The Zenith AAA Endovascular Graft for Abdominal Aortic
`Aneurysms: Clinical Update
`
`By Roy Greenberg on Behalf of the Zenith Investigators
`
`Purpose: The aim of this study was to provide an update regarding the US clinical trial assessing the
`performance of the Zenith AAA Endovascular Graft in the treatment of abdominal aortic aneurysms.
`Materials and Methods: A prospective, nonrandomized, concurrent control-based study design was
`used to contrast conventional repair of infrarenal aneurysms with endovascular repair in patients that
`would otherwise be candidates for open surgical procedures. Additional study arms allow high-
`physiologic-risk patients and roll-in patients to be treated using the same endovascular device in a
`registry format. Patients were evaluated clinically and radiographically computed tomography (CT)
`and abdominal radiographs) before any intervention, at hospital discharge, 30 days, 6 months, 12
`months, and yearly thereafter. Data were analyzed with the intent of assessing acute and chronic
`morbidity and mortality, radiographic parameters indicative of successful aneurysm repair, and
`device integrity.
`Results: A total of 352 patients were treated with the Zenith graft (one patient did not receive an
`implant in the standard-risk group), and 80 patients underwent conventional surgical repair. Two
`hundred patients were enrolled in the standard-risk group, 100 in the high-risk group, and 52
`underwent endovascular repair as roll-in patients. All cause mortality, aneurysm-related deaths, and
`ruptures were statistically identical between the groups. Procedural morbidity was significantly lower
`for patients treated with endovascular grafts with respect to cardiac, pulmonary, renal, and vascular
`complications. Secondary interventions were more commonly required in the endovascular group.
`The endoleak rate was 4.9% at 12 months in the standard-risk endovascular group. There was one
`rupture in the high-risk subset of patients and 3 elective conversions. A total of 1.6% of the
`endovascular patients were noted to have a barb separation without evidence of significant migration
`or clinical events. No other device integrity issues were observed.
`Conclusions: The safety of the Zenith endovascular graft was superior to conventional management
`with respect to morbidity and clinical utility. The short-term device efficacy was satisfactory; how-
`ever, longer-term follow-up will be necessary to establish the duration of this observation.
`© 2003 Elsevier Inc. All rights reserved.
`
`THE ZENITH ENDOGRAFT has evolved
`
`from a worldwide collaborative effort that
`began in 1991. Observations of early complica-
`tions1-5 associated with endovascular grafting have
`resulted in a series of modifications leading to the
`current Zenith AAA Endovascular Graft and
`H&L-B One Shot Introduction System (Blooming-
`ton, IN). Initial studies in animals and humans1,2,4
`utilized prostheses fashioned with Gianturco Z-
`stents and dacron fabric. The implantable portion
`of this device has progressed from an unsupported
`monoiliac device to a fully supported modular
`bifurcated system. The current delivery system has
`undergone developmental changes that parallel
`prosthesis evolution. Sophisticated deployment
`techniques allow for extremely accurate position-
`ing of the device and the controlled release of a
`series of suprarenal barbs designed to mimic a
`surgical anastomosis.5,6 The endovascular system
`has incorporated a number of modifications to
`avert the failure modes detected in early implants
`
`as well as other devices. However, additional iter-
`ations of the device are expected owing to the
`careful observations of potential areas of improve-
`ment. A United States Food and Drug Administra-
`tion pivotal study was undertaken beginning in
`January 2000, with the intention of evaluating the
`safety and effectiveness of the Zenith endograft.
`
`MATERIAL AND METHODS
`Device Description
`The Zenith AAA Endovascular Graft is a bifur-
`cated, modular system capable of multiple endo-
`
`From the Department of Vascular Surgery, The Cleveland
`Clinic Foundation, Cleveland, OH.
`Address reprint requests to, Roy K. Greenberg, MD, Director
`of Endovascular Research, Department of Vascular Surgery,
`The Cleveland Clinic Foundation, Desk S-41, 9500 Euclid Ave,
`Cleveland, OH 44195.
`© 2003 Elsevier Inc. All rights reserved.
