population Between March 2000 and August 2001 there were nine patients, seven male and two female patients, with acute type B dissections who were treated using transluminal thoracic stent-graft insertion. The mean age of these patients was 63 years (range, 48 85 years). Six patients (66.7%) had a classic entry tear localization and the three remaining patients (33.3%) had a primary tear that was identified in the distal part of the descending aorta. The dissection extended into the abdominal aorta in three patients (33.3%) and distally into the iliac arteries in two patients (22.2%). In one case, the inferior mesenteric artery and right renal artery were perfused from the false lumen, in all other cases, abdominal vessels were perfused from the true lumen of the dissected aorta. Eight patients (88.9%) were hemodynamically stable under antihypertensive therapy, and no rupture or bleeding was observed. Indications for invasive therapy were recurrent pain and increasing pleural effusion. One patient with the signs of a contained rupture was treated under emergency conditions. Diagnosis was performed by multislice computed tomography (CT) scans in all patients (Fig. 1). Exact identification of primary entry site was feasible using this technique. Required anatomical criteria for stent-graft insertion were an entry site at least 1 cm distal to the left subclavian artery and suitable access with no severe tortuosity, and dissection or narrowness of the iliac or femoral arteries that could impede introduction of the stent-graft delivery system. The maximum diameters of the true and the false lumen of the thoracic aorta were measured, evaluated from crosssectional CT scans. Hypertension was the most common preoperative medical disorder (97.2%) followed by pulmonary dysfunction (33.3%) and coronary artery disease (22.2%). Two patients (22.2%) suffered from myocardial dysfunction with an ejection fraction of less than 0.25. Obstructive pulmonary dysfunction with a forced expiratory volume in Fig. 1. CT scan shows dissected thoracic aorta with a small true lumen and large false lumen. Note significant pleural effusion at the left side. 1 s of less than 50% of predicted values was present in three patients (33.3%). One patient had undergone previous coronary artery bypass grafting. The mean time between diagnosis of the dissection and stent-grafting was 4.75 days (range, 1 9 days). 2.2. Stent-graft systems Two different stent-graft systems are available in our department. The GORE w Thoracic Excluder Endoprothesis (W.L. Gore & Associates, Sunnyvale, CA, USA) was used in eight patients (88.9%). Exclusion of thoracic aortic aneurysms with the TALENT w endoluminal stent-graft system (Medtronic, Sunrise, FL, USA) was performed in one patient (11.1%). The TALENT w stent-graft consists of a nitinol wire stent shaped in a zigzag formation, which is covered with extrathin polyester (Dacron). A straight nitinol wire severs the length of the device and avoids twisting or kinking. This self-expandable stent-graft is compressed over a placement catheter. Both the stent-graft and the catheter are loaded into a polyurethane sheath for insertion. The endoluminal stentgraft system is passed over the guidewire and positioned at the desired location as determined by intraoperative angiography. After exact positioning, the stent-graft was released by removing the sheath. The GORE w Thoracic Excluder stents are constructed differently. The system is placed into the vasculature through an introduction sheath. The stent-graft itself is mounted on a placement catheter. Development of the stent-graft is achieved by pulling on a string at the end of the placement catheter. GORE w stent-grafts are available in standard sizes. GORE w stent-grafts were used in eight patients (88.9%), TALENT w stent-grafts were placed in one patient (11.1%).

