- Email: [email protected]
Sealing of Type III Endoleaks with Ethylene Vinyl Alcohol Copolymer in a Canine Model
Laboratory Investigation
Sealing of Type III Endoleaks with Ethylene Vinyl Alcohol Copolymer in a Canine Model Romualdo Maffra, Jr, MD, Alvin U. Anene, BS, Mark Sands, MD, Yong-Hua Dong, MD, William Davros, PhD, Lucas H. Brennecke, DVM, DACVP, and Bart L. Dolmatch, MD
PURPOSE: To test ethylene vinyl alcohol copolymer (EVOH) as a sealing agent for persistent abdominal aortic aneurysm (AAA) endograft leaks. MATERIALS AND METHODS: Twelve dogs underwent creation of AAAs with a Palmaz P4014 stent. A 10-mm ⴛ 5-cm Wallgraft endoprosthesis with a 4-mm-diameter hole cut into its side was deployed within the AAA. One week later, computed tomography (CT) and angiography were performed and the aneurysm sac was catheterized through the 4-mm hole. Then, EVOH was injected into the sac and lumbar arteries. Four weeks thereafter, all surviving animals underwent repeat CT scanning and angiography and were then euthanized. The AAA underwent gross and microscopic study. RESULTS: Three dogs died from aortic rupture within 24 hours of AAA creation and the remaining nine dogs survived to receive EVOH. All nine dogs had persistent flow into the sac and lumbar arteries at the time of EVOH delivery. Seven dogs survived to the end of the experiment, and all aneurysm sacs and lumbar arteries remained occluded on angiography and CT. Histologic examination revealed EVOH and thrombus admixed, with thrombus in varying stages of organization filling the aneurysm sac and lumbar arteries. CONCLUSIONS: Embolization of type III endoleaks with EVOH proved to be feasible in a canine model. Further work is warranted to determine its therapeutic utility. J Vasc Interv Radiol 2007; 18:763–769 Abbreviations:
AAA ⫽ abdominal aortic aneurysm, EVOH ⫽ ethylene vinyl alcohol copolymer
AN abdominal aortic aneurysm (AAA) is defined as focal aortic dilation greater than 50% compared with the expected normal diameter of the aorta (1). The natural history of AAA is one of progressive enlargement resulting in rupture in a significant percentage of patients. Five to eight percent of men older than 65 years have an abdominal aneurysm, and the total pop-
From the Department of Radiology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390. Received September 6, 2006; final revision received March 5, 2007; accepted March 12, 2007. From the SCVIR 2000 Annual Meeting. Address correspondence to A.U.A.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2007 DOI: 10.1016/j.jvir.2007.03.003
ulation-based mortality rate from ruptured AAA is approximately 85% (2,3). The goal of AAA treatment is to depressurize the aneurysm sac, which prevents expansion, encourages thrombus formation within the sac, and often leads to shrinkage of the sac. Open surgical repair, the standard AAA treatment, carries a mortality risk between 1.1% 7.0% (4 – 6). This may increase to 20% when severe comorbid illness is present (7–9). In addition, conventional exposure of the infrarenal aorta necessitates a large abdominal incision, mobilization of the abdominal viscera, and retroperitoneal dissection. Endoluminal stent-graft insertion is a less invasive method that is being used with increasing frequency to treat AAA. Compared with conventional surgery, it is less painful and has a faster recovery time (10 –12). It is
hoped that endografting will reduce operative risks and provide a durable repair. However, in some cases, there are problems associated with endograft placement that are not seen with conventional surgery. One particular problem is persistent blood flow between the endograft and aneurysm sac, which is referred to as endoleak (13). To understand the types of endoleak treatments available, it is important to understand the different causes of endoleaks. One widely accepted classification defines five types of endoleak (14). In this schema, endoleaks related to the attachment site (proximal or distal) are called type IA and type IB, respectively. Those caused by a retrograde flow from collateral vessel leaks are type II. Those caused by a defect in the graft fabric or distraction of endograft components are considered type III. Type IV en-
763
764
•
June 2007
Type III Endoleak Seal with EVOH in a Canine Model
doleaks are caused by graft fabric porosity and are nearly always self-limited and do not require treatment. Those caused by tension exerted on the aneurysm wall (ie, endotension) are type V. Type I and III endoleaks are typically approached by placement of a stent or covered stent at the attachment site or endograft defect. When this does not work, transcatheter occlusion procedures may be attempted. For endoleaks that are not occluded with additional stents or covered stents, or with transcatheter techniques, open surgery may be necessary because endoleaks may lead to aneurysm growth and possible rupture (15,16). Many type I, II, and III endoleaks involve branch vessels arising from the AAA sac. In these situations, translumbar or transarterial embolization has been used with procedural success typically delivering occlusive coils into the AAA sac or collateral vessels (16 –18). However, coil embolization has been questioned for its long-term durability because experimental evidence shows that, after coil embolization, the intraaneurysmal pressure may not be reduced (19 –21). If this is true, coil-embolized AAAs may remain at risk for delayed rupture. In addition, coil placement may be nonuniform, leaving large interstices where thrombus may be prone to recanalization. Recognizing that endoleaks are part of endograft procedures and that the ideal treatment of endoleaks may not yet exist, we designed the present experiment with a type III endoleak canine model with persistent lumbar artery flow. AAAs were created by overdilation of balloon-expandable stents in the abdominal aorta; then, a fenestrated covered stent was placed within the AAA. To seal the type III endoleak created by the fenestration, we used ethylene vinyl alcohol copolymer (EVOH) 8% (Onyx; ev3, Plymouth, Minn). EVOH is a copolymer dissolved in dimethyl sulfoxide with suspended micronized tantalum power for radiopacity. It is delivered as a liquid, and on contact with aqueous solution (blood in this case), changes state into a caulk-like semisolid. Our primary purpose was to study the use of EVOH to provide a durable occlusion of AAA endoleaks. Four secondary goals of this study were evaluated as well: (i) validation
JVIR
Figure 1. Chronologic procedure and mortality graph.
of a model for the study of leaking AAA endografts, (ii) assessment of EVOH delivery, (iii) confirmation that EVOH can acutely occlude leaking stent-grafts, and (iv) evaluation of the tissue response and inflammatory changes related to EVOH. We created AAA endoleaks based on a canine AAA model previously reported (1,20,22) with some changes we thought appropriate, as described in further sections.