`0895-7967/03/1602-0008$30.00/0
`doi:10.1016/S0895-7967(03)00009-7
`
`Seminars in Vascular Surgery, Vol 16, No 2 (June), 2003: pp 151-157
`
`151
`
`TMT 2108
`Medtronic v. TMT
`IPR2021-01532
`
`

`

`152
`
`ROY GREENBERG
`
`Fig 1. The 3-piece Zenith system. The components are constructed from stainless steel stents and polyester fabric. The main
`body consists of an uncovered suprarenal stent with barbs followed by an internal sealing stent. The contralateral limb is designed
`to reside 15 mm above the aortic bifurcation. Each of the common iliac limbs can be tromboned (by increasing the overlap region),
`to a certain extent, into the main body to give the interventionalist a range of iliac coverage for a given length of the device. The
`customized 2-piece version and aortomonoiliac designs were used infrequently in the national trial.
`
`vascular graft configurations and is preloaded onto
`the H&L-B One-Shot Introduction System. The
`main endovascular graft (Fig 1) is a 3 component
`device (aortic main body and two iliac legs), the
`sizes of which are selected depending on patient
`anatomy. The main graft and ancillary components
`consist of woven Dacron and are fully stented with
`self-expanding stainless steel Z-stents. An uncov-
`ered stent with staggered barbs at the top of the
`graft is used for suprarenal attachment. Ancillary
`components (main body extenders, iliac leg ex-
`tenders, converters, and occluders) are available to
`provide additional length to their respective por-
`tions of the graft or to convert a bifurcated graft
`into an aorto-uni-iliac graft, if necessary.
`
`Study Design
`
`A total of 352 patients were enrolled prospec-
`tively at 15 centers within the United States. Three
`endovascular arms were established with the intent
`of comparing physiologically similar endovascular
`and surgical control patients. Patients that were felt
`to be appropriate candidates for traditional open
`repair or endovascular procedures were considered
`standard-risk patients (SRG). Individuals that were
`felt to be at a higher physiologic-risk, potentially
`unable to tolerate conventional treatments, were
`placed in a high-physiologic-risk arm (HRG). Fi-
`nally, each center was allotted a number of patients
`they were allowed to treat before the accumulation
`of data within the pivotal study, arms with the goal
`of establishing device and procedural familiarity.
`The anatomic inclusion criteria were identical for
`
`all patients treated with an endovascular approach
`and are depicted in Fig 1. A total of 80 concurrent
`controls (CG) were enrolled with the intent of
`contrasting the morbidity and mortality to the SR
`endovascular group. Although the medical inclu-
`sion criteria were identical for the CG and SRG,
`the CG patients need not have met the anatomic
`inclusion criteria for endovascular grafting. Demo-
`graphic information, comorbid conditions, aortic
`morphology, implant information, complications,
`and follow-up information were collected. An as-
`signed core laboratory (The Cleveland Clinic
`Foundation, Cleveland, OH) independently ana-
`lyzed radiographic information, while a clinical
`events committee (Harvard Clinical Research In-
`stitute, Boston, MA) adjudicated clinical events for
`purposes of interpretation. A data safety monitor-
`ing board was established and reviewed the data
`intermittently throughout the course of the trial. All
`participating centers were required to have institu-
`tional review board approval for the study and
`informed consent from each patient.
`
`Preprocedural Information
`
`All patients were required to undergo high-
`resolution helical CT evaluation of their abdomen
`and pelvis. The observed expansion of the self-
`expanding stent required arterial diameter calcula-
`tions to be observed from outer wall to outer wall.
`Length measurements were obtained from the axial
`images in nontortuous anatomy, 3-dimensional re-
`constructions, or angiography, which was not re-
`quired before placement of the endoprosthesis.
`
`

`

`DEVICE SIZING
`
`153
`
`Device diameters were intended to be sized 10 to
`15% larger than the measured arterial diameter.
