The Ancure endografting system: An update
`
`Michel S. Makaroun, MD, Pittsburgh, Pa
`
`The Ancure endografting system (Guidant Cardiac and Vascular Division, Menlo Park, Calif) features a unibody, non-
`supported woven polyester graft designed to treat abdominal aortic aneurysms. It is constructed in tube, bifurcated, and
`aortoiliac configurations. The attachment system consists of a frame with four independent V-shaped double hooks that
`penetrate the arterial wall for fixation. There are separate attachment systems at the proximal and distal ends of the endo-
`prosthesis. In September 1999, the Food and Drug Administration (FDA) approved the tube and bifurcated devices for
`general use. The aortoiliac device is under present consideration of the FDA. Phase II and III clinical trials of the sys-
`tem enrolled over 870 patients from the end of 1995 to the summer of 1999. The device was deployed successfully in
`90% to 96% of cases, depending on the configuration and the phase of the trial. Mortality rates were similar to those of
`concurrent open surgical control rates, but serious morbidity was reduced. Long-term follow-up of the bifurcated group
`from phase II showed only one migration and no ruptures. Aneurysm size reduction in this group was noted in 51.3%
`of patients at 1 year and 68.5% at 2 years. In the same subset, type I endoleaks were noted in 2.7% at 1 year and 1.3%
`at 2 years. All postoperative imaging studies were reviewed by a core laboratory facility. The advantages of the Ancure
`system include solid fixation, flexibility in accommodating morphologic changes, and excellent long-term clinical per-
`formance. The disadvantages include a large introducer system and the potential for limb obstruction by compression or
`angulation. However, limb compromise responds well to intraluminal stenting. The expected FDA approval of the aor-
`toiliac device and a larger variety of graft sizes should expand the number of patients who can be treated with this sys-
`tem. (J Vasc Surg 2001;33:S129-34.)
`
`Endovascular Technologies (EVT) was founded in
`1989 to develop an endovascular graft to treat abdominal
`aortic aneurysms (AAAs). The concept was based on the
`work of Harrison Lazarus, MD, from the early 1980s
`when he successfully implanted a polyester tube graft with
`attached staples into a sheep aorta using a modified endo-
`tracheal tube and followed some of the implants for 6
`months. Wesley Moore, MD, performed the first human
`implantation of an EVT tube graft at the University of
`California, Los Angeles, Medical Center in February
`1993. It was the first industry-manufactured device to
`enter clinical trials in the United States. A bifurcated
`design was introduced in late 1994, and later an aorto-
`monoiliac configuration was added (Fig 1). Guidant
`Corporation acquired EVT in 1997 and continued the
`development of the system.
`Trial enrollment was halted in early 1995 after hook
`breaks were discovered on an explanted graft during a late
`open surgical conversion for a persistent endoleak. Many
`additional patients were subsequently discovered to have
`hook and frame breaks in the attachment systems during
`routine radiography. The second generation of grafts was
`
`From the Division of Vascular Surgery, University of Pittsburgh School of
`Medicine.
`Competition of interest: MSM serves on the Clinical Advisory Panel of
`Guidant Cardiac and Vascular Division, Menlo Park, Calif.
`Reprint requests: Michel S. Makaroun, MD, Division of Vascular Surgery,
`A-1011 PUH/ 200 Lothrop St, Pittsburgh, PA 15213.
`Copyright © 2001 by The Society for Vascular Surgery and The American
`Association for Vascular Surgery, a Chapter of the International Society
`for Cardiovascular Surgery.
`0741-5214/2001/$35.00 + 0 24/0/111673
`doi:10.1067/mva.2001.111673
`
`reintroduced at the end of 1995 after a redesign of the
`attachment system and the hook configuration made the
`hooks 280 times stronger during fatigue bench testing.
`No new fractures have been noted with the new design in
`nearly 3000 implants since that time. The new graft has
`not been modified, although the deployment system and
`handle underwent a significant overhaul before phase III
`started in 1998. The new deployment catheter, the
`Ancure (Guidant Cardiac and Vascular Division, Menlo
`Park, Calif) endograft system, has serial interlocks to guide
`a simplified deployment process. This system was used
`during the phase III trials. The results of patients enrolled
`in phase II and the first 89 patients enrolled in phase III
`that were presented to a Food and Drug Administration
`(FDA) panel in June 1999 resulted in approval for distri-
`bution in September 1999.
`
`DESCRIPTION OF THE DEVICE
`The implanted endografts deployment during phase II
`that used the old deployment system and during phase III
`that used the Ancure system are identical. The fabric is a
`woven polyester material that is similar to the standard
`surgical grafts. The attachment systems are manufactured
`from elgiloy and are hand sutured to the graft. The prox-
`imal attachment system consists of a Z frame and four V-
`shaped double hooks (Fig 2). The hooks are intended to
`penetrate the arterial wall with the assistance of a balloon
`inflated to 2 atm. A variation of this system is present on
`the distal end of all configurations. The bifurcated implant
`is a unibody single-piece construction similar to surgically
`implanted bifurcated grafts. The lateral edges of the
`
`S129
`
`TMT 2107
`Medtronic v. TMT
`IPR2021-01532
`
`