MATERIALS AND METHODS This study was approved by our institutional animal care and use committee. Twelve male mongrel dogs weighing 20 –32 kg were housed and maintained in facilities approved by the American Association for the Accreditation of Laboratory Animal Care. All animals were quarantined for 1 week before the first procedure. A summary of the experiment is presented graphically in Figure 1. Aneurysm Creation and Endoleak Each dog fasted for 12 hours before each invasive procedure and was anesthetized with intravenous sodium pentobarbital (18 –20 mg/kg body weight), intubated, and given inhalation anesthesia maintained with 1.5%– 2.5% isoflurane. Then, 1 g of cefazolin was administered, the neck was prepared in a sterile manner, and a midline neck incision was made to expose the and right common carotid artery. Arteriotomy was made after proximal and distal control of the carotid artery had been obtained, and a 30-cm, 11-F introducer sheath (Hemostatic Introducer; Daig, Minnetonka, Minn) was introduced. Intravenous heparin (75 U/kg) was administered to prevent
pericatheter thrombosis, and an 8-F guide catheter (Multipurpose; Bard, Galway, Ireland) was introduced over a 0.035-inch, 3-mm J-tipped guide wire into the descending thoracic aorta. An abdominal aortogram was obtained with use of 20 mL of contrast medium. The diameter of the infrarenal aorta was measured with use of a radiopaque ruler placed over the abdomen of the dog. The mean aortic diameter measured 8.3 mm, with a range of 8 –9 mm. These measurements were confirmed by subsequent spiral computed tomography (CT). To create the AAA, a P4014 Palmaz stent (Cordis/Johnson & Johnson, Warren, NJ), previously cut to 3 cm in length and resterilized, was deployed into the infrarenal abdominal aorta on a 16-mm ⫻ 2-cm angioplasty balloon (XXL balloon dilation catheter; Boston Scientific, Watertown, Mass) under fluoroscopic guidance. The site of stent placement was chosen according to the projection of the lumbar arteries so that at least one pair of lumbar arteries arose from the middle of the stent-implanted aneurysm. The balloon was removed and an angiogram was obtained. Diameter of the aneurysm was determined by spiral CT 1 week after completion of the stent implantation procedure. The mean diameter at the middle of the stent-implanted aneurysm was 15.6 mm, with a range of 14.2–17.0 mm. The mean aneurysm length was 40 mm and the mean length of the infrarenal aorta was 9.4 mm. Aneurysm creation was immediately followed by endograft placement. A 10-mm ⫻ 5-cm Wallgraft endoprosthesis (Schneider, Minneapolis, Minn) with a 4-mm-diameter hole at
Volume 18
Number 6
Maffra et al
the middle of the graft was advanced over a guide wire to the aneurysm site. The Wallgraft was deployed so the hole was positioned toward the posterior aorta wall. Another angiogram was obtained to confirm the presence of endoleak through the hole in the Wallgraft. In all cases, endoleak was noted with filling of the lumbar arteries (Fig 2). Among the nine dogs included in the final cohort, we found three dogs with two pairs of lumbar arteries rising from the aneurysm sac. All catheters were then removed and the carotid artery was repaired with running 6 – 0 polypropylene suture. The neck incision was closed with 3– 0 polygalactin suture. All dogs received cefazolin 1 g three times a day and aspirin 325 mg three times a day for 72 hours and were housed at our animal research facility for the duration of the experiment. Heparin was not reversed at the end of the procedure.
765
Figure 3. Aortogram following EVOH treatment of the AAA sac. The large arrow indicates EVOH in the sac, which appears less radiodense than the angiographic contrast agent. The small arrow indicates EVOH in one of the lumbar arteries.
Endoleak Seal One week after AAA creation and implantation of the endograft, the dogs were anesthetized and CT scanning of the abdominal aorta was performed. Thereafter, each dog was taken to the angiography suite, where the left common carotid artery was exposed and catheterized. An angiogram was obtained to confirm persistence of the endoleak, noted in all surviving dogs. A 5-F RIM catheter (Infiniti angiography catheter; Cordis Endovascular, Warren, NJ) was advanced from the common carotid artery to the 4-mm hole in the Wallgraft where contrast agent injection allowed visualization of the persistent endoleak, the AAA sac, and the runoff lumbar arteries. Next, a 0.018-inch guide wire and a dimethyl sulfoxide– compatible microcatheter (Micro Therapeutics, Irvine, Calif) was passed through the 5-F catheter and across the hole in the Wallgraft. A priming volume of 0.35 mL dimethyl sulfoxide was injected into the microcatheter to prevent earlier solidification of the EVOH. Then, EVOH 8% was injected into the lumbar arteries and aneurysm sac under fluoroscopic guidance. We decided to stop the EVOH injection at the time that the lumbar arteries and the aneu-
•
Figure 2. Leaky Wallgraft within AAA model. Arrows point to lumbar arteries filling from the patent AAA sac.
rysm sac were completely filled. In our experiment, a mean volume of 0.7 mL of EVOH was injected (range, 0.4 – 1.0 mL). At the end of the procedure, an aortogram was obtained to confirm occlusion of the endoleak (Fig 3). The carotid arteriotomy and incision were sutured, and once again the dogs were awakened from anesthesia and allowed to recover. All surviving animals received cefazolin 1 g three times per day and aspirin 325 mg three times per day for 72 hours. CT Scanning All surviving dogs underwent CT imaging immediately before EVOH delivery (1 week after stent-implanted AAA creation and leaky Wallgraft placement) and then 1 week and 5 weeks after EVOH delivery. Scanning was performed on a Somatom Plus 4 scanner (Siemens, Erlangen, Germany) that obtained 3.0-mm-thick contiguous slices with and without intravenous contrast agent administration.
Images were reviewed and measured by an independent investigator who had no knowledge of the angiographic findings (W.D.). Because of unanticipated mortality in two of the nine dogs that survived to the EVOH procedure, final CT data at 5 weeks after embolization were available for seven dogs. CT data 1 week after EVOH were available for the dog that died of anesthetic complications, and CT data at 2 days after EVOH were available for the dog that developed paraplegia. Termination After CT at 5 weeks after EVOH delivery, each animal underwent final angiography and was euthanized. The retroperitoneum was inspected for any evidence of hematoma and the segment of the infrarenal aorta containing the aneurysm, from the renal arteries to the aortic trifurcation, was harvested and preserved in 10% formalin for histologic examination. Histopathologic Methods Gross pathologic and histologic examination of the treated AAA was performed by one author (L.H.B.) at an independent laboratory without information regarding angiographic or CT findings. Lumbar arteries were processed through and embedded in
766
•
June 2007
Type III Endoleak Seal with EVOH in a Canine Model
JVIR
Figure 4. Cross-section schematic of histologic analysis sites.
paraffin and then sectioned and stained with hematoxylin and eosin or Masson trichrome stains. Photomicrographs and gross photographs were prepared. Five sites for each explant were evaluated (Fig 4).
RESULTS We succeeded in creating stent-implanted AAAs with a patent Wallgraft endoleak in nine of our 12 dogs; the remaining three dogs died of aortic rupture related to AAA creation. Regarding the three deaths, one occurred at the end of the aneurysm/endoleak procedure, one the night of the procedure, and one 3 days after the procedure. Necropsy confirmed aortic rupture in all three animals. All nine dogs that survived AAA creation and Wallgraft placement had endoleaks into the sac and lumbar arteries, confirmed by angiography and CT imaging (Fig 2) 1 week later. After EVOH delivery, acute occlusion of the lumbar arteries and aneurysm sac was noted in all nine dogs by angiography. For the seven dogs that survived to the end of the experiment, CT and angiography confirmed occlusion of the AAA sac and lumbar arteries, even in the three dogs with two pairs of lumber arteries. There were three complications in the nine-dog cohort that survived AAA creation and Wallgraft placement. The first dog treated with EVOH manifested paraplegia on the first day after treatment as a result of an excess of EVOH infused into the distal lumbar arteries. The veterinary director of the animal research facility suggested observation for several
days. Therefore, the animal was observed for 3 additional days. After showing no improvement, this dog was euthanized after CT scanning and the aneurysm was resected on the fourth day after EVOH delivery. Another dog died 1 week after delivery of EVOH after undergoing a scheduled 1-week CT examination. The cause of death was believed related to complications of anesthesia. No anatomic abnormalities were noted at necropsy and the abdominal aortic aneurysm had not ruptured. A nonfatal complication occurred during successful delivery of EVOH when a small amount of EVOH refluxed through the 4-mm Wallgraft hole and embolized into the femoral artery. Reflux occurred because of overinjection of EVOH, and this complication was avoided in the remaining procedures. The EVOH embolus was easily seen by fluoroscopy and uneventfully removed with a sterile urologic stone retrieval basket. CT Scanning In all CT scans after EVOH delivery, the AAA sac was filled with radiodense EVOH, outflow lumbar arteries contained EVOH, and no endoleaks were seen (Fig 5). Histopathologic Results Gross examination of all specimens showed a smooth layer of neointima covering the luminal surface of the Wallgraft. In many cases the 4-mm hole at the middle of the Wallgraft was covered with tissue and could not be found. There was no evidence of rup-
Figure 5. CT scan shows EVOH in the posterior aspect of the AAA sac (large arrow) and in the lumbar arteries (small arrows).
