The AneuRx stent graft: Four-year results and
`worldwide experience 2000
`
`Christopher K. Zarins, MD,a Rodney A. White, MD,c Frans L. Moll, MD,d Tami Crabtree, MS,e, Daniel
`A. Bloch, PhD,b Kim J. Hodgson, MD,f Mark F. Fillinger, MD,g and Thomas J. Fogarty, MD,a Stanford,
`Torrance, and Santa Rosa, Calif, Nieuwegein, The Netherlands, Springfield, Ill, and Lebanon, NH
`
`Objective: The objective was to review the current results of endovascular abdominal aortic aneurysm repair with the
`AneuRx stent graft and to determine the effectiveness of the device in achieving the primary objective of preventing
`aneurysm rupture.
`Methods: The outcome of all patients treated during the past 4 years in the U.S. AneuRx clinical trial was determined,
`and the worldwide clinical experience was reviewed.
`Results: A total of 1192 patients were treated with the AneuRx stent graft during all phases of the U.S. Clinical Trial
`from June 1996 to November 1999, with follow-up extending to June 2000. Ten (0.8%) patients have had aneurysm
`rupture, with most ruptures (n = 6) occurring in 174 (3.4%) patients treated with an early stiff bifurcation stent graft
`design used in phase I and in the initial stages of phase II. Since the current, flexible, segmented bifurcation stent graft
`design was introduced, four (0.4%) ruptures have occurred among 1018 patients treated. Of these, one was during
`implantation, two were placed too far below the renal arteries, and one patient refused treatment of a type I endoleak.
`Kaplan-Meier analysis of all 1192 patients treated with the AneuRx stent graft including both stent graft designs revealed
`the patient survival rate to be 93% at 1 year, 88% at 2 years, and 86% at 3 years, freedom from conversion to open repair
`to be 98% at 1 year, 97% at 2 years, and 93% at 3 years, and freedom from secondary procedure to be 94% at 1 year, 92%
`at 2 years, and 88% at 3 years. Freedom from aneurysm rupture with the commercially available segmented bifurcation
`stent graft was 99.7% at 1 year, 99.5% at 2 years, and 99.5% at 3 years. The presence or absence of endoleak on contrast
`computed tomography scanning after stent graft placement was not found to be a significant predictor of long-term out-
`come measures. Worldwide experience with the AneuRx device now approaches 10,000 patients.
`Conclusions: Endovascular management of abdominal aortic aneurysms with the AneuRx stent graft has markedly reduced
`the risk of aneurysm rupture while eliminating the need for open aneurysm surgery in 98% of patients at 1 year and 93%
`of patients at 3 years. The device was effective in preventing aneurysm rupture in 99.5% of patients over a 3-year period.
`The overall patient survival rate was 93% at 1 year and 86% at 3 years. (J Vasc Surg 2001;33:S135-45.)
`
`The primary objective in treating patients with aortic
`aneurysms is to prevent aneurysm rupture and death from
`aneurysm rupture. The effectiveness of any treatment of
`abdominal aortic aneurysms must be balanced against the
`risk of the treatment itself and the risk of no treatment,
`that is, the natural history of the aneurysm to rupture.
`Open repair with direct exposure of the aneurysm and
`transmural suture fixation of a prosthetic graft is generally
`effective in preventing aortic aneurysm rupture but
`requires major abdominal surgery and is associated with
`significant morbidity and mortality.1,2 The long-term
`
`From the Division of Vascular Surgery, Stanford University Medical
`Center, Stanforda; Health Research and Policy, Stanford University
`Medical Centerb; Harbor-UCLA Medical Center, Torrance, Californiac;
`St. Antonius Hospital, The Netherlandsd; Medtronic AVE, Santa Rosa,
`Californiae; SIU School of Medicine, Springfield, Illinoisf; and
`Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire.g
`Competition of interest: nil.
`Reprint requests: Christopher K. Zarins, MD, Division of Vascular Surgery,
`Stanford University Medical Center, 300 Pasteur Dr, H3642, Stanford,
`CA 94305-5642.
`Copyright © 2001 by The Society for Vascular Surgery and The American
`Association for Vascular Surgery, a Chapter of the International Society
`of Cardiovascular Surgery.
`0741-5214/2001/$35.00 + 0 24/0/111676
`doi:10.1067/mva.2001.111676
`
`aneurysm-related risk after open repair is greatly reduced
`but persists with the potential for subsequent anastomotic
`or true aneurysmal degeneration. Endovascular aneurysm
`repair with endoluminal placement of a stent graft provides
`a less invasive alternative to standard open surgery with
`reduced morbidity and more rapid patient recovery.3-7
`However, the long-term effectiveness of this approach com-
`pared with that of open surgery or with observation and no
`treatment is not well defined. This report will review the
`current clinical results of abdominal aortic aneurysm treat-
`ment with the AneuRx stent graft and consider issues
`related to its effectiveness against open surgery and the
`natural history of abdominal aortic aneurysms. We will
`also provide a current perspective on the worldwide use of
`the AneuRx stent graft for the treatment of patients with
`abdominal aortic aneurysms.
`
`DEVELOPMENT OF THE ANEURX STENT
`GRAFT
`The AneuRx stent graft was designed and engineered
`by Dr. Thomas J. Fogarty and Fogarty Engineering in
`1993.8-10 The device and delivery system were perfected
`by AneuRx, Inc (Sunnyvale, Calif), and animal experi-
`
`S135
`
`TMT 2106
`Medtronic v. TMT
`IPR2021-01532
`
`