`

`S130 Makaroun
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`Fig 1. The Ancure endograft is manufactured in three configurations: tube, bifurcated, and aortoiliac.
`
`the bifurcated devices presently account for more than
`90% of the grafts used, their results will be highlighted.
`Two-year data on a large number of patients have now
`been collected and analyzed for the bifurcated graft; the
`salient features will be outlined. Two-year outcome analy-
`sis for patients with the tube and aortoiliac grafts has not
`been completed yet, but the results at 12 months are
`included.
`All imaging-based data reflect the interpretation of
`independent observers at a core laboratory facility that was
`established at the Cleveland Clinic, unless otherwise
`noted. Clinical data was collected at the investigation sites
`and duly monitored. Our group at the University of
`Pittsburgh Medical Center participated actively in both
`phase II and III trials.1-5
`
`Phase I clinical trial
`The devices implanted during this phase coincided
`mostly with the use of the first generation attachment sys-
`tem that developed hook and frame fractures. Patients who
`were treated during this phase from 1993 to early 1995
`received 102 tubes and 12 bifurcated grafts. Several patients
`experienced problems with migration and endoleaks and
`required conversion. Four patients in this group experi-
`enced rupture of the AAA during the last 5 years.
`
`Phase II trials
`Separate multicenter phase II trials were conducted
`for each configuration. A prospectively enrolled, concur-
`rent open surgical group served as a control arm for all
`three trials. This cohort was enrolled during the first phase
`II trial that evaluated the tube endograft. All patients
`
`Fig 2. The proximal attachment system of the Ancure graft.
`
`implant contain radiopaque markers to facilitate visualiza-
`tion of the endograft during deployment and follow-up.
`The Ancure catheter is typically advanced through a
`sheath with dual valves to reduce blood loss. The deploy-
`ment of the graft with the Ancure system is achieved from
`a single handle with multiple controls for the independent
`release of the proximal and ipsilateral distal attachment
`systems and the manipulation of the included aortic bal-
`loon. The contralateral attachment system deployment in
`bifurcated grafts is achieved through a special proprietary
`torque catheter that is introduced from the contralateral
`femoral artery.
`
`CLINICAL TRIALS
`An exhaustive description of all the clinical trials and
`their results is beyond the scope of this update. Because
`
`