`The length measurements were used to calculate a
`main body length that was designed to have the
`proximal aspect of the graft material immediately
`below the lowest renal artery and place the con-
`tralateral limb 15 mm cranial to the contralateral
`iliac orifice, allowing the ipsilateral limb to extend
`15 mm into the ipsilateral common iliac artery. The
`iliac extension lengths were calculated to allow
`coverage of as much of the common iliac artery as
`possible (Fig 1). The overall intention was to line
`as much of the arterial region from the renal
`arteries to the hypogastric arteries with graft
`material. Bilateral hypogastric preservation was
`desired, but unilateral hypogastric patency was
`mandatory. Although customized designs were
`available and intended for use in tortuous anato-
`mies or situations in which an aortomonoiliac graft
`was felt to be superior, they were infrequently
`utilized in the trial.
`
`Procedural Techniques
`Devices were inserted through either femoral
`artery. A contralateral angiographic catheter was
`utilized to intermittently show the location of the
`lowest renal artery, allowing the graft portion of
`the main body to be deployed immediately below
`the renal artery. Before deployment of the uncov-
`ered suprarenal Z-stent, the contralateral limb was
`cannulated. An angiographic injection within the
`partially deployed main body allowed visualization
`of the renal arteries and final device position ad-
`justments. The cap constraining the barbed supra-
`renal stent then was advanced allowing the barbs to
`engage the aortic wall. A stiff wire was advanced
`through the contralateral limb, and the contralateral
`endoprosthesis was inserted and deployed. The
`remainder of the main body was deployed, the top
`cap was retrieved, and the ipsilateral limb was
`inserted and deployed. The sheath design allowed
`for direct arterial puncture and instrumentation of
`the arterial system without placement of additional
`sheaths. A compliant balloon was inflated at all of
`the modular joints and the arterial sealing sections
`to ensure graft apposition.
`
`Follow-up
`All patients underwent a predischarge, 30-day,
`6-month, and 12-month clinical evaluation; CT
`scan; and 4-view abdominal radiography. Annual
`
`Table 1. Zenith US Multicenter Phase II Trial Study Arms
`
`Standard-Risk
`Group
`
`Surgical Control
`Group*
`
`200*
`
`80†
`
`High-Risk
`Group†
`
`100‡
`
`Roll-in
`Group‡
`
`52§
`
`*One standard-risk patient did not receive the Zenith device.
`†Surgical patients may have had anatomy unsuitable for
`endovascular repair.
`‡High-risk patients had high medical risk and anatomy suit-
`able for endovascular repair.
`§Initial Zenith patients at sites without prior device experi-
`ence.
`
`examinations were performed thereafter. Clinical
`evaluation included a physical examination, an
`assessment of the lower extremity blood flow, and
`measurement of the serum blood urea nitrogen and
`creatinine. CT scans were assessed to determine
`maximal aneurysm size, the patency of the renal
`arteries, presence of renal infarcts, migration of the
`endoprosthesis, patency of the hypogastric arteries,
`as well as the presence and attempted classification
`of an endoleak. Standard definitions for endoleaks
`were utilized.7,8 Aneurysmal shrinkage or growth
`was considered significant when a change greater
`than 5 mm was observed. Abdominal radiographs
`were assessed for device integrity and continuity of
`the overlap region of modular components.
`
`RESULTS
`A total of 351 patients underwent placement of
`a Zenith endovascular prosthesis from January
`2000 through July of 2001 (one patient did not
`receive the Zenith device. Table 1). The mean
`patient age was 71 years, and approximately 93%
`were men. There were no significant differences
`across the investigative sites allowing data pooling
`to occur. Aneurysms were confined to the infrare-
`nal abdominal aorta in 80% of
`the patients,
`whereas 20% of the patients had aneurysms ex-
`tending from the aorta into one of the common
`iliac arteries. Only one patient treated with this
`device had an isolated iliac artery aneurysm. The
`mean aneurysm size for the SRG, CG, HRG, and
`roll-in group was 56.2 ⫾ 8.8 mm, 63.8 ⫾ 11.3 mm.
`57.5 ⫾ 9.2 mm, and 58.1 ⫾ 6.6 mm, respectively.