`

`S136 Zarins et al
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`A
`
`B
`
`Fig 1. AneuRx stent graft composed of Nitinol exoskeleton joined to woven polyester graft with multiple polyester sutures. A, Original
`bifurcation module included single 5-cm long, unbending bifurcation stent. B, Currently available device is constructed with individual
`1 cm long stent rings throughout its length and provides flexibility throughout length of stent graft.
`
`ments were concluded at Stanford University and Harbor
`UCLA in 1995.12 European CE mark was granted in
`March 1997. U.S. clinical trials, in accord with Food and
`Drug Administration (FDA) guidelines, were begun in
`1996. The first AneuRx device implanted in the United
`States was by Dr. Rodney White at Harbor UCLA on
`June 3, 1996. The first AneuRx device in Europe was
`implanted by Dr. Frans Moll in the Netherlands in
`September 1996. AneuRx, Inc. was acquired by
`Medtronic, Inc (Minneapolis, Minn) in 1996 and is now
`operating as a part of Medtronic AVE (Santa Rosa, Calif).
`The U.S. phase II clinical trial comparing the AneuRx
`stent graft with standard open surgical aneurysm repair
`began in April 1997 and was carried out at 13 clinical sites
`throughout the United States. The 1-year results were
`reviewed on June 23, 1999, by the FDA Advisory Panel,
`which recommended approval. The FDA granted market
`approval for the AneuRx device on September 28, 1999.
`
`THE ANEURX STENT GRAFT SYSTEM
`The AneuRx stent graft is a modular, bifurcated
`endovascular device designed to treat infrarenal aortic
`aneurysms. Each stent graft module consists of a thin-
`walled, noncrimped, woven polyester graft that is joined
`
`to a nickel-titanium (Nitinol) exoskeleton by numerous
`individual polyester sutures (more than 2000 suture con-
`nections in the bifurcation module). The self-expanding
`Nitinol stent rings provide both radial and columnar
`structural support throughout the length of the graft, and
`the polyester graft provides a smooth, impervious blood
`flow conduit. Each individual stent graft module is loaded
`inside a delivery catheter, which is introduced over a stiff
`0.035-inch guidewire and positioned with x-ray fluoro-
`scopic control.
`The primary aortic bifurcation module is contained in a
`21F (OD) delivery catheter, and the iliac modules are con-
`tained in 16F delivery catheters. A nosecone orientation
`marker and radio-opaque markers on the stent graft allow
`radial and longitudinal orientation of the stent graft under
`fluoroscopic guidance. The self-expanding stent graft is
`deployed by retraction of the delivery catheter-covering
`sheath with a deployment handle. A new tapered nosecone
`and new integrated deployment-delivery system is being
`introduced and will eventually replace the two-component
`delivery system.
`In April 1998, early in the AneuRx clinical trial, the
`FDA approved a manufacturing change of the bifurcation
`stent graft module. The initial bifurcation stent graft mod-
`
`