966 D. Hutschala et al. / European Journal of Cardio-thoracic Surgery 21 (2002) 964 969 2.3. Surgical procedure Stent-grafting was performed in an angiosuite by a team of cardiovascular surgeons and interventional radiologists. Eight patients were operated in general anesthesia attained by using tracheal intubation and mechanical ventilation, one patient was treated in spinal anesthesia. Patients were prepared to undergo surgery in case the procedure failed. First a 5-French calibrated angiographic pigtail catheter was advanced via the right subclavian artery into the ascending aorta to permit an arteriographic evaluation of the distance between the left subclavian artery and the entry tear. In all nine patients, it was possible to identify the entrance site precisely. An angiography with an automatic injector was performed to reconfirm characterization of the morphology and extent of the dissection (Fig. 2). After the arteriographic evaluation, 5000 UI of heparin sodium was administered intravenously and antibacterial medication was given prophylactically. The femoral (n ¼ 6; 66.7%) or distal iliac artery (n ¼ 2; 22.2%) was surgically exposed. In one patient an aortobifemoral prothesis had to be used for vascular access. After the insertion of the appropriately sized introducer sheath, a guidewire (Backup Meier, Boston Scientific, Natick, MA, USA) was placed into the ascending aorta. Next, the delivery system was introduced through the true lumen and was carefully advanced into the optimal position under fluoroscopic guidance. After correct positioning of the device the stent-graft was deployed. Reduction of the systolic blood pressure was not considered necessary, as deployment of the GORE w Thoracic Excluder stent-graft spreads within milliseconds from the middle to both ends of the graft. After deployment, a TAG w Balloon Catheter (W.I. Gore & Associates, Inc., Sunnyvale, CA, USA) was inflated inside the stent-graft to achieve full expansion and to anchor the stent in the aortic wall. In contrast to the GORE w device, the balloon of the TALENT w device is included in the delivery system. Finally, an angiogram was performed to confirm the position of the device relative to the entry tear, to evaluate patency of the left subclavian artery and blood flow in the visceral arteries and to verify the effectiveness of the stentgrafting. 2.4. Follow-up Follow-up was performed during a time period between 1 day and 6 months (mean 89 days) after intervention. In all nine patients, the first follow-up was done within 72 h after operation. Spiral CT scans were then obtained 6 months after stent-grafting, yearly thereafter, and at the onset of any new symptoms. Because of the actuality of this study we could not perform a second or third follow-up in all patients. The second follow-up was realized in five cases (55.6%) and the third follow-up in three cases (33.3%). For follow-up investigation, spiral CT scans of the aorta were analyzed for correct placement of the stent-device and the perfusion of the false lumen. Correct placement in the follow-up was defined as the stent-graft being in the same position, without altered overlaps of the proximal and distal aorta and complete expansion without kinking or stenosis. Diameters of the true and the false lumen were measured along a line perpendicular to the intimal flap. Clinical data concerning early mortality and morbidity were available in all patients at least over a period of 30 days, either in the hospital or after discharge. 3. Results Fig. 2. Intraoperative angiography is performed to evaluate the morphology of the dissection, especially for identification of the primary entry tear. Stent-graft deployment within the true lumen of the aorta and complete coverage of the primary entry tear were technically and clinically successful in all cases. There was no hospital mortality and all patients were doing fine during the follow-up period. Furthermore, no cases of aneurysm formation or aortic rupture were documented and no additional aortic operation was necessary.

D. Hutschala et al. / European Journal of Cardio-thoracic Surgery 21 (2002) 964 969 967 Fig. 3. Postoperative angiography confirms closure of the false lumen. the false lumen of the dissected aorta decreased from 2.34 ^ 0.58 to 0.7 ^ 0.44 cm, and the true lumen increased from 1.56 ^ 0.5 to 4.10 ^ 0.60 cm in the thoracic region. Complete thrombosis of the thoracic aortic false lumen was evident in two patients and partial thrombosis was evident in seven patients on angiographic studies or CT images obtained within 6 months after stent-graft placement (Figs. 3 and 4). In these seven cases, the false lumen was obliterated in the area of the thoracic aorta but perfused via re-entries in the abdominal region (Fig. 5). Except for one case, all abdominal vessels were perfused from the true lumen of the dissection. In this patient, the perfusion of the inferior mesenteric artery and the right renal artery from the false lumen of the dissection did not change during follow-up, and there were no ischemic symptoms of the abdominal organs. No substantial kinking or migration of the stent-grafts was noted when comparing the multislice CT images of the stent-grafts obtained during the follow-up. All implanted stent-grafts were fully expanded and perfused, with no sign of an endoleak. In the thoracic part of the aorta no antegrade perfusion of the false lumen was seen. One procedure-related neurological complication occurred. A patient treated with stent-grafts for acute dissection type B developed bilateral incomplete paraplegia within 14 h postoperatively. A cerebrospinal fluid drainage (CSF) was initiated 8 h after onset of symptoms by inserting a lumbar catheter in the height of L3 L4 [8]. Initial liquor pressure was 45 mmhg. To reach the target pressure of 15 mmhg liquor was drained for 96 h. A total volume of 800 ml was collected and neurologic symptoms decreased continuously. After an uneventful postoperative course the patient could be discharged with no remaining neurologic deficit. After the intervention, the patients were transferred to the intermediate care unit where routine examinations were performed. Four patients presented a postimplantation syndrome, described by Blum et al. [9], with elevated inflammatory parameters such as leukocytosis, elevated C- reactive protein levels and slight back pain, which normalized without therapy between 5 and 20 days. The stay at the intermediate care unit averaged 2.3 days (between 1 and 5 days), mean postoperative hospital stay was 7.4 days (between 5 and 21 days). In six patients (66.7%), deployment of two stent-grafts in an overlapping manner was necessary to cover additional entry tears into the false lumen, which were located more distally. The mean diameter of the implanted stent-grafts ðn ¼ 13Þ was 28 44 mm (mean, 34 mm), the mean length of the stent-grafts was 100 mm (range, 50 200 mm). Total procedure time was 89 300 min (mean, 121 min); total fluoroscopy time was 9 32 min (mean, 19 min). During follow-up investigations, the mean diameter of 4. Discussion Our preliminary study indicates that stent-grafting of acute type B dissections may represent a very effective and promising new method by closure of the primary entry tear, thereby minimizing the risk of rupture of the thoracic aorta and optimizing distal perfusion by decompression of the true lumen. Clinical application of stent-grafting in acute aortic dissections was first reported by Dake et al.[5,7] and Niena- Fig. 4. CT scan reveals regularly perfused stent-graft in the true lumen and completely obliterated false lumen.