`Follow-up rates were more than 90% for SRG and
`more than 80% for SG at predischarge, 30 days, 6
`months, and 12 months (Table 2).
`
`Comparison of SRG and CG
`In the analysis of the SRG and CG, the only
`significant differences in comorbid conditions be-
`
`

`

`154
`
`ROY GREENBERG
`
`Table 2. The Percentage of Patients With Completed
`Clinical and Radiographic Follow-Up at 12 Months
`
`Follow-up
`
`Standard Risk
`
`Surgical
`
`High Risk
`
`Roll-in
`
`Clinical
`Imaging
`
`91%
`90%
`
`81%
`81%
`
`85%
`81%
`
`80%
`77%
`
`Clinical follow-up refers to clinical examination and labora-
`tory studies. Radiographic follow-up includes abdominal and
`pelvic CT analysis and flat plate abdominal radiography.
`
`tween the groups were an increased age in the SRG
`(71 v 69; P ⫽ .03) and the presence of hyperten-
`sion (64% v 83%; P ⬍ .001). Anatomic differences
`inherent to the study design were present. Surgical
`patients had shorter proximal necks (P ⬍ .001),
`with a larger proximal neck diameter (P ⬍ .001),
`more tortuousity (P ⬍ .01) and occlusive disease
`(P ⬍ .01) of the iliac anatomy, larger aneurysms
`(P ⬍ .001), and differences in aneurysm type
`(more iliac involvement P ⬍ .001). The proximal
`clamp was placed in an infrarenal position in 81%
`of the surgical patients, and suprarenal (including
`supraceliac) in 18% of the patients. All of the
`surgical procedures were done under general anes-
`thesia, whereas 53% of the endovascular grafts
`were placed using epidural or 2% using local
`anesthesia. The median procedure duration was
`140 minutes for the SRG and 210 minutes (P ⬍
`.001) for the CG. Estimated blood loss was greater
`for the CG compared with the SRG (1,676 mL and
`299 mL respectively, P ⬍ .001). The median
`fluoroscopy time was 25 minutes.
`
`Mortality Rate
`The acute mortality, defined as that occurring
`within 30 days of the procedure, was 0.5% in the
`SRG (1 of 199), and 2.5% in the CG (2 of 80). All
`cause mortality (through 12 months) was 3.5% in
`SRG, and 3.8% in the CG. AAA-related mortality,
`defined as any mortality within 30 days of the
`primary procedure, or a secondary procedure ad-
`dressing an aneurysmal issue as well as deaths
`related to the aneurysm (⬎30 days postprocedure),
`was 0.5% in SRG and 1.3% for the CG. Using a
`log-rank test (Tarone-Ware), the difference in an-
`eurysm related survival was significant between
`the 2 groups (P ⫽ .04). Kaplan Meier analysis of
`all cause mortality and aneurysm-related mortality
`showed no differences between the 2 groups.
`
`cified femoral artery precluding a single patient in
`the SRG from receiving a device; however, the
`patient did not undergo a surgical conversion.
`There were 3 late conversions during the first 12
`months. The first was performed electively for a
`persistent proximal endoleak that was unsolvable
`with endovascular means. The second was for the
`development of a supraceliac aneurysm that ex-
`panded rapidly in a patient with systemic lupus
`erythematosis. A thoracoabdominal repair was car-
`ried out with resection of the proximal portion of
`the prosthesis and a sutured anastomosis of the
`surgically implanted graft to the distal portion of
`the main body of the endograft. There was one
`conversion in the HRG, at 222 days postprocedure,
`as a result of rupture. This patient from the HRG
`represents the single rupture in the pivotal trial.
`The patient underwent successful surgical repair.
`At the time of the procedure, the right iliac limb of
`the endograft was noted to be within the aneurysm
`sac. The aneurysm had contracted from 6 cm to 4.6
`cm over the 6 months after implantation. Despite
`the security of the proximal fixation, the iliac limb
`had presumably retracted into the aneurysm, re-
`pressurizing the sac and causing the rupture.