`

`JOURNAL OF VASCULAR SURGERY
`Volume 33, Number 2
`
`Makaroun S131
`
`A
`
`B
`Fig 3. Survival by life table analysis of tube (A) and bifurcated (B) configurations in patients (broken line) compared with open surgical
`control subjects (solid line).
`
`enrolled in the trial were acceptable surgical candidates
`and well matched, except for anatomic considerations.
`Patients with a proximal aortic neck at least 15 mm in
`length and a distal neck at least 12 mm in length under-
`went implantation with a tube endograft. An aortic neck
`size not to exceed 26 mm in diameter and at least one suit-
`able iliac artery to introduce the device were also required.
`Patients who were anatomically not suited for the tube
`endograft were enrolled in the standard surgical arm, if
`they fit all other inclusion criteria and signed an informed
`consent. Most patients did not qualify for the tube endo-
`graft because of the absence of a satisfactory distal neck
`length.
`The bifurcated and aortoiliac trials were activated later
`than the tube trial. Investigative centers participated in
`one, two, or all three trials. Anatomic considerations dic-
`
`tated which endograft configuration to use and in which
`phase II trial subsequent patients would be enrolled. A
`total of 542 patients were enrolled to receive 153 tubes
`and 268 bifurcated and 121 aortoiliac grafts. Standard
`open procedures were performed on 111 patients as part
`of the control group.
`Demographics and risk factors were similar among
`groups. A notable exception is that few female patients
`qualified for the endovascular treatment arms, mainly
`because of small or diseased iliac arteries. The deployment
`success rate (defined as successful implantation in the
`desired location) exceeded 90% with all endograft config-
`urations. Results were based on an intent to treat analysis,
`and all patients who were enrolled were included in the
`analysis whether or not they received an endograft. Blood
`loss, use of the intensive care unit, and hospital length of
`
`

`

`S132 Makaroun
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`Table I. Death and selected morbidity in phase II tube and bifurcated configurations in patients compared with open
`surgical control subjects
`
`Tube (n = 153)
`
`Bifurcated (n = 268)
`
`Control (n = 111)
`
`P value
`
`Deaths (%)
`Respiratory complications (%)
`Cardiac complications (%)
`Bleeding (%)
`Bowel complications (%)
`Wound complications (%)
`Renal insufficiency (%)
`Graft thrombosis (%)
`
`0
`11 (7)*
`16 (11)*
`13 (9)*
`8 (5)
`8 (5)
`5 (3)
`0
`
`NS, Not significant.
`*Statistical significance is compared with control subjects.
`
`7 (2.6)
`27 (10)*
`36 (13)
`42 (16)
`8 (3)
`9 (3)
`22 (8)*
`7 (3)*
`
`3 (2.7)
`25 (22)
`23 (21)
`44 (40)
`9 (8)
`2 (2)
`2 (2)
`0
`
`NS
`<.01*
`<.01*
`<.01*
`NS
`NS
`.02*
`<.01*
`
`Table II. Late results of Ancure grafts in patients enrolled in phase II trials
`
`Tube
`
`Bifurcated
`
`Aortoiliac
`
`%AAA shrinkage
`12 Months
`24 Months
`%AAA expansion (%)
`12 Months
`24 Months
`%Endoleaks at 12 months (%)
`Type I