`

`JOURNAL OF VASCULAR SURGERY
`Volume 33, Number 2
`
`Zarins et al S137
`
`A
`Fig 2. Curved planar reformation of spiral CT angiograms of tortuous abdominal aortic aneurysms treated with AneuRx stent graft (A).
`Stiff bifurcation stent graft module used in 174 patients in clinical trial. Arrows denote stiff bifurcation segment (B). Segmented bifur-
`cation stent graft was used in 1018 patients in clinical trial and is currently FDA-approved for clinical use. Arrows define flexible bifur-
`cation segment.
`
`B
`
`ules were manufactured with a single-unit Nitinol bifurca-
`tion stent, 5 cm in length, coupled proximally and distally
`to individual 1 cm long Nitinol rings joined together end
`to end (Fig l, A). After April 1998, the entire length of the
`bifurcation stent body was constructed with individual 1
`cm long Nitinol rings joined together end to end (Fig l,
`B). The segmented body construction resulted in flexibil-
`ity of the bifurcated graft throughout the length of the
`stent graft and eliminated the stiff, unbending 5 cm long
`proximal bifurcation segment (Fig 2). The commercially
`available AneuRx stent graft has the segmented body con-
`struction.
`
`U. S. MULTICENTER CLINICAL TRIAL
`The AneuRx Stent Graft System was evaluated under
`the FDA Investigational Device Exemption (IDE)
`Application Number G960016 for Endovascular Stent
`Graft and Delivery Catheter. The study was designed in
`three phases. Phase I began in June 1996 and was
`designed as a feasibility study involving 40 patients at four
`study sites. Only one follow-up year was required, but
`
`most patients in this phase continue to be monitored.
`Phase II began in April 1997 and was a prospective, non-
`randomized, controlled clinical investigation comparing
`endovascular repair using the AneuRx stent graft with
`standard open surgical repair. Phase II included 13 study
`sites and included a 1-year follow up, which has been
`extended to 5 years for the patients receiving stent grafts.
`Phase III began in August 1998 and included additional
`patients with stent grafts treated at a total of 19 clinical
`sites (13 phase II sites and 6 new sites). These patients will
`be monitored for 5 years. The clinical study sites are listed
`in the Appendix.
`The numbers of patients treated in the clinical trial as
`of June 3, 2000, are as follows: phase I - 40 patients, phase
`II stent graft - 424 patients, phase II surgical control - 66
`patients, and phase III stent graft - 641 patients. In addi-
`tion, 87 patients during phases II and III were treated
`with the AneuRx stent graft on a compassionate use/high-
`risk basis. The total number of patients treated with the
`AneuRx stent graft during all phases of the clinical trial
`was 1192.
`
`

`

`S138 Zarins et al
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`Fig 3. Freedom from aneurysm rupture. Kaplan-Meier analysis.
`
`Patient selection. Patients with nonruptured infrarenal
`aortic and aortoiliac aneurysms were candidates for the trial
`if the aneurysm met one of the following criteria: larger
`than 5 cm in diameter, between 4 and 5 cm in diameter
`with a documented increase in diameter of 0.5 cm in the
`past 6 months, twice the diameter of the infrarenal neck, or
`saccular. Additional requirements included an infrarenal
`neck between 18 and 26 mm in diameter with a length
`below the most inferior renal artery of at least 1 cm.
`Patient exclusions included ruptured or leaking aneurysms
`and suprarenal, thoracic, and inflammatory aneurysms.
`
`Patient selection criteria in the clinical trial were the same
`for patients in both the stent graft and surgical control
`groups. Patients were evaluated with preoperative and
`postoperative computed tomography (CT) scans and
`Duplex ultrasound scanning and were monitored at inter-
`vals of 1 month, 6 months, and 12 months and annually
`thereafter with clinical evaluation, plain abdominal x-ray
`evaluation, and CT scans.
`Data at each study site are managed by a study coor-
`dinator and externally audited with interval site visits. Data
`are entered into a central database. A data safety and mon-
`
`