ture or retroperitoneal hematoma in any of the specimens of the seven dogs that survived to the end of the experiment, the dog that died 1 week after EVOH, or the dog that was killed 4 days after EVOH. EVOH was easily identified in the AAA sac and the transected lumbar arteries by its dark gray appearance (Fig 6). Histologic analysis revealed that the space between the aneurysm wall and the Wallgraft was filled with EVOH mixed with thrombus of varying states of organization (Fig 7a,c). In some sections, inflammatory cells (typically macrophages) were found, mixed with thrombus. There were few changes in mural architecture of the aorta or the lumbar arteries. There was some degeneration of the smooth muscle of the aortic wall, in some cases replaced by fibrous connective tissue. Those findings were related to the stent placement and aneurysm creation. Lumbar arteries were filled with a plug of EVOH admixed with fibrous thrombus (Fig 7b). There was no evidence of intramural changes in the lumbar arteries. There were no vascular channels or evidence of persistent flow in any of the sections taken from the AAAs or lumbar arteries.
DISCUSSION Endoleaks are a significant technical problem associated with endoluminal repair of AAAs. In different reports, the incidences of endoleak
Volume 18
Number 6
Maffra et al
Figure 6. Bivalved en-bloc AAA explant at 5 weeks. EVOH is seen in this gross specimen as a homogenous gray filler in the wall of the AAA sac. EVOH runs nearly the entire length of the AAA (arrow).
have ranged between 8% and 44% (14,15,23). For treatment of endoleaks, many authors prefer to follow early endoleaks because they believe that a great majority of them will seal spontaneously. If spontaneous closure of the endoleak does not occur, it is often necessary to determine the type of endoleak. Type II endoleaks without AAA enlargement may be followed, whereas persistent type I and III endoleaks warrant treatment, as do type II endoleaks with enlarging AAA sacs. These types of endoleaks are considered unstable and pose risk for aneurysm rupture (24). Certainly, endografts are not designed and inserted to leak, and an endoleak represents failure of primary therapy directed at the AAA. Amongst those who believe persistent endoleaks warrant treatment, there is still debate as to the best type of intervention. Endoleaks may result from incomplete sealing at the end(s) of the endograft, delayed detachment at the terminal attachment sites, persistent or delayed retrograde collateral filling of the aneurysm sac, or delayed defects in the endograft itself (13,25). Transcatheter occlusion of endoleaks has been performed with coils, thrombin, and glue. The use of coils has been most widely reported (16 – 18). However, in one report, coils were not able to successfully reduce the intraaneurysmal sac pressure in a canine model (13) and the potential for aneurysm rupture was thought to persist after angiographic sac thrombosis. In the present study we tested the ability
of a novel embolic agent to seal a type III endoleak in a canine model. This agent, EVOH, changes its physical state on contact with aqueous solution (ie, blood) and becomes a plastic semisolid after it mixes with blood. We found that EVOH was capable of sealing leaks within the aneurysm sac and runoff lumbar collateral vessels, and maintained stable occlusion for 5 weeks with no evidence of recurrent endoleak. Histologic examination showed that EVOH mixed with thrombus and that the thrombus showed varying degrees of organization. We recognize that one of the limitations of our study is the short time span (5 weeks) for the observation of changes in the occluded sac and lumber arteries. In addition, our model is very “synthetic,” in that we created an aneurysm with a very small sac but a large endoleak. The endoleak, a 4-mm hole in the midportion of the Wallgraft, was artificially large and designed to sustain the endoleak until treatment with EVOH and permit easy catheterization of the AAA sac. For some endoleaks, such as those type II leaks that arise from collateral vessels without any attachment or graft defect leak, direct catheterization may be difficult. However, it is possible to employ a translumbar embolization technique as described by Baum et al (26,27). Another limitation of this work is the lack of pressure data from within the sac. Certainly, the concerns raised in previous studies (19 –21) regarding persistently increasd sac pressures af-
•
767
ter occlusive coil embolization could also apply to EVOH embolization if sac pressures were to remain increased. However, although our methodology did not allow for placement of sac pressure transducers, the ability of EVOH embolization to successfully reduce intraaneurysmal sac pressure has been described (28). This is most likely attributed to the ability of EVOH to form a durable and complete seal of the endoleak, in contrast to the use of coils, which may allow persistent perfusion amid the coils resulting in continued pressurization of the sac (28,29). Despite limitations caused by the study’s short time span, there are certain advantages to our model. These include quick and easy AAA creation, patency of endoleak flow into the AAA sac and lumbar arteries, straightforward explantation of the treated AAA, and most importantly the ability of the liquid embolic agent to completely seal the endoleak. However, these advantages are tempered by the loss of three dogs (ie, 25% mortality rate) during the early phase of the experiment when the AAA was created. In regard to the three animals that died of aneurysm rupture during AAA creation, the model of creating AAAs in dogs called for the use of oversized balloon-dilated stents. Gross inspection showed that the combined effects of overstretching resulting from the the balloon and the rigid ends of the stent caused aortic laceration. This is the singular drawback to this method of AAA creation. But although a 25% mortality rate related to AAA formation is substantial, other investigators have described the use of a similar technique of AAA creation with comparable mortality (1,22,30). Animals that survived AAA creation had stable AAAs with preserved lumbar arteries. Our final cohort of nine animals were ultimately treated for type III endoleaks, and we remain convinced that balloon stent overdilation is a reasonable method for the study of type III endoleaks in canine AAAs. Advantages of this method include simple creation of AAAs with use of percutaneous methods without the need for surgical techniques other than carotid artery cutdown, and also preservation of lumbar arteries. Regarding the delivery of EVOH, there was no difficulty in visualizing
768
•
June 2007
Type III Endoleak Seal with EVOH in a Canine Model
JVIR
Figure 7. (a) Cross-section micrograph (hematoxylin and eosin, magnification ⫻2.5) corresponding to location 2 in the histology schematic (Fig 4). EVOH appears black and lacelike, and it is admixed within thrombus in varying stages of organization. The transition between EVOH and the aneurysm sac wall (arrow) is free from inflammatory infiltrates. (b) Cross-section micrograph (hematoxylin and eosin, magnification ⫻10) corresponding to location 1 in the histology schematic (Fig 4). EVOH fills the lumbar artery lumen and is mixed with thrombus in varying stages of organization (arrow).