`
`Endoleaks and Secondary Interventions
`The acute (30-day) endoleak rate was 17%. The
`majority of
`these were type II
`leaks (9.5%),
`whereas 4.5% were classified as type I endoleaks.
`The study design required the treatment of all type
`I and III endoleaks whenever possible. Type II
`endoleaks were observed and treated in the setting
`of aneurysmal growth or based on physician pref-
`erence. The endoleak rate decreased to 7.4% at 12
`months. Only one type I leak was ascribed to
`failure of a distal iliac seal. There were consider-
`ably more secondary interventions in the SRG
`compared with the CG (11% and 2.5%; P ⫽ .03).
`The treatment of endoleaks constituted the major-
`ity of the secondary interventions (6%) for the
`SRG. The remainder of the secondary interven-
`tions were for iliac limb or native arterial stenoses
`or occlusion, renal issues, and other miscellaneous
`interventions. The 2 secondary interventions in the
`CG were acute and performed in an effort
`to
`control intraabdominal hemorrhage.
`
`Conversions and Rupture
`There were no acute conversions in the study.
`Technical difficulties were encountered with a cal-
`
`Morbidity
`Significant decreased morbidity was noted in the
`SRG at 30 days for cardiac (P ⫽ .02), pulmonary
`
`

`

`DEVICE SIZING
`
`155
`
`(P ⬍ .001), renal (P ⫽ .01) and vascular (P ⬍
`.001) organ systems. Additionally, diminished
`blood loss, fewer transfusion requirements, shorter
`hospital stay, decreased intensive care unit time,
`and faster return to daily activities were associated
`with the endovascular procedure. There was no
`evidence of deterioration of renal function over the
`course of the follow-up. The observed renal in-
`farcts (3 in SRG, one in HRG, and 2 CG, 1 RI)
`were attributed to the coverage or ligation of ac-
`cessory renal arteries and occurred after device
`implantation or surgical repair.
`
`Aneurysm Size
`Aneurysm growth was noted in 3 patients (2
`SRG, 1 HRG; 1.5%). Two were associated with
`graft infections and were converted electively. The
`third was attributed to a distal endoleak. A signif-
`icant decrease in aneurysm size (⬎5 mm) was
`present in 68% of the patients at 12 months. The
`remainder of the aneurysms remained stable in
`size.
`
`DISCUSSION
`Careful consideration has gone into the design
`of both the endograft and the delivery device.
`Much of the evolution of this device was influ-
`enced by early experiences. For example, after
`the implantation of non–fully supported Z-stent
`endografts into a group of patients, the Malmo
`group detected a proximal stent migration rate of
`approximately 40%.1 Early migrations were at-
`tributed to a number of factors including fixation
`mechanisms and device placement in close prox-
`imity to the renal arteries.3 It was noted also that
`devices placed low in the proximal neck had a
`higher incidence of migration and subsequent
`proximal endoleak rate and subsequent rupture.
`Patients treated for aneurysms with open surgi-
`cal and endovascular approaches have been
`shown to undergo proximal neck dilation over
`time.9-11 Potential explanations relate to the in-
`stability of the infrarenal aorta in the setting of
`an infrarenal aneurysm. This raised great con-
`cern among the investigative groups and resulted
`in the development of suprarenal aortic fixation
`and a delivery system that allows for extremely
`precise positioning of the endograft at the level
`of the lowest renal artery—similar to the surgi-
`cal practice of suturing an anastomosis as close
`to the renal arteries as possible. However, this
`
`requires one to cross the renal arteries with a
`bare stent, a process that has elicited a great deal
`of debate. A series of animal studies,12 in addi-
`tion to the evaluations of many clinical implants,
`has allowed advocates of suprarenal stenting to
`proclaim this practice as safe.1,5,13 Other devices
`have utilized similar means of extension into the
`suprarenal aorta.14,15
`The clinical utility of the device was improved
`over the CG. Significant decreases in blood loss;
`blood transfusion; pulmonary, cardiac, and bowel-
`related complications; reduced hospital stay; and
`minimized intensive care unit days mimic the re-
`sults of other endograft studies.16,17 Specific issues
`with respect to the presence of a suprarenal stent
`deserve additional discussion. No deleterious ef-
`fect of the suprarenal stent was detected by mea-
`suring blood urea nitrogen levels and creatinine
`concentration. Radiographic analysis of the status
`of the renal artery and prevalence of renal infarcts
`support this conclusion and concur with prior re-
`ports.13,18 However, the incidence of secondary
`procedures in the SRG was significantly higher
`than in the CG. The majority of these were under-
`taken to treat endoleaks. Although the indication
`for a secondary procedure is subjective in many
`cases, both the incidence and magnitude of the
`secondary intervention must be considered when
`assessing the results. The 3 conversions that were
`performed ultimately constitute failure of endovas-
`cular treatment. However, the remainder of the
`secondary interventions were percutaneous in na-
`ture and must be viewed in the context of alterna-
`tive therapeutic modalities. Secondary procedures
`pertaining to renal artery issues have the potential
`to be more complex with this device. A brachial
`approach in many cases may provide one with a
`geometric advantage with respect to renal cannu-
`lation.