`

`JOURNAL OF VASCULAR SURGERY
`Volume 33, Number 2
`
`Zarins et al S139
`
`Fig 4. Survival after endovascular aneurysm repair. Kaplan-Meier analysis.
`
`itoring board reviews adverse event information. Imaging
`data from patients in phase II of the study (CT scans and
`abdominal x-ray evaluations) are independently evaluated
`by a Radiologic Core Laboratory with expertise in radio-
`logic imaging and with no affiliation to any of the study
`sites or investigators. Statistical analysis of the data is over-
`seen by D.A.B. of the Stanford University Medical Center,
`Department of Health Research and Policy.
`Early results. The initial experience of 190 patients
`treated with the AneuRx stent graft (patients in both
`phase I and phase II) was compared with that of 60
`patients in a surgical control group with 1-year follow-up
`and was presented to the International Society for
`Cardiovascular Surgery in June 1998 and published in
`1999.4 There were no significant differences in preoper-
`ative risk factors of comorbidities between the patients in
`the surgical control and stent graft groups. Stent graft
`deployment was successful in 97% of patients. There was
`no difference in operative mortality between the groups
`(0% in the surgery group and 1% in the stent graft
`group). Patients who underwent stent graft repair had
`60% less blood loss compared with patients who under-
`
`went open surgery (P < .001), and these patients
`required 80% less blood transfusion (P < .05). There was
`a marked reduction in the time to extubation, discharge
`from the intensive care unit, ambulation without assis-
`tance, and eating a regular diet in the stent graft group
`compared with the surgery group. Major morbidity was
`reduced by 50% in the stent graft group (P < .05), and
`hospital length of stay was reduced from 9.3 to 3.4 days
`(P < .001).4
`Phase II clinical trial. The phase II controlled trial of
`416 patients in the stent graft group compared with 66
`patients undergoing open aneurysm repair was presented
`to the FDA Advisory Panel on June 23, 1999. The results
`did not differ significantly from the previously published
`results and demonstrated the following. Successful graft
`deployment was achieved in 98% of patients, and the sur-
`gical conversion rate was 1.5%; there was no difference in
`the 30-day mortality rate between patients in the stent
`graft group (2%) and those in the surgical group (0%).
`There was a 50% reduction in major morbidity compared
`with open repair, a 66% reduction in blood loss, and a 63%
`reduction in hospital length of stay. Patients recovered
`
`

`

`S140 Zarins et al
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`Fig 5. Freedom from surgical conversion. Kaplan-Meier analysis.
`
`more quickly and had earlier return to function compared
`with patients undergoing open surgery. The physician-
`reported endoleak rate at the time of hospital discharge
`was 38%, which was reduced to 13% at 1 month, 16% at 6
`months, and 11% at 12 months. Two (0.5%) patients had
`aneurysm rupture, one during the implantation procedure
`and one 14 months after implantation of a stiff bifurcation
`stent graft. These ruptures are included in the later analy-
`sis in this article. Five percent of patients required a sec-
`ondary endovascular procedure for endoleak, and 2%
`required a secondary procedure for nonpatency of the
`stent graft. The primary graft patency rate was 98%, and
`the secondary graft patency rate was 99%.
`
`FOUR-YEAR RESULTS OF U.S. ANEURX
`CLINICAL TRIAL
`All patients entered into the endovascular treatment
`arm beginning with the first patient in phase I (June 3,
`1996) and ending with the last patient treated in phase III
`on November 4, 1999, were reviewed. The follow-up
`period extended for 4 years through June 3, 2000, with a
`
`mean of 73.4 + 8.0 SD months. All patients were included
`in this analysis, with no patient exclusions for the initial
`learning curve, device manufacturing changes, compas-
`sionate use, or other reasons.
`A total of 1192 patients were reviewed to determine
`long-term outcome on intent-to-treat basis. Patients
`included 1058 (89%) men and 134 (11%) women with a
`mean age of 73.4 ± 8.0 SD years (range 45 to 96 years).
`Patients had multiple risk factors and comorbidities, and 92%
`of patients had American Society of Anesthesiologists (ASA)
`risk classification of III or IV. The mean preoperative
`aneurysm diameter was 5.6 + 0.3 SD cm. The mean
`aneurysm neck length was 27.3 + 12.2 SD mm, and the
`mean neck diameter was 22.3 + 2.9 SD mm. The primary
`end points for analysis included aneurysm rupture, death
`from aneurysm rupture, death from any cause, surgical con-
`version, and secondary procedures for endoleak/migration
`or graft nonpatency.
`Aneurysm rupture. The first report of aneurysm rup-
`ture after successful AneuRx stent graft repair was in
`1999.4 This was a patient in phase I with a 5.5-cm
`
`