or infusing this agent. However, overly eager delivery probably contributed to the one case of paraplegia and one case of reflux embolization. We supposed that the case of paraplegia was a result of some kind of communication between the distal lumbar arteries and the medullary arteries in dogs. We believe careful and judicious use of any embolic or occlusive agent is warranted during occlusion of lumbar arteries. There is no embolic agent in clinical use today that carries no risk of inadvertent misplacement or unanticipated ischemia. However, EVOH can be delivered in a slower, more controlled manner that may reduce those risks compared with coil embolization or the use of other liquid embolic agents (31). Histologic evaluation demonstrated varying degrees of inflammation. There are many reasons for inflammatory cell infiltrates in the model AAA sac at 5 weeks. Some possibilities include the inflammatory nature of polyethylene terephthalate (ie, the Wallgraft), inflammation from aortic stretch injury after placement of the Palmaz stent, ongoing clearance and organization of thrombus within the sac, and the presence of EVOH itself. Nevertheless, macrophage infiltrates may not be “bad” and may actually represent the early phase in long-term occlusion of the endoleak, with thrombus organization and adherence to the
polyethylene terephthalate of the endograft. The observation of organized thrombus in many histologic sections is encouraging, but long-term studies are needed to declare this procedure a durable fix for endoleaks. Our model shows the feasibility of endoleak sealing with EVOH for type III endoleaks. The polymeric seal caused by EVOH was durable to 5 weeks without recurrence of endoleak by CT or angiography or at postmortem evaluation. Longer studies are needed to address the question of whether EVOH will ultimately prove successful in providing a durable seal of endoleak after endovascular AAA treatment in humans. References 1. Hallisey MJ. 1997 SCVIR Gary J. Becker Young Investigator Award: a transluminally created abdominal aortic aneurysm model. J Vasc Interv Radiol 1997; 8:305–312. 2. Balm R, Eikelboom BC, May J, Bell PR, Swedenborg J, Collin J. Early experience with transfemoral endovascular aneurysm management (TEAM) in the treatment of aortic aneurysms. Eur J Vasc Endovasc Surg 1996; 11:214 –220. 3. Beard JD. Screening for abdominal aortic aneurysm. Br J Surg 2003; 90: 515–516. 4. Ernst CB. Abdominal aortic aneurysm. N Engl J Med 1993; 328:1167– 1172. 5. Gorham TJ, Taylor J, Raptis S. Endovascular treatment of abdominal
6.
7.
8.
9.
10.
11.
12.
aortic aneurysm. Br J Surg 2004; 91: 815– 827. Prinssen M, Verhoeven EL, Buth J, et al. A randomized trial comparing conventional and endovascular repair of abdominal aortic aneurysms. N Engl J Med 2004; 351:1607–1618. Becker GJ, Kovacs M, Mathison MN. Transluminal repair of abdominal aortic aneurysm: a call for selective use, careful surveillance, new device design, and systematic study of transrenal fixation. J Vasc Surg 2002; 35:611– 615. Mortality results for randomised controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. The UK Small Aneurysm Trial Participants. Lancet 1998; 352:1649 –1655. Hausegger KA, Schedlbauer P, Deutschmann HA, Tiesenhausen K. Complications in endoluminal repair of abdominal aortic aneurysms. Eur J Radiol 2001; 39:22–33. Schlensak C, Doenst T, Hauer M, et al. Serious complications that require surgical interventions after endoluminal stent-graft placement for the treatment of infrarenal aortic aneurysms. J Vasc Surg 2001; 34:198 –203. Greenhalgh RM, Brown LC, Kwong GP, Powell JT, Thompson SG. Comparison of endovascular aneurysm repair with open repair in patients with abdominal aortic aneurysm (EVAR trial 1), 30-day operative mortality results: randomised controlled trial. Lancet. 2004; 364:843– 848. Maher MM, McNamara AM, MacEneaney PM, Sheehan SJ, Malone DE. Abdominal aortic aneurysms: elective
Volume 18
13. 14.
15.
16.
17.
18.
19.
Number 6
endovascular repair versus conventional surgery—evaluation with evidence-based medicine techniques. Radiology. 2003; 228:647– 658. Golzarian J, Maes EB, Sun S. Endoleak: treatment options. Tech Vasc Interv Radiol 2005; 8:41– 49. Baum RA, Stavropoulos SW, Fairman RM, Carpenter JP. Endoleaks after endovascular repair of abdominal aortic aneurysms. J Vasc Interv Radiol 2003; 14:1111–1117. Wain RA, Marin ML, Ohki T, et al. Endoleaks after endovascular graft treatment of aortic aneurysms: classification, risk factors, and outcome. J Vasc Surg 1998; 27:69 –78; Choke E, Thompson M. Endoleak after endovascular aneurysm repair: current concepts. J Cardiovas Surg 2004; 45:349 –366. Gorich J, Rilinger N, Sokiranski R, et al. Treatment of leaks after endovascular repair of aortic aneurysms. Radiology 2000; 215:414 – 420. Baum RA, Carpenter JP, Tuite CM, et al. Diagnosis and treatment of inferior mesenteric arterial endoleaks after endovascular repair of abdominal aortic aneurysms. Radiology 2000; 215: 409 – 413. Solis MM, Ayerdi J, Babcock GA, et al. Mechanism of failure in the treatment of type II endoleak with percutaneous
Maffra et al
20.
21.
22.
23.
24.
25.
coil embolization. J Vasc Surg 2002; 36:485– 491. Marty B, Sanchez LA, Ohki T, et al. Endoleak after endovascular graft repair of experimental aortic aneurysms: does coil embolization with angiographic “seal” lower intraaneurysmal pressure? J Vasc Surg 1998; 27:454 – 461. Sheehan MK, Barbato J, Compton CN, Zajko A, Rhee R, Makaroun MS. Effectiveness of coiling in the treatment of endoleaks after endovascular repair. J Vasc Surg 2004; 40:430 – 434. Pavcnik D, Andrews RT, Yin Q, et al. A canine model for studying endoleak after endovascular aneurysm repair. J Vasc Interv Radiol 2003; 14:1303–1310. White GH, Yu W, May J, Chaufour X, Stephen MS. Endoleak as a complication of endoluminal grafting of abdominal aortic aneurysms: classification, incidence, diagnosis, and management. J Endovasc Surg 1997; 4:152–168. Faries PL, Cadot H, Agarwal G, Kent KC, Hollier LH, Marin ML. Management of endoleak after endovascular aneurysm repair: cuffs, coils, and conversion. J Vasc Surg 2003; 37: 1155–1161. White GH, May J, Waugh RC, Chaufour X, Yu W. Type III and type IV endoleak: toward a complete definition of blood flow in the sac after endolu-
26.
27.
28.
29.
30.
31.
•
769
minal AAA repair. J Endovasc Surg 1998; 5:305–309. Baum RA, Carpenter JP, Golden MA, et al. Treatment of type 2 endoleaks after endovascular repair of abdominal aortic aneurysms: comparison of transarterial and translumbar techniques. J Vasc Surg 2002; 35:23–29. Baum RA, Cope C, Fairman RM, Carpenter JP. Translumbar embolization of type 2 endoleaks after endovascular repair of abdominal aortic aneurysms. J Vasc Interv Radiol 2001; 12: 111–116. Martin ML, Dolmatch BL, Fry PD, Machan LS. Treatment of type II endoleaks with EVOH. J Vasc Interv Radiol 2001; 12:629 – 632. Abdala N, Levitin A, Dawson A, et al. Use of ethylene vinyl alcohol copolymer for tubal sterilization by selective catheterization in rabbits. J Vasc Interv Radiol 2001; 12:979 –984. Skillern CS, Stevens SL, Piercy KT, et al. Endotension in an experimental aneurysm model. J Vasc Surg 2002; 36: 814 – 817. Cantasdemir M, Kantarci F, Mihmanli I, Numan F. Embolization of profunda femoris artery branch pseudoaneurysms with ethylene vinyl alcohol copolymer (EVOH). J Vasc Interv Radiol 2002; 13:725–728.