`Patient selection, in the setting of a suprarenal
`stent, must be addressed. The angulation of the
`suprarenal aorta is viewed in context with angu-
`lation of the proximal neck and the aneurysm
`sac. Opposing angulations can result in the su-
`prarenal segment displacing the sealing stent in
`the infrarenal aorta resulting in a proximal leak,
`greatly complicating the initial procedure. For
`this reason, we do not advocate the use of this
`graft in the setting of opposing supra-and infra-
`renal angulations. Aortic penetration by barbs
`has been the focus of some concern. The poten-
`
`

`

`156
`
`ROY GREENBERG
`
`tial for surrounding organ damage may exist but
`has not been recognized clinically to date. De-
`vice repositioning after barb deployment can
`represent a challenge. Although a number of
`investigators have pulled a bifurcated device
`distally with an up-and-over wire/catheter sys-
`tem, we prefer to place renal stents in the setting
`of a high deployment rather than moving the
`main body.
`The Zenith device is designed to accommo-
`date larger necks and iliac arteries than the 3
`commercially available devices. Proximal necks
`larger than 28 mm were felt to be potentially
`unstable9 whereas the current Z-stent design
`may provide an inadequate degree of radial force
`in an overly large infrarenal neck. The addition
`of barbs to the suprarenal stent was undertaken
`to diminish the incidence of migration. Fixation
`of the prosthesis to the delivery device itself
`combined with a separate mechanism utilized to
`deploy the barbs allows for movement of the
`endograft after deployment of the graft portion
`of the prosthesis and extremely accurate posi-
`tioning. The distal fixation size was intended to
`allow for maximal preservation of the hypogas-
`tric circulation. The main body of the device is
`intentionally long and optimally designed to
`place the ostium of the contralateral limb 15 mm
`above the aortic bifurcation. This will place the
`ipsilateral limb within the proximal aspect of the
`ipsilateral iliac artery and is intended to diminish
`the risk of component separation, limit the mi-
`gration effect of blood hitting the bifurcation,
`and facilitate limb cannulation.
`Conversion to open repair in the setting of
`suprarenal fixation with barbs has been per-
`formed in 3 patients in this series. Two cases
`involved supraceliac clamping to avoid damag-
`ing the aortic wall by crushing the stent material.
`After an aortotomy, an umbilical tape was tight-
`ened at the infrarenal component of the suprare-
`nal stent in one case, and the stent was pulled out
`without much difficulty. This technique forces
`the barbs to bend in an upward direction. Per-
`haps a more comfortable solution was exercised
`in 2 other cases in which the graft material was
`transected at the level of the suprarenal stent or
`below and removed. The proximal anastomosis
`then was constructed in such a manner to incor-
`porate the suprarenal stent. Although one can
`argue that the presence of barbs may complicate
`
`explant procedures, to design an implant so that
`it is easily removable would appear to make it
`more prone to migration during the lifetime of
`the patient. The displacement force of the en-
`dograft was tested in a cadaver model with a
`variety of hooks and barb configurations.19 The
`findings of this study support the use of large and
`numerous barbs. This observation coincided
`with the clinical findings in 3 patients who
`suffered from endograft migration when a sys-
`tem with only 4 weak barbs was used. The
`subsequent addition of more barbs, bringing the
`total to 10, increases the required displacement
`force in the cadaver model.20 The incidence of
`barb separation was 2.5% at 12 months. This
`issue is likely to be addressed in subsequent
`design modifications. No other device integrity
`issues were identified in the SRG during the
`course of the trial.