`

`JOURNAL OF VASCULAR SURGERY
`Volume 33, Number 2
`
`Zarins et al S141
`
`Fig 6. Freedom from secondary procedure. Kaplan Meier analysis.
`
`aneurysm and a short angulated neck who had a rupture
`23 months after endovascular repair. The main bifurcation
`module used in this patient was the early stiff body design,
`which is no longer manufactured. Since that time there
`have been seven additional cases of aneurysm rupture after
`successful AneuRx stent graft repair. Three of these
`patients were reported by Politz et al13 in the March 2000
`issue of the Journal of Vascular Surgery. Seven of the eight
`patients who had rupture including the three reported by
`Politz et al and two additional patients who had aneurysm
`rupture during the implantation procedure were reported
`by Zarins et al14 in the May 2000 issue of the Journal of
`Vascular Surgery. Thus as of June 3, 2000, 4 years since
`the first device implantation, there have been a total of 10
`aneurysm ruptures (two during the implant procedure,
`eight after implantation) among the 1192 (0.8%) patients
`entered into the endovascular treatment arm of the U.S.
`AneuRx Clinical Trial.
`
`Two of the ruptures occurred during the treatment
`period, and both underwent successful surgical conver-
`sion. One was related to intraoperative balloon dilation of
`the proximal aortic neck in an attempt to seal a type I
`endoleak, and the other was related to instrumentation
`and perforation of the aneurysm sac. The eight ruptures
`after successful endovascular repair occurred 3 weeks to 26
`months (mean 17 months) after endovascular repair. Five
`patients survived open surgical repair, two died in the peri-
`operativc period, and one died after refusing operation
`(rupture-related mortality: 0.3%).
`Before rupture occurred, two patients had type I iliac
`endoleaks and aneurysm enlargement and refused recom-
`mended treatment with iliac extender modules. Six patients
`had no documented endoleak and no aneurysm enlarge-
`ment. Four ruptured as a result of proximal fixation failure
`in angulated necks, and two ruptured as a result of
`aneurysm tortuosity and separation of the iliac limb from
`
`