Sealing of Type III Endoleaks with Ethylene Vinyl Alcohol Copolymer in a Canine Model Romualdo Maffra, Jr, MD, Alvin U. Anene, BS, Mark Sands, MD, Yong-Hua Dong, MD, William Davros, PhD, Lucas H. Brennecke, DVM, DACVP, and Bart L. Dolmatch, MD
PURPOSE: To test ethylene vinyl alcohol copolymer (EVOH) as a sealing agent for persistent abdominal aortic aneurysm (AAA) endograft leaks. MATERIALS AND METHODS: Twelve dogs underwent creation of AAAs with a Palmaz P4014 stent. A 10-mm ⴛ 5-cm Wallgraft endoprosthesis with a 4-mm-diameter hole cut into its side was deployed within the AAA. One week later, computed tomography (CT) and angiography were performed and the aneurysm sac was catheterized through the 4-mm hole. Then, EVOH was injected into the sac and lumbar arteries. Four weeks thereafter, all surviving animals underwent repeat CT scanning and angiography and were then euthanized. The AAA underwent gross and microscopic study. RESULTS: Three dogs died from aortic rupture within 24 hours of AAA creation and the remaining nine dogs survived to receive EVOH. All nine dogs had persistent flow into the sac and lumbar arteries at the time of EVOH delivery. Seven dogs survived to the end of the experiment, and all aneurysm sacs and lumbar arteries remained occluded on angiography and CT. Histologic examination revealed EVOH and thrombus admixed, with thrombus in varying stages of organization filling the aneurysm sac and lumbar arteries. CONCLUSIONS: Embolization of type III endoleaks with EVOH proved to be feasible in a canine model. Further work is warranted to determine its therapeutic utility. J Vasc Interv Radiol 2007; 18:763–769 Abbreviations:
AAA ⫽ abdominal aortic aneurysm, EVOH ⫽ ethylene vinyl alcohol copolymer
AN abdominal aortic aneurysm (AAA) is defined as focal aortic dilation greater than 50% compared with the expected normal diameter of the aorta (1). The natural history of AAA is one of progressive enlargement resulting in rupture in a significant percentage of patients. Five to eight percent of men older than 65 years have an abdominal aneurysm, and the total pop-
From the Department of Radiology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390. Received September 6, 2006; final revision received March 5, 2007; accepted March 12, 2007. From the SCVIR 2000 Annual Meeting. Address correspondence to A.U.A.; E-mail: [email protected] None of the authors have identified a conflict of interest. © SIR, 2007 DOI: 10.1016/j.jvir.2007.03.003
ulation-based mortality rate from ruptured AAA is approximately 85% (2,3). The goal of AAA treatment is to depressurize the aneurysm sac, which prevents expansion, encourages thrombus formation within the sac, and often leads to shrinkage of the sac. Open surgical repair, the standard AAA treatment, carries a mortality risk between 1.1% 7.0% (4 – 6). This may increase to 20% when severe comorbid illness is present (7–9). In addition, conventional exposure of the infrarenal aorta necessitates a large abdominal incision, mobilization of the abdominal viscera, and retroperitoneal dissection. Endoluminal stent-graft insertion is a less invasive method that is being used with increasing frequency to treat AAA. Compared with conventional surgery, it is less painful and has a faster recovery time (10 –12). It is
hoped that endografting will reduce operative risks and provide a durable repair. However, in some cases, there are problems associated with endograft placement that are not seen with conventional surgery. One particular problem is persistent blood flow between the endograft and aneurysm sac, which is referred to as endoleak (13). To understand the types of endoleak treatments available, it is important to understand the different causes of endoleaks. One widely accepted classification defines five types of endoleak (14). In this schema, endoleaks related to the attachment site (proximal or distal) are called type IA and type IB, respectively. Those caused by a retrograde flow from collateral vessel leaks are type II. Those caused by a defect in the graft fabric or distraction of endograft components are considered type III. Type IV en-
763
764
•
June 2007
Type III Endoleak Seal with EVOH in a Canine Model
doleaks are caused by graft fabric porosity and are nearly always self-limited and do not require treatment. Those caused by tension exerted on the aneurysm wall (ie, endotension) are type V. Type I and III endoleaks are typically approached by placement of a stent or covered stent at the attachment site or endograft defect. When this does not work, transcatheter occlusion procedures may be attempted. For endoleaks that are not occluded with additional stents or covered stents, or with transcatheter techniques, open surgery may be necessary because endoleaks may lead to aneurysm growth and possible rupture (15,16). Many type I, II, and III endoleaks involve branch vessels arising from the AAA sac. In these situations, translumbar or transarterial embolization has been used with procedural success typically delivering occlusive coils into the AAA sac or collateral vessels (16 –18). However, coil embolization has been questioned for its long-term durability because experimental evidence shows that, after coil embolization, the intraaneurysmal pressure may not be reduced (19 –21). If this is true, coil-embolized AAAs may remain at risk for delayed rupture. In addition, coil placement may be nonuniform, leaving large interstices where thrombus may be prone to recanalization. Recognizing that endoleaks are part of endograft procedures and that the ideal treatment of endoleaks may not yet exist, we designed the present experiment with a type III endoleak canine model with persistent lumbar artery flow. AAAs were created by overdilation of balloon-expandable stents in the abdominal aorta; then, a fenestrated covered stent was placed within the AAA. To seal the type III endoleak created by the fenestration, we used ethylene vinyl alcohol copolymer (EVOH) 8% (Onyx; ev3, Plymouth, Minn). EVOH is a copolymer dissolved in dimethyl sulfoxide with suspended micronized tantalum power for radiopacity. It is delivered as a liquid, and on contact with aqueous solution (blood in this case), changes state into a caulk-like semisolid. Our primary purpose was to study the use of EVOH to provide a durable occlusion of AAA endoleaks. Four secondary goals of this study were evaluated as well: (i) validation
JVIR
Figure 1. Chronologic procedure and mortality graph.
of a model for the study of leaking AAA endografts, (ii) assessment of EVOH delivery, (iii) confirmation that EVOH can acutely occlude leaking stent-grafts, and (iv) evaluation of the tissue response and inflammatory changes related to EVOH. We created AAA endoleaks based on a canine AAA model previously reported (1,20,22) with some changes we thought appropriate, as described in further sections.