`Type II
`Indeterminate
`%Endoleaks at 24 months
`Type I
`Type II
`Late surgical conversions* (n)
`Ruptures (n)
`Attachment system failures (n)
`Proximal system migration (n)
`
`43 (44/103)
`N/A
`
`5.8 (6/103)
`N/A
`
`3.8 (4/104)
`19.2 (20/104)
`1.9 (2/104)
`
`N/A
`N/A
`1
`0
`0
`0
`
`51.3 (96/187)
`68.5 (61/89)
`
`2.1 (4/187)
`1.1 (1/89)
`
`2.7 (5/196)
`21.9 (43/196)
`5.1 (10/196)
`
`1.1 (1/92)
`19.6 (18/92)
`2
`0
`0
`1
`
`58 (38/65)
`N/A
`
`4.6 (3/65)
`N/A
`
`2.9 (2/69)
`21.7 (15/69)
`7.2 (5/69)
`
`N/A
`N/A
`1
`0
`0
`0
`
`N/A, Final analysis not available yet.
`*Twelve months for tube and aorto-iliac configurations; 24 months for bifurcated configuration.
`
`stay were reduced in all endograft groups compared with
`the control subjects. The reductions were statistically sig-
`nificant.
`Mortality and morbidity. Perioperative mortality
`and selected morbidity parameters are presented in Table I
`for the patients who were treated with tube or bifurcated
`endografts and control patients. Mortality rates were similar
`among groups, with no deaths noted in the tube endograft
`group. Blood loss and cardiac and pulmonary morbidity
`were significantly reduced in the endograft groups. Graft
`thrombosis and renal complications, most of them mild and
`temporary, were increased in the bifurcated group.
`Aneurysmal sac shrinkage. The salient late results
`are detailed in Table II. Reduction in AAA size was
`defined as at least a 5-mm reduction in the largest AAA
`diameter on computed tomography scan. Of the bifur-
`cated grafts, 51.3% grafts were associated with an AAA size
`reduction at 1 year and 68.5% at 2 years. AAA shrinkage
`occurred less often in patients with endoleaks. A limited
`number of patients showed sac expansion that was associ-
`ated with an endoleak. One AAA enlarged in the absence
`of any demonstrable endoleak.
`
`Endoleaks. Endoleak rates presented in Table II were
`tabulated from core laboratory data that were based on
`review of contrast computed tomography scans and color
`Duplex ultrasound scans. At 12 months endoleaks were
`present in 25% of patients who were treated with a tube
`graft and 30% of patients who were treated with a bifur-
`cated graft. The incidence of type I endoleaks still present
`at 1 and 2 years, however, is very small, with a consider-
`able number of type II endoleaks. These reported rates
`deserve additional discussion. Discrepancies in reported
`endoleaks between the core laboratory and investigators
`persisted throughout the study. Investigators diagnosed
`endoleaks less than one half as often as the core laboratory.
`The availability of all computed tomography raw data to
`investigators may have allowed a better analysis of ques-
`tionable cases. The high rate of shrinkage of the AAA
`(30%-40%) in the presence of the endoleaks reported by
`the core laboratory seems to support the diagnosis of the
`investigators.
`Secondary interventions to correct endoleaks were not
`common. Only two late open conversions at 4 months and
`7 months were performed for proximal endoleaks with
`
`