`

`S142 Zarins et al
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`the junction gate. Retrospective analysis of all patients with
`aneurysm rupture after successful stent graft repair revealed
`evidence of insecure fixation either proximally, distally, or
`at the junction gate, which could have been treated with
`stent graft extender modules before rupture.14
`Single-unit versus segmented bifurcation stent
`body. All 40 patients in phase I were treated with the orig-
`inal, single-unit, stiff bifurcation body, as were 134 patients
`in the beginning of the phase II clinical trail. Subsequently,
`all patients treated with the AneuRx stent graft have
`received flexible, segmented bifurcation stent grafts. Thus
`174 (15%) patients were treated with the stiff body stent
`graft, whereas 1018 (85%) patients were treated with the
`flexible stent graft. Among the 10 patients with rupture, six
`had received the stiff body stent graft, and four had
`received the flexible stent graft. Thus the risk of aneurysm
`rupture with the stiff body stent graft was much higher (6
`of 174, 3.4%) than with the flexible body stent graft (4 of
`1018, 0.4%). Kaplan-Meier analysis revealed a significant
`reduction in the risk of aneurysm rupture among patients
`receiving the flexible, segmented stent graft compared with
`those treated with the stiff body stent graft (P < .002,
`Mantel-Haenszel test, Fig 3). Freedom from aneurysm
`rupture with the flexible, segmented stent graft was 99.7%
`at 1 year, 99.5% at 2 years, and 99.5% at 3 years (Fig 3).
`One of the two patients with periprocedural rupture
`received a stiff device, whereas the other received a flexible
`device. Among the eight patients with rupture after suc-
`cessful repair, five received the stiff body graft and three
`received the flexible graft. Two of the five stiff devices had
`been placed in aneurysms with short, angulated necks with
`ultimate loss of proximal fixation, and two experienced
`progressive anterior angulation of the stent graft over time,
`resulting in separation of the iliac limb from the junction
`gate. It is likely that the inability of the stiff bifurcation
`module to flex in tortuous or shrinking aneurysms exacer-
`bated the angulation forces at either the proximal or distal
`end of the rigid 5 cm long segment, resulting in ultimate
`fixation failure. The fifth patient had a type I iliac endoleak
`and refused treatment. Among the three patients with flex-
`ible stent grafts who had rupture, two had severely angu-
`lated necks with initial positioning of the device low in the
`neck, well below the renal arteries. The third patient had a
`type I iliac endoleak and refused treatment.
`Survival. Among the 1192 patients, a total of 119
`(10%) patients have died during the 4-year time course of
`study. The 30-day mortality rate was 2% (23 of 1192).
`Four (0.3%) patients died of aneurysm rupture after suc-
`cessful endovascular repair, three after emergent opera-
`tion, and one refused operation. One patient died of
`aneurysm rupture before treatment and one of a ruptured
`arch aneurysm after treatment. The remaining patients
`died of nonaneurysm causes. The causes of death were
`cardiac in 49, cancer in 20, pulmonary in 19, renal in 7,
`
`gastrointestinal in 4, sepsis and multisystem organ failure
`in 5, stroke in 3, pulmonary embolism in 1, and mesen-
`teric thrombosis in 1. The cause of death for four patients
`has not been determined. Kaplan-Meier survival analysis
`(Fig 4) reveals a 1-year survival rate of 93%, a 2-year sur-
`vival rate of 88%, and a 3-year survival of 86%.
`Secondary procedures. A total of 33 (2.8%) patients
`have undergone conversion to open surgical repair includ-
`ing the nine patients who underwent open repair for
`aneurysm rupture. There were 15 (1.3%) surgical conver-
`sions within the first 30 days and 18 (1.5%) late surgical
`conversions at an average time of 22 months after endovas-
`cular repair. The Kaplan-Meier estimate of freedom from
`surgical conversion is 98% at 1 year, 97% at 2 years, and
`93% at 3 years (Fig 5). A total of 107 secondary procedures
`have been performed in 94 (8%) patients; 82 patients had
`one procedure, 11 patients had two procedures, and one
`patient had more than two procedures. Freedom from sec-
`ondary procedure (Kaplan-Meier estimate) was 94% at 1
`year, 92% at 2 years, and 88% at 3 years (Fig 6).
`
`ENDOLEAKS
`Evidence of blood flow in the aneurysm sac
`(endoleak) is commonly seen in patients after endovascu-
`lar aneurysm repair. Although some have considered this
`to be evidence of an unsuccessful aneurysm repair,15-18 the
`true significance of this finding remains unclear.19,20 To
`determine whether evidence of blood flow in the
`aneurysm sac (endoleak) was a meaningful predictor of
`clinical outcome after successful endovascular aneurysm
`repair, we reviewed all patients in phase II of the AneuRx
`Multicenter Clinical Trial with successful stent graft
`implantation and predischarge contrast CT imaging. The
`clinical outcome of patients with evidence of endoleak was
`compared with that of patients without evidence of
`endoleak. The results were presented at the Western
`Vascular Society Meeting in Lake Tahoe in September
`1999 and were published in the July 2000 issue of the
`Journal of Vascular Surgery.21
`The determination of whether an endoleak was pres-
`ent after endovascular repair was independently deter-
`mined by (1) the 13 clinical centers before hospital
`discharge, at 1 month, 6 months, 12 months, and 24
`months, and (2) a Radiologic Core Laboratory that inde-
`pendently reviewed the contrast CTs at predischarge, 6
`months, 12 months, and 24 months. The Centers
`reported endoleaks in 152 (38%) of 398 patients on pre-
`discharge CT, whereas the Core Lab reported endoleaks
`in 50% of these patients (P < .001). Follow-up extended
`to 2 years (mean 10 ± 4 months). There were no differ-
`ences between patients with and patients without
`endoleak before discharge in the following outcome mea-
`sures: patient survival, aneurysm rupture, surgical conver-
`sion, need for a secondary procedure, aneurysm
`
`