MATERIALS AND METHODS This study was approved by our institutional animal care and use committee. Twelve male mongrel dogs weighing 20 –32 kg were housed and maintained in facilities approved by the American Association for the Accreditation of Laboratory Animal Care. All animals were quarantined for 1 week before the first procedure. A summary of the experiment is presented graphically in Figure 1. Aneurysm Creation and Endoleak Each dog fasted for 12 hours before each invasive procedure and was anesthetized with intravenous sodium pentobarbital (18 –20 mg/kg body weight), intubated, and given inhalation anesthesia maintained with 1.5%– 2.5% isoflurane. Then, 1 g of cefazolin was administered, the neck was prepared in a sterile manner, and a midline neck incision was made to expose the and right common carotid artery. Arteriotomy was made after proximal and distal control of the carotid artery had been obtained, and a 30-cm, 11-F introducer sheath (Hemostatic Introducer; Daig, Minnetonka, Minn) was introduced. Intravenous heparin (75 U/kg) was administered to prevent
pericatheter thrombosis, and an 8-F guide catheter (Multipurpose; Bard, Galway, Ireland) was introduced over a 0.035-inch, 3-mm J-tipped guide wire into the descending thoracic aorta. An abdominal aortogram was obtained with use of 20 mL of contrast medium. The diameter of the infrarenal aorta was measured with use of a radiopaque ruler placed over the abdomen of the dog. The mean aortic diameter measured 8.3 mm, with a range of 8 –9 mm. These measurements were confirmed by subsequent spiral computed tomography (CT). To create the AAA, a P4014 Palmaz stent (Cordis/Johnson & Johnson, Warren, NJ), previously cut to 3 cm in length and resterilized, was deployed into the infrarenal abdominal aorta on a 16-mm ⫻ 2-cm angioplasty balloon (XXL balloon dilation catheter; Boston Scientific, Watertown, Mass) under fluoroscopic guidance. The site of stent placement was chosen according to the projection of the lumbar arteries so that at least one pair of lumbar arteries arose from the middle of the stent-implanted aneurysm. The balloon was removed and an angiogram was obtained. Diameter of the aneurysm was determined by spiral CT 1 week after completion of the stent implantation procedure. The mean diameter at the middle of the stent-implanted aneurysm was 15.6 mm, with a range of 14.2–17.0 mm. The mean aneurysm length was 40 mm and the mean length of the infrarenal aorta was 9.4 mm. Aneurysm creation was immediately followed by endograft placement. A 10-mm ⫻ 5-cm Wallgraft endoprosthesis (Schneider, Minneapolis, Minn) with a 4-mm-diameter hole at
Volume 18
Number 6
Maffra et al
the middle of the graft was advanced over a guide wire to the aneurysm site. The Wallgraft was deployed so the hole was positioned toward the posterior aorta wall. Another angiogram was obtained to confirm the presence of endoleak through the hole in the Wallgraft. In all cases, endoleak was noted with filling of the lumbar arteries (Fig 2). Among the nine dogs included in the final cohort, we found three dogs with two pairs of lumbar arteries rising from the aneurysm sac. All catheters were then removed and the carotid artery was repaired with running 6 – 0 polypropylene suture. The neck incision was closed with 3– 0 polygalactin suture. All dogs received cefazolin 1 g three times a day and aspirin 325 mg three times a day for 72 hours and were housed at our animal research facility for the duration of the experiment. Heparin was not reversed at the end of the procedure.
765
Figure 3. Aortogram following EVOH treatment of the AAA sac. The large arrow indicates EVOH in the sac, which appears less radiodense than the angiographic contrast agent. The small arrow indicates EVOH in one of the lumbar arteries.
Endoleak Seal One week after AAA creation and implantation of the endograft, the dogs were anesthetized and CT scanning of the abdominal aorta was performed. Thereafter, each dog was taken to the angiography suite, where the left common carotid artery was exposed and catheterized. An angiogram was obtained to confirm persistence of the endoleak, noted in all surviving dogs. A 5-F RIM catheter (Infiniti angiography catheter; Cordis Endovascular, Warren, NJ) was advanced from the common carotid artery to the 4-mm hole in the Wallgraft where contrast agent injection allowed visualization of the persistent endoleak, the AAA sac, and the runoff lumbar arteries. Next, a 0.018-inch guide wire and a dimethyl sulfoxide– compatible microcatheter (Micro Therapeutics, Irvine, Calif) was passed through the 5-F catheter and across the hole in the Wallgraft. A priming volume of 0.35 mL dimethyl sulfoxide was injected into the microcatheter to prevent earlier solidification of the EVOH. Then, EVOH 8% was injected into the lumbar arteries and aneurysm sac under fluoroscopic guidance. We decided to stop the EVOH injection at the time that the lumbar arteries and the aneu-
•
Figure 2. Leaky Wallgraft within AAA model. Arrows point to lumbar arteries filling from the patent AAA sac.
rysm sac were completely filled. In our experiment, a mean volume of 0.7 mL of EVOH was injected (range, 0.4 – 1.0 mL). At the end of the procedure, an aortogram was obtained to confirm occlusion of the endoleak (Fig 3). The carotid arteriotomy and incision were sutured, and once again the dogs were awakened from anesthesia and allowed to recover. All surviving animals received cefazolin 1 g three times per day and aspirin 325 mg three times per day for 72 hours. CT Scanning All surviving dogs underwent CT imaging immediately before EVOH delivery (1 week after stent-implanted AAA creation and leaky Wallgraft placement) and then 1 week and 5 weeks after EVOH delivery. Scanning was performed on a Somatom Plus 4 scanner (Siemens, Erlangen, Germany) that obtained 3.0-mm-thick contiguous slices with and without intravenous contrast agent administration.
Images were reviewed and measured by an independent investigator who had no knowledge of the angiographic findings (W.D.). Because of unanticipated mortality in two of the nine dogs that survived to the EVOH procedure, final CT data at 5 weeks after embolization were available for seven dogs. CT data 1 week after EVOH were available for the dog that died of anesthetic complications, and CT data at 2 days after EVOH were available for the dog that developed paraplegia. Termination After CT at 5 weeks after EVOH delivery, each animal underwent final angiography and was euthanized. The retroperitoneum was inspected for any evidence of hematoma and the segment of the infrarenal aorta containing the aneurysm, from the renal arteries to the aortic trifurcation, was harvested and preserved in 10% formalin for histologic examination. Histopathologic Methods Gross pathologic and histologic examination of the treated AAA was performed by one author (L.H.B.) at an independent laboratory without information regarding angiographic or CT findings. Lumbar arteries were processed through and embedded in
766
•
June 2007
Type III Endoleak Seal with EVOH in a Canine Model
JVIR
Figure 4. Cross-section schematic of histologic analysis sites.
paraffin and then sectioned and stained with hematoxylin and eosin or Masson trichrome stains. Photomicrographs and gross photographs were prepared. Five sites for each explant were evaluated (Fig 4).
RESULTS We succeeded in creating stent-implanted AAAs with a patent Wallgraft endoleak in nine of our 12 dogs; the remaining three dogs died of aortic rupture related to AAA creation. Regarding the three deaths, one occurred at the end of the aneurysm/endoleak procedure, one the night of the procedure, and one 3 days after the procedure. Necropsy confirmed aortic rupture in all three animals. All nine dogs that survived AAA creation and Wallgraft placement had endoleaks into the sac and lumbar arteries, confirmed by angiography and CT imaging (Fig 2) 1 week later. After EVOH delivery, acute occlusion of the lumbar arteries and aneurysm sac was noted in all nine dogs by angiography. For the seven dogs that survived to the end of the experiment, CT and angiography confirmed occlusion of the AAA sac and lumbar arteries, even in the three dogs with two pairs of lumber arteries. There were three complications in the nine-dog cohort that survived AAA creation and Wallgraft placement. The first dog treated with EVOH manifested paraplegia on the first day after treatment as a result of an excess of EVOH infused into the distal lumbar arteries. The veterinary director of the animal research facility suggested observation for several
days. Therefore, the animal was observed for 3 additional days. After showing no improvement, this dog was euthanized after CT scanning and the aneurysm was resected on the fourth day after EVOH delivery. Another dog died 1 week after delivery of EVOH after undergoing a scheduled 1-week CT examination. The cause of death was believed related to complications of anesthesia. No anatomic abnormalities were noted at necropsy and the abdominal aortic aneurysm had not ruptured. A nonfatal complication occurred during successful delivery of EVOH when a small amount of EVOH refluxed through the 4-mm Wallgraft hole and embolized into the femoral artery. Reflux occurred because of overinjection of EVOH, and this complication was avoided in the remaining procedures. The EVOH embolus was easily seen by fluoroscopy and uneventfully removed with a sterile urologic stone retrieval basket. CT Scanning In all CT scans after EVOH delivery, the AAA sac was filled with radiodense EVOH, outflow lumbar arteries contained EVOH, and no endoleaks were seen (Fig 5). Histopathologic Results Gross examination of all specimens showed a smooth layer of neointima covering the luminal surface of the Wallgraft. In many cases the 4-mm hole at the middle of the Wallgraft was covered with tissue and could not be found. There was no evidence of rup-
Figure 5. CT scan shows EVOH in the posterior aspect of the AAA sac (large arrow) and in the lumbar arteries (small arrows).