968 D. Hutschala et al. / European Journal of Cardio-thoracic Surgery 21 (2002) 964 969 Fig. 5. After stent deployment control angiography shows retrograde perfusion of the false lumen via a re-entry in the height of the renal arteries. ber et al. [7]. These two studies report preliminary findings with acute aortic dissections treated by catheter-delivered endovascular stent-grafts. The objective of their approach was to seal the entry tear, promote thrombosis of the false lumen and stabilize the dissection. Both these studies, which yielded favorable early results, represent a major step forward in the treatment of this potentially lethal disorder. Recent investigations consider endoluminal stent-graft application for the treatment of type B dissection as a potential alternative to surgical repair [10,11]. The basic principle for the established surgical repair of the dissected aorta is to obviate the most frequent cause of death, due to local rupture mostly of the false lumen [12]. However, in this series we focused on patients, who otherwise have been subjected to conservative treatment. In these patients, a significant risk for development of late complications such as increasing aortic diameter and aortic rupture is reported. The authors conclude that a more aggressive therapeutic strategy is warranted [13]. Our approach is similar to surgical obliteration of the entry tear because it can exclude flow through the initial tear in the intima and redirect aortic blood flow exclusively into the true lumen. Our experience suggests that stent-graft placement over the primary entry tear in patients with acute type B dissections may be an alternative to both open surgery and conservative treatment. In addition, stent-graft placement over the intimal tear can prevent the development of an aneurysm by facilitating complete thrombosis of the false lumen. In this series, thrombosis and shrinkage of the false lumen in the descending thoracic aorta were observed in two of nine patients after stent-grafting and partial thrombosis in the remaining seven patients. In these seven patients the false lumen was obliterated in the area of the thoracic aorta but perfused via reentries in the abdominal region. Even if only partial thrombosis of the false lumen is achieved, it still can be advantageous: it may protect the false lumen from enlarging over time, since systemic blood pressure is no longer directly transmitted from the aorta through a large primary tear in the intima. For interventional therapy, restrictions including anatomical conditions especially the distal vascular access through the iliac or femoral arteries and a sufficient proximal neck between the primary entry tear and the aortic arch are necessary. Since most aortic dissections type B begin immediately distal from the left subclavian artery, proximal anchoring of the prosthesis with simultaneous sealing of the primary entry can be problematic. However, overstenting of the left subclavian artery may be performed under emergency conditions. Furthermore, distal vascular access can be complicated by dissection of the iliac or femoral arteries. In all our patients, dissection extended only unilaterally in peripheral arteries, which was identified by preoperative CT scans. The position of the guidewire in the true lumen was verified by administration of contrast media during insertion of the catheter. In contrast to the studies of Dake et al. [5]and Nienhaber et al. [7], two-thirds of our patients received a second stentgraft in an overlapping manner. The aim of stent grafting over a longer distance is not only the complete sealing of the primary entry, but also the stabilization of the dissected thoracic aorta. In our opinion, this leads to a better decompensation of the false lumen and therefore to better perfusion of the true lumen. Furthermore, the risk of perigraft leakage should be minimized. In accordance with Czermark et al. [14], we hypothesize, that an incomplete closure of entry sites and non-clotting of the false lumen are factors adversely influencing outcome. Endoluminal repair of type B dissection obviates thoracotomy, thereby preventing respiratory dysfunction, especially in patients with preexisting pulmonary insufficiency. After intervention, no pulmonary or cardiac complications occurred. We observed one severe neurologic complication. This patient suffered from bilateral paraparesis with motor and sensory deficits affecting completely the right leg and partially the left leg. Cerebrospinal fluid drainage was initiated 8 h after onset of symptoms via a lumbar catheter. A total volume of 800 ml was drained within 96 h. Subsequently, neurologic symptoms declined and the patient came to full recovery. In our series, no central neurologic deficit could be observed. It can be suggested that the catheter technique applied is responsible for this positive