`

`JOURNAL OF VASCULAR SURGERY
`Volume 33, Number 2
`
`Zarins et al S143
`
`enlargement more than 5 mm, and appearance of a new
`endoleak or stent graft migration. Despite a higher
`endoleak rate reported by the Core Lab, neither Core
`Lab-defined endoleaks nor Center-defined endoleaks at
`discharge were significantly related to subsequent out-
`come measures. The outcome of patients with type I or
`type II endoleaks before discharge was no different from
`that of patients with no endoleak.21
`At 1 month the endoleak rate had decreased to 13%.
`Although patients with persisting endoleaks were more
`likely to have aneurysm enlargement at 1 year, there was
`no difference in patient survival, aneurysm rupture rate,
`surgical conversion, new endoleak, or stent graft migra-
`tion between patients with and those without endoleak at
`1 month. Kaplan-Meier survival of all patients undergoing
`endovascular aneurysm repair was 96% at 1 year and was
`independent of endoleak status.21
`Thus the presence or absence of endoleak on CT scan
`after AneuRx stent graft aneurysm repair does not appear
`to predict long-term outcome. Although the identification
`of blood flow in the aneurysm sac after endovascular repair
`is a meaningful finding, the usefulness of endoleak as a pri-
`mary indicator of procedural success or failure remains
`unproved. Thus all patients who have undergone endovas-
`cular aneurysm repair should be carefully monitored
`regardless of endoleak status.
`
`WORLDWIDE EXPERIENCE
`Since the initiation of clinical trials in the United
`States and the introduction of the device in Europe in
`1996, use of the AneuRx stent graft has increased on an
`annual basis. The worldwide experience of endovascular
`abdominal aortic aneurysm repair with the AneuRx stent
`graft now approaches 10,000 patients. A total of 1192
`patients were included in the U.S. clinical trial and are
`undergoing monitored 5-year follow-up. During the past
`8 months, since FDA market approval, 895 physicians
`have been trained to implant the device in the United
`States, and 370 hospitals have begun using the AneuRx
`stent graft. A total of 6700 stent grafts have been shipped
`for implantation in the United States. In Europe, the
`Middle East, and Australia, more than 2500 patients have
`been treated. There have been eight aneurysm ruptures in
`the United States and one aneurysm rupture in Europe
`after successful implantation, plus two implantation proce-
`dure-related ruptures. Six of these 11 ruptures were with
`the early prototype stiff body bifurcation stent design dur-
`ing the U.S. clinical trial. Therefore as of June 2000 the
`worldwide reported aneurysm rupture rate after endovas-
`cular repair with the commercially available AneuRx stent
`graft is 5 (0.06%) among more than 9000 patients treated
`thus far.
`As of December 1999 a total of 386 patients treated
`with the AneuRx graft have been entered in the Eurostar
`
`Registry from 35 European and Middle East Centers.
`Follow-up evaluation has been carried out on 124 patients
`at 12 months and 30 patients at 24 months. Patient sur-
`vival at 12 months is 87% (Kaplan Meier estimate) and at
`24 months is 83% (Kaplan Meier estimate). In the U.K.
`experience there has been one reported aneurysm rupture
`after AneuRx stent graft repair with successful surgical
`conversion.
`
`RUPTURE RISK OF UNTREATED
`ANEURYSMS
`The risk of aneurysm rupture of untreated aneurysms
`is a continuous function of aneurysm size. The annual
`rupture rate is near zero for aneurysms less than 4 cm. The
`risk is approximately 1% per year for aneurysms 4 to 4.9
`cm in diameter, 11% per year for aneurysms 5 to 5.9 cm,
`and 25% per year or more for aneurysms greater than 6
`cm.22-24 The mean diameter for aneurysms treated in the
`AneuRx clinical trial was 5.6 ± 0.3 cm. Thus the 1-year
`risk of aneurysm rupture was reduced by more than 98%
`compared with the expected risk of rupture of untreated
`aneurysms.
`Although the risk of aneurysm rupture after endovas-
`cular aneurysm repair is very low and markedly reduced
`compared with the likelihood of rupture of untreated
`aneurysms,24 the possibility of rupture persists even in
`patients with no endoleak after the procedure.25,26 The
`cause of rupture with the AneuRx stent graft can be traced
`to poor proximal, distal, or junctional fixation of the stent
`graft, with a higher risk of rupture in patients who were
`treated with the early stiff body prototype stent graft and
`who had significant aneurysm or neck angulation.
`Therefore all patients treated with endovascular aneurysm
`repair should continue to be monitored after the proce-
`dure with follow-up clinical evaluation and imaging stud-
`ies. Patients with evidence of insecure stent graft fixation
`or evidence of aneurysm enlargement should undergo fur-
`ther endovascular treatment or open surgical repair. The
`cause of rupture of aneurysms after endovascular repair
`with other devices may be different from the AneuRx
`experience and include fabric tears and strut failures.27-30
`
`COMPARISON TO OPEN REPAIR
`Endovascular repair compares favorably to open surgi-
`cal repair in the short term, with a significant reduction in
`morbidity, reduced blood loss, shorter hospital stay, and
`earlier return to function.4,7 There was no difference in 1-
`year patient survival between patients treated with
`endovascular procedures and those treated with open
`surgery in the phase II AneuRx clinical trial. In a similar
`fashion, concurrent comparison of endoluminal and open
`repair of aneurysms revealed no differences in survival
`rate.7 Long-term controlled trials comparing open with
`endovascular repair have not yet been reported. Although
`
`