`

`JOURNAL OF VASCULAR SURGERY
`Volume 33, Number 2
`
`Makaroun S133
`
`bifurcated grafts, and only one was performed with a tube
`graft. Because the protocols did not dictate or suggest
`possible remedies for endoleaks, treatment regimens var-
`ied considerably among centers. Watchful follow-up, coil-
`ing, limited revisions, or conversions were all used to a
`varying degree. The collective experience of all investiga-
`tors is not available because interventions were typically
`late, with no review of interventions by the core labora-
`tory. At the University of Pittsburgh, we performed arte-
`riography and coiling of all persistent endoleaks at 6
`months (10% of patients with implants), which resulted in
`a low incidence of endoleaks at 1 year (3%) and none at 2
`years. This treatment modality has been very effective in
`our hands.2,3,5
`Late results. During the follow-up of all patients who
`were enrolled in the trials after 1995, no attachment sys-
`tem failures or ruptures have been noted so far. One 10-
`mm migration in a bifurcated proximal attachment system
`was noted at 1 year. The migration was associated with
`neck dilatation (>6 mm). No endoleak or AAA expansion
`was noted with this migration. At least one patient who
`was treated with a tube endograft had a cephalad migra-
`tion of the inferior end of the graft at 2 years. He was suc-
`cessfully treated with a second graft that was carried to the
`iliac level. This type of migration is due to effacement of
`the distal neck that is known to be prone to significant
`enlargement over time.6 This complication can easily be
`avoided by the use of only tube grafts in patients with a
`long distal neck.
`
`Phase III trials
`The phase III trials have used the new deployment sys-
`tem of the Ancure endograft but the identical implant.
`Enrollment started in early 1998 and terminated with the
`FDA approval in September 1999. Preoperative parame-
`ters were similar to those in the phase II trials. Tube
`devices were implanted in 41 patients, and bifurcated
`endografts were implanted in 290 of 352 patients who
`enrolled in the phase III study. Deployment success was
`94%. Operating time was initially higher for the new
`Ancure deployment system, presumably because of the
`learning curve. Because the deployment technique was
`different, there were more reported twists (9%) with the
`early Ancure experience. This improved significantly over
`time. It is our distinct feeling at the University of
`Pittsburgh that a greater accuracy of placement of the
`proximal attachment system is achieved with the new
`deployment system.
`An analysis of the results after 182 patients with the
`bifurcated Ancure grafts were enrolled in phase III
`showed some significant differences from phase II patients
`with bifurcated endografts. The median hospital stay
`decreased from 3 to 2 days; cardiac complications
`decreased from 13.4% to 5.5%, and pulmonary complica-
`
`tions decreased from 10.1% to 1.1%. Most of the differ-
`ences are likely due to better case selection and more
`familiarity with the new technology. The incidence of
`endoleaks as estimated by the investigators at discharge
`among 290 patients (26.6%) was similar to the incidence
`during phase II (24.3%). The author, however, noted a
`considerable reduction of endoleaks during this phase of
`the trials.5
`
`SPECIAL ISSUES
`Rupture. Rupture of the AAA after endovascular
`treatment has been reported by many authors, including
`the recent publication of a significant series after repair
`with a commercially available device in the United States.7
`No ruptures have yet been encountered in the United
`States with the use of the Ancure grafts that were
`implanted after the attachment system was revised at the
`end of 1995. However, a case of rupture has recently been
`encountered in Germany. The patient underwent a tube
`endograft implantation in June 1996. Follow-up in
`August 1997 showed no abnormalities. The patient
`refused further evaluation until he experienced a rupture
`in March 2000. The distal attachment had migrated
`cephalad because of effacement of the distal neck during a
`period of 3 years with no observation. The patient sur-
`vived the emergency conversion to open repair. This case
`highlights the necessity for lifelong follow-up.
`Limb obstruction. The limbs of the Ancure graft are
`not supported and are thus prone to the development of
`luminal compromise from a variety of sources. Heavy dis-
`tal aortic calcifications or proximal iliac stenoses can result
`in compression of the lumen. Graft folds that develop
`from oversized limbs or angulated and tortuous iliac arter-
`ies can also be the cause of obstruction. Patients were eval-
`uated early in phase II with limb occlusions before this was
`commonly appreciated. Limb occlusions were treated by a
`variety of interventional means, such as lytic therapy or
`surgical options (eg, femorofemoral bypass grafting).
`Increased awareness of the problem prompted routine
`careful assessment of limb flow in the operating room
`before the termination of the procedure. Intravascular
`ultrasound scanning is best suited for this task, but other
`methods have been used. When luminal compromise is
`noted, kissing balloon dilatation or elgiloy stent placement
`(such as WallStent; Boston Scientific, Boston, Mass) in the
`lumen usually prevents further problems and has been suc-
`cessful in more than 97% of patients. During phase II tri-
`als, 38% of patients with bifurcated implants were noted to
`have at least some degree of limb abnormalities that were
`sufficient to require additional intervention. However,
`only 10.3% of patients (one fourth of those affected) man-
`ifested problems after operation. In the early part of the
`phase III trial, only two of 77 patients (3%) needed treat-
`ment for limb abnormalities after operation.
`
`

`

`S134 Makaroun
`
`JOURNAL OF VASCULAR SURGERY
`February 2001
`
`Long-term survival. One of the yardsticks in com-
`paring therapies is the effect on long-term survival. So far,
`endovascular AAA treatment with the Ancure endograft
`has resulted in essentially identical long-term survival
`when compared with standard surgical repair. Survival
`curves for patients with tube and bifurcated implants that
`are compared with control subjects are shown in Fig 3.
`
`TOTAL CLINICAL EXPERIENCE
`The Ancure endograft system has been available to
`selected centers in Europe and Australia for many years; it
`is estimated that more than 500 patients have undergone
`implantation procedures outside the United States.
`Because of the FDA approval, more than 2000 cases have
`been performed in the United States, expanding the total
`experience to more than 3000 patients. More than one
`half of the total implantations occurred recently.
`Training, as mandated by the FDA, has so far con-
`sisted of a 2-day course that includes didactic sessions,
`hands on deployment in flow models, and observation of
`cases in the operating room. Typically, a two-member
`team with satisfactory AAA surgical volume and interven-
`tional skills is trained at the same time. From October
`through May 1, 2000, Guidant Cardiac and Vascular
`Division has trained 828 physicians: 537 surgeons, 