ture or retroperitoneal hematoma in any of the specimens of the seven dogs that survived to the end of the experiment, the dog that died 1 week after EVOH, or the dog that was killed 4 days after EVOH. EVOH was easily identified in the AAA sac and the transected lumbar arteries by its dark gray appearance (Fig 6). Histologic analysis revealed that the space between the aneurysm wall and the Wallgraft was filled with EVOH mixed with thrombus of varying states of organization (Fig 7a,c). In some sections, inflammatory cells (typically macrophages) were found, mixed with thrombus. There were few changes in mural architecture of the aorta or the lumbar arteries. There was some degeneration of the smooth muscle of the aortic wall, in some cases replaced by fibrous connective tissue. Those findings were related to the stent placement and aneurysm creation. Lumbar arteries were filled with a plug of EVOH admixed with fibrous thrombus (Fig 7b). There was no evidence of intramural changes in the lumbar arteries. There were no vascular channels or evidence of persistent flow in any of the sections taken from the AAAs or lumbar arteries.
DISCUSSION Endoleaks are a significant technical problem associated with endoluminal repair of AAAs. In different reports, the incidences of endoleak
Volume 18
Number 6
Maffra et al
Figure 6. Bivalved en-bloc AAA explant at 5 weeks. EVOH is seen in this gross specimen as a homogenous gray filler in the wall of the AAA sac. EVOH runs nearly the entire length of the AAA (arrow).
have ranged between 8% and 44% (14,15,23). For treatment of endoleaks, many authors prefer to follow early endoleaks because they believe that a great majority of them will seal spontaneously. If spontaneous closure of the endoleak does not occur, it is often necessary to determine the type of endoleak. Type II endoleaks without AAA enlargement may be followed, whereas persistent type I and III endoleaks warrant treatment, as do type II endoleaks with enlarging AAA sacs. These types of endoleaks are considered unstable and pose risk for aneurysm rupture (24). Certainly, endografts are not designed and inserted to leak, and an endoleak represents failure of primary therapy directed at the AAA. Amongst those who believe persistent endoleaks warrant treatment, there is still debate as to the best type of intervention. Endoleaks may result from incomplete sealing at the end(s) of the endograft, delayed detachment at the terminal attachment sites, persistent or delayed retrograde collateral filling of the aneurysm sac, or delayed defects in the endograft itself (13,25). Transcatheter occlusion of endoleaks has been performed with coils, thrombin, and glue. The use of coils has been most widely reported (16 – 18). However, in one report, coils were not able to successfully reduce the intraaneurysmal sac pressure in a canine model (13) and the potential for aneurysm rupture was thought to persist after angiographic sac thrombosis. In the present study we tested the ability
of a novel embolic agent to seal a type III endoleak in a canine model. This agent, EVOH, changes its physical state on contact with aqueous solution (ie, blood) and becomes a plastic semisolid after it mixes with blood. We found that EVOH was capable of sealing leaks within the aneurysm sac and runoff lumbar collateral vessels, and maintained stable occlusion for 5 weeks with no evidence of recurrent endoleak. Histologic examination showed that EVOH mixed with thrombus and that the thrombus showed varying degrees of organization. We recognize that one of the limitations of our study is the short time span (5 weeks) for the observation of changes in the occluded sac and lumber arteries. In addition, our model is very “synthetic,” in that we created an aneurysm with a very small sac but a large endoleak. The endoleak, a 4-mm hole in the midportion of the Wallgraft, was artificially large and designed to sustain the endoleak until treatment with EVOH and permit easy catheterization of the AAA sac. For some endoleaks, such as those type II leaks that arise from collateral vessels without any attachment or graft defect leak, direct catheterization may be difficult. However, it is possible to employ a translumbar embolization technique as described by Baum et al (26,27). Another limitation of this work is the lack of pressure data from within the sac. Certainly, the concerns raised in previous studies (19 –21) regarding persistently increasd sac pressures af-
•
767
ter occlusive coil embolization could also apply to EVOH embolization if sac pressures were to remain increased. However, although our methodology did not allow for placement of sac pressure transducers, the ability of EVOH embolization to successfully reduce intraaneurysmal sac pressure has been described (28). This is most likely attributed to the ability of EVOH to form a durable and complete seal of the endoleak, in contrast to the use of coils, which may allow persistent perfusion amid the coils resulting in continued pressurization of the sac (28,29). Despite limitations caused by the study’s short time span, there are certain advantages to our model. These include quick and easy AAA creation, patency of endoleak flow into the AAA sac and lumbar arteries, straightforward explantation of the treated AAA, and most importantly the ability of the liquid embolic agent to completely seal the endoleak. However, these advantages are tempered by the loss of three dogs (ie, 25% mortality rate) during the early phase of the experiment when the AAA was created. In regard to the three animals that died of aneurysm rupture during AAA creation, the model of creating AAAs in dogs called for the use of oversized balloon-dilated stents. Gross inspection showed that the combined effects of overstretching resulting from the the balloon and the rigid ends of the stent caused aortic laceration. This is the singular drawback to this method of AAA creation. But although a 25% mortality rate related to AAA formation is substantial, other investigators have described the use of a similar technique of AAA creation with comparable mortality (1,22,30). Animals that survived AAA creation had stable AAAs with preserved lumbar arteries. Our final cohort of nine animals were ultimately treated for type III endoleaks, and we remain convinced that balloon stent overdilation is a reasonable method for the study of type III endoleaks in canine AAAs. Advantages of this method include simple creation of AAAs with use of percutaneous methods without the need for surgical techniques other than carotid artery cutdown, and also preservation of lumbar arteries. Regarding the delivery of EVOH, there was no difficulty in visualizing
768
•
June 2007
Type III Endoleak Seal with EVOH in a Canine Model
JVIR
Figure 7. (a) Cross-section micrograph (hematoxylin and eosin, magnification ⫻2.5) corresponding to location 2 in the histology schematic (Fig 4). EVOH appears black and lacelike, and it is admixed within thrombus in varying stages of organization. The transition between EVOH and the aneurysm sac wall (arrow) is free from inflammatory infiltrates. (b) Cross-section micrograph (hematoxylin and eosin, magnification ⫻10) corresponding to location 1 in the histology schematic (Fig 4). EVOH fills the lumbar artery lumen and is mixed with thrombus in varying stages of organization (arrow).