`

`S144 Zarins et al
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`some have assumed that patients who have undergone
`open surgical repair are no longer at risk of aneurysm rup-
`ture once they have recovered from the operation, this has
`not proved to be the case. Patients are at risk of pseudo-
`aneurysm rupture, suprarenal and iliac aneurysm forma-
`tion, graft infection, aortoenteric fistula, and graft
`thrombosis after open surgical repair.31-36 The risk of
`death from late rupture of abdominal aneurysms and
`pseudoaneurysms after elective open surgery was 5% in
`three large series of 1126 patients monitored for an aver-
`age of 5 years.31,35,36 The long-term late aneurysm-related
`death rate after open aneurysm repair including deaths
`from ruptured true and false aneurysms, aortoenteric fis-
`tula, and graft infections among 1474 patients is 6%
`(range 1.5% to 7.5%).31,34-36 Recent reports have demon-
`strated that “endoleaks” and ruptures can occur after con-
`ventional open aneurysm repair37 just as they can after
`endovascular repair. Thus the true differences in long-
`term outcome between endovascular and open aneurysm
`repair remain to be determined. However, the long-term
`risk of aneurysm rupture after endovascular treatment
`appears, thus far, to be no higher and perhaps to be lower
`than standard open surgical repair.
`
`SUMMARY
`The AneuRx stent graft has markedly reduced the mor-
`bidity of aortic aneurysm repair and is effective in prevent-
`ing aneurysm rupture in the great majority of patients. In
`clinical trial follow-up extending to 4 years, the device is
`effective in preventing aneurysm rupture in 99.5% of
`patients. Most aneurysm ruptures have occurred in patients
`in early clinical trials who were treated with a stiff bifurca-
`tion stent gr