`In contrast to unsupported grafts22 a relatively
`low incidence of limb thrombosis was noted. The
`exclusion of patients with severe iliac tortuosity,
`the use of fully supported limbs, and the selective
`use of additional stents for kinked limbs provide us
`with potential explanations.
`The endoleak incidence was extremely low in
`contrast to other reported series. It is difficult to
`explain why there would be variable rates of type
`II leaks between graft designs unless there are a
`significant number of leaks that are incorrectly
`identified as type II. However, the 30-day leak rate
`of 17% was largely composed of type II leaks that
`spontaneously thrombosed. Investigators were en-
`couraged to treat all type I and III endoleaks,
`leaving us with a 7.4% incidence of endoleak at 12
`months.
`Aneurysm size changes were dramatic with this
`device. A fairly steep rate of size change was
`noted. Nearly 70% of the patients in the SRG
`experienced a minimum of 5 mm of diameter
`reduction within a 12-month period. Only 3 pa-
`tients suffered from aneurysm enlargement, 2 in
`the setting of infected grafts and one as a result of
`an untreated type I distal endoleak. The inherent
`reassurance of decreasing aneurysm size must be
`balanced with a careful assessment of follow-up
`radiographs to detect potential component migra-
`tion.
`The placement of the Zenith endovascular graft
`can be accomplished with minimal morbidity and
`mortality. The patients have relatively short hospi-
`
`

`

`DEVICE SIZING
`
`157
`
`tal stays, minimal blood loss, and return to normal
`function quickly. Aneurysm size appears to de-
`crease in the majority of cases during the follow-up
`period, and rupture is extremely rare; however, we
`have not eliminated the potential for late en-
`doleaks. Graft failure and fatigue are possible and
`have been documented in a few cases; however, we
`feel we have minimized this risk by utilizing time-
`tested materials in a manner that closely resembles
`our practices of open surgical repair.
`
`ACKNOWLEDGEMENTS
`
`The authors acknowledge the Zenith Investigators: Timothy
`A.M. Chuter, Daniel Clair, Sunita Srivastava, Stephan Lalka,
`W. Charles Sternbergh, III, Barry S. George, Albert Hakaim,
`Ronald Fairman, Linda M. Reilly, Frank J. Criado, Richard M.
`Green, David J. Porter, Frank J. Veith, Richard P. Cambria,
`Frederick S. Keller, Frnak J. Miller, Mark Sarfati, Luis
`Sanchez; the Cleveland Clinic Angiographic Core Laboratory
`Study PI, Sean Lyden, MD; the Clinical Events Committee
`Chief Scientific Officer, Richard E. Kuntz, MD; and DSMB
`Chairman, Willis Tacker, MD, PhD.