or infusing this agent. However, overly eager delivery probably contributed to the one case of paraplegia and one case of reflux embolization. We supposed that the case of paraplegia was a result of some kind of communication between the distal lumbar arteries and the medullary arteries in dogs. We believe careful and judicious use of any embolic or occlusive agent is warranted during occlusion of lumbar arteries. There is no embolic agent in clinical use today that carries no risk of inadvertent misplacement or unanticipated ischemia. However, EVOH can be delivered in a slower, more controlled manner that may reduce those risks compared with coil embolization or the use of other liquid embolic agents (31). Histologic evaluation demonstrated varying degrees of inflammation. There are many reasons for inflammatory cell infiltrates in the model AAA sac at 5 weeks. Some possibilities include the inflammatory nature of polyethylene terephthalate (ie, the Wallgraft), inflammation from aortic stretch injury after placement of the Palmaz stent, ongoing clearance and organization of thrombus within the sac, and the presence of EVOH itself. Nevertheless, macrophage infiltrates may not be “bad” and may actually represent the early phase in long-term occlusion of the endoleak, with thrombus organization and adherence to the
polyethylene terephthalate of the endograft. The observation of organized thrombus in many histologic sections is encouraging, but long-term studies are needed to declare this procedure a durable fix for endoleaks. Our model shows the feasibility of endoleak sealing with EVOH for type III endoleaks. The polymeric seal caused by EVOH was durable to 5 weeks without recurrence of endoleak by CT or angiography or at postmortem evaluation. Longer studies are needed to address the question of whether EVOH will ultimately prove successful in providing a durable seal of endoleak after endovascular AAA treatment in humans. References 1. Hallisey MJ. 1997 SCVIR Gary J. Becker Young Investigator Award: a transluminally created abdominal aortic aneurysm model. J Vasc Interv Radiol 1997; 8:305–312. 2. Balm R, Eikelboom BC, May J, Bell PR, Swedenborg J, Collin J. Early experience with transfemoral endovascular aneurysm management (TEAM) in the treatment of aortic aneurysms. Eur J Vasc Endovasc Surg 1996; 11:214 –220. 3. Beard JD. Screening for abdominal aortic aneurysm. Br J Surg 2003; 90: 515–516. 4. Ernst CB. Abdominal aortic aneurysm. N Engl J Med 1993; 328:1167– 1172. 5. Gorham TJ, Taylor J, Raptis S. Endovascular treatment of abdominal
6.
7.
8.
9.
10.
11.
12.
aortic aneurysm. Br J Surg 2004; 91: 815– 827. Prinssen M, Verhoeven EL, Buth J, et al. A randomized trial comparing conventional and endovascular repair of abdominal aortic aneurysms. N Engl J Med 2004; 351:1607–1618. Becker GJ, Kovacs M, Mathison MN. Transluminal repair of abdominal aortic aneurysm: a call for selective use, careful surveillance, new device design, and systematic study of transrenal fixation. J Vasc Surg 2002; 35:611– 615. Mortality results for randomised controlled trial of early elective surgery or ultrasonographic surveillance for small abdominal aortic aneurysms. The UK Small Aneurysm Trial Participants. Lancet 1998; 352:1649 –1655. Hausegger KA, Schedlbauer P, Deutschmann HA, Tiesenhausen K. Complications in endoluminal repair of abdominal aortic aneurysms. Eur J Radiol 2001; 39:22–33. Schlensak C, Doenst T, Hauer M, et al. Serious complications that require surgical interventions after endoluminal stent-graft placement for the treatment of infrarenal aortic aneurysms. J Vasc Surg 2001; 34:198 –203. Greenhalgh RM, Brown LC, Kwong GP, Powell JT, Thompson SG. Comparison of endovascular aneurysm repair with open repair in patients with abdominal aortic aneurysm (EVAR trial 1), 30-day operative mortality results: randomised controlled trial. Lancet. 2004; 364:843– 848. Maher MM, McNamara AM, MacEneaney PM, Sheehan SJ, Malone DE. Abdominal aortic aneurysms: elective
Volume 18
13. 14.
15.
16.
17.
18.
19.
Number 6
endovascular repair versus conventional surgery—evaluation with evidence-based medicine techniques. Radiology. 2003; 228:647– 658. Golzarian J, Maes EB, Sun S. Endoleak: treatment options. Tech Vasc Interv Radiol 2005; 8:41– 49. Baum RA, Stavropoulos SW, Fairman RM, Carpenter JP. Endoleaks after endovascular repair of abdominal aortic aneurysms. J Vasc Interv Radiol 2003; 14:1111–1117. Wain RA, Marin ML, Ohki T, et al. Endoleaks after endovascular graft treatment of aortic aneurysms: classification, risk factors, and outcome. J Vasc Surg 1998; 27:69 –78; Choke E, Thompson M. Endoleak after endovascular aneurysm repair: current concepts. J Cardiovas Surg 2004; 45:349 –366. Gorich J, Rilinger N, Sokiranski R, et al. Treatment of leaks after endovascular repair of aortic aneurysms. Radiology 2000; 215:414 – 420. Baum RA, Carpenter JP, Tuite CM, et al. Diagnosis and treatment of inferior mesenteric arterial endoleaks after endovascular repair of abdominal aortic aneurysms. Radiology 2000; 215: 409 – 413. Solis MM, Ayerdi J, Babcock GA, et al. Mechanism of failure in the treatment of type II endoleak with percutaneous
Maffra et al
20.
21.
22.
23.
24.
25.
coil embolization. J Vasc Surg 2002; 36:485– 491. Marty B, Sanchez LA, Ohki T, et al. Endoleak after endovascular graft repair of experimental aortic aneurysms: does coil embolization with angiographic “seal” lower intraaneurysmal pressure? J Vasc Surg 1998; 27:454 – 461. Sheehan MK, Barbato J, Compton CN, Zajko A, Rhee R, Makaroun MS. Effectiveness of coiling in the treatment of endoleaks after endovascular repair. J Vasc Surg 2004; 40:430 – 434. Pavcnik D, Andrews RT, Yin Q, et al. A canine model for studying endoleak after endovascular aneurysm repair. J Vasc Interv Radiol 2003; 14:1303–1310. White GH, Yu W, May J, Chaufour X, Stephen MS. Endoleak as a complication of endoluminal grafting of abdominal aortic aneurysms: classification, incidence, diagnosis, and management. J Endovasc Surg 1997; 4:152–168. Faries PL, Cadot H, Agarwal G, Kent KC, Hollier LH, Marin ML. Management of endoleak after endovascular aneurysm repair: cuffs, coils, and conversion. J Vasc Surg 2003; 37: 1155–1161. White GH, May J, Waugh RC, Chaufour X, Yu W. Type III and type IV endoleak: toward a complete definition of blood flow in the sac after endolu-
26.
27.
28.
29.
30.
31.
•
769
minal AAA repair. J Endovasc Surg 1998; 5:305–309. Baum RA, Carpenter JP, Golden MA, et al. Treatment of type 2 endoleaks after endovascular repair of abdominal aortic aneurysms: comparison of transarterial and translumbar techniques. J Vasc Surg 2002; 35:23–29. Baum RA, Cope C, Fairman RM, Carpenter JP. Translumbar embolization of type 2 endoleaks after endovascular repair of abdominal aortic aneurysms. J Vasc Interv Radiol 2001; 12: 111–116. Martin ML, Dolmatch BL, Fry PD, Machan LS. Treatment of type II endoleaks with EVOH. J Vasc Interv Radiol 2001; 12:629 – 632. Abdala N, Levitin A, Dawson A, et al. Use of ethylene vinyl alcohol copolymer for tubal sterilization by selective catheterization in rabbits. J Vasc Interv Radiol 2001; 12:979 –984. Skillern CS, Stevens SL, Piercy KT, et al. Endotension in an experimental aneurysm model. J Vasc Surg 2002; 36: 814 – 817. Cantasdemir M, Kantarci F, Mihmanli I, Numan F. Embolization of profunda femoris artery branch pseudoaneurysms with ethylene vinyl alcohol copolymer (EVOH). J Vasc Interv Radiol 2002; 13:725–728.