D. Hutschala et al. / European Journal of Cardio-thoracic Surgery 21 (2002) 964 969 969 outcome. At first, the aortic arch is advanced by a soft pigtail catheter. Thereafter, a stiff guidewire (Backup Meier) is inserted via the pigtail catheter into the ascending aorta. Advancement of the stent-graft into the distal aortic arch over the extra-stiff guidewire could be performed avoiding extensive manipulation in the aortic arch. Recently, due to the rising number of stent-graft insertions as an alternative treatment in a selected group of patients, attention has been focused on the possibility of neurologic deficit occurring after stent-graft placement [15,16]. The accepted mechanism of spinal cord protection achieved by CSF drainage is decompression of the spinal compartment and restoration of spinal cord perfusion. In our opinion, paraplegia was a result of increased liquor pressure in the spinal cord. We suggested that the increased liquor pressure was a consequence of reperfusion edema resulting from obliteration of three pairs of intercostal arteries by the stent-graft, which was deployed in the distal portion of the thoracic aorta. CSF drainage was able to restore the microcirculatory flow in the spinal cord. In conclusion, we demonstrate that stent-grafting over the primary entry tear is feasible and may represent an effective therapeutic option of acute type B dissection by minimizing the risk of rupture and by optimizing perfusion of the distal aorta by decompression of the true lumen. However, further studies need to be done to evaluate long-term effectiveness of this method. References [1] Coselli JS, Plestis KA, La-francesa S, Cohen S. Results of contemporary surgical treatment of descending thoracic aortic aneurysms: experience in 198 patients. Ann Vasc Surg 1996;10(2):131 137. [2] Svensson LG, Hess KR, Coselli JS, Safi HJ. Influence of segmental arteries, extent, and atriofemoral bypass on postoperative paraplegia after thoracoabdominal aortic operations. J Vasc Surg 1994;20:255 262. [3] DeBakey ME, McCollum CH, Graham JM. Surgical treatment of aneurysms of the descending thoracic aorta: long-term results in 500 patients. J Cardiovasc Surg 1978;19:571 576. [4] Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms. Ann Vasc Surg 1991;5:491 499. [5] Dake MD, Miller DC, Semba CP, Mitchell RS, Walker PJ, Liddell RP. Transluminal placement of endovascular stent-grafts for the treatment of descending thoracic aortic aneurysms. N Engl J Med 1994;331:1729 1734. [6] Dake MD, Kato N, Michell RS, Semba CP, Miller DC. Endovascular stent-graft placement for the treatment of acute aortic dissection. N Engl J Med 1999;340:1546 1552. [7] Nienhaber CA, Fattori R, Lund G, Dieckmann CH, Wolf W, Von Kodolitsch Y, Pierangeli A. Nonsurgical reconstruction of thoracic aortic dissection by stent-graft placement. N Engl J Med 1999;340: 1539 1545. [8] Tefera G, Acher C, Wynn M. Clamp and sew techniques in thoracoabdominal aortic surgery using naloxone and CSF drainage. Semin Vasc Surg 2000;13(4):325 330. [9] Blum U, Voshage G, Lammer J. Endoluminal stent-grafts für infrarenal abdominal aortic aneurysms. N Engl J Med 1997;336:13 20. [10] Mitchell RS, Dake MD, Semba CP. Endovascualar stent-graft repair of thoracic aortic aneurysm. J Thorac Cardiovasc Surg 1996;111: 1054 1062. [11] Razavi MK, Nishimura E, Slonim S, Zeigler W, Kee S, Witheerall HL, Semba CP, Dake MD. Percutaneous creation of acute type B aortic dissection : an experimental model for endoluminal therapy. J Vasc Interv Radiol 1998;9:626 632. [12] Miller DC. Surgical management of aortic dissections: indications, perioperative management, and long-term results. Doroghazi RM, Slater EE, editors. Aortic dissection. New York, NY: McGraw-Hill, 1983. pp. 193 243. [13] Juvonen T, Ergin MA, Galla JD, et al. Risk factors for rupture of chronic type B dissections. J Thorac Cardiovasc Surg 1999; 117(4):776 786. [14] Czermark BV, Waldenberger P, Fraedrich G, Dessl AH, Roberts KE, Bale RJ, Perkmann R, Jaschke WR. Treatment of Stanford type B aortic dissection with stent-grarfts: preliminary results. Radiology 2000;217:544 550. [15] Coselli J, LeMaire S, Schmittling Z, Köksoy C. Cerebrospinal fluid drainage in thoracoabdominal aortic surgery. Semin Vasc Surg 2000;13(4):308 314. [16] Huynh D, Miller III C, Safi H. Delayed onset of neurologic deficit: significance and management. Semin Vasc Surg 2000;13(4):325 330.