`
`REFERENCES
`
`1. Ivancev K, Malina M, Lindblad B, et al: Abdominal aortic
`aneurysms: Experience with the Ivancev-Malmo endovascular
`system for aortomonoiliac stentgrafts. J Endovasc Sur 4:242-
`251, 1997
`2. Chuter TA, Wendt G, Hopkinson BR, et al: Bifurcated
`stent-graft for abdominal aortic aneurysm. Cardiovasc Surg
`5:388-392, 1997
`3. Ivancev K, Lindblad B, Resch T, et al: Factors contribut-
`ing to migration of stent-grafts in true AAA. J Vasc Interv
`Radiol 9:175, 1998 (Suppl)
`4. Chuter TA, Wendt G, Hopkinson BR, et al: Transfemoral
`insertion of a bifurcated endovascular graft for aortic aneurysm
`repair: The first 22 patients. Cardiovasc Surg 3:121-128, 1995
`5. Lawrence-Brown M, Hartley D, MacSweeny S, et al: The
`Perth endoluminal bifurcated graft system—Development and
`early experience. Cardiovasc Surg 4:706-712, 1996
`6. van Schie G, Sieunarine K, Lawrence-Brown M, et al:
`The Perth bifurcated endovascular graft for infrarenal aortic
`aneurysms. Semin Interv Radiol 15:63-69, 1998
`7. White GH, Yu W, May J, et al: Endoleak as a complica-
`tion of endoluminal grafting of abdominal aortic aneurysms:
`Classification, incidence, diagnosis, and management. J Endo-
`vasc Surg 4:152-168, 1997
`8. White G, May J, Waugh R, et al: Type III and Type IV
`endoleak: Toward a complete definition of blood flow in the sac
`after endoluminal AAA repair. J Endovasc Surg 5:305-309,
`1998
`9. Illig K, Green R, Ouriel K, et al: Fate of the proximal
`aortic cuff: Implications for endovascular aneurysm repair. J
`Vasc Surg 26:492-499, 1997
`10. Sonnesson B, Resch T, Lanne T, et al: The fate of the
`infrarenal aortic neck after open aneurysm surgery. J Vasc Surg
`28:889-894, 1998
`11. Sonnesson B, Hansen F, Lanne T: Compliance and
`diameter in the human abdominal aorta—The influence of age
`and sex. Eur J Vasc Endovasc Surg 7:690-697, 1993
`12. Malina M, Lindh M, Ivancev K, et al: The effect of
`
`endovascular aortic stents placed across the renal arteries. Eur J
`Vasc Endovasc Surg 13:207-213, 1997
`13. Malina M, Brunkwall J, Ivancev K, et al: Renal arteries
`covered by aortic stents: Clinical experience from endovascular
`grafting of aortic aneurysms. Eur J Vasc Endovasc Surg 14:
`109-113, 1997
`14. Marin ML, Veith FJ, Cynamon J, et al: Initial experience
`with transluminally placed endovascular grafts for the treatment
`of complex vascular lesions. Ann Surg 222:449-465, 1995
`15. Criado F, Fry P: The Talent endolumnial stent-graft
`system: Technical success and acute clinical outcome: Report
`of a collective international experience. J Endovasc Surg 5:1-9,
`1998
`16. Zarins C, White R, Schwarten D, et al: AneuRx stent
`graft versus open surgical repair of abdominal aortic aneu-
`rysms: Multicenter prospective clinical trial. J Vasc Surg 29:
`292-308, 1999
`17. Franco T, Zajko A, Federle M, et al: Endovascular repair
`of the abdominal aortic aneurysm with the ancure endograft: CT
`follow-up of perigraft flow and aneurysm size at 6 months. J
`Vasc Intervent Radiol 11:429-435, 2000
`18. Endovascular repair of abdominal aortic aneurysms with
`an uncovered suprarenal stent does not impair renal function. 98
`A.D. Dec 13, 1998
`19. Resch T, Malina M, Lindblad B, et al: The impact of
`stent design on proximal stent-graft fixation in the abdominal
`aorta: An experimental study. Eur J Vasc Endovasc Surg
`20:190-195, 2000
`20. Resch T, Ivancev K, Brunkwall J, et al: Distal migration
`of stent-grafts after endovascular repair of abdominal aortic
`aneurysms. J Vasc Intervent Radiol 10:257-264, 1999
`21. Malina M, Lindblad B, Ivancev K, et al: Endovascular
`AAA exclusion: Will stents with hooks and barbs prevent
`stent-graft migration? J Endovasc Surg 5:310-317, 1998
`22. Moore WS, Rutherford R: Transfemoral endovascular
`repair of abdominal aortic aneurysm: results of the North
`American EVT phase I trial. J Vasc Surg 23:543-553, 1996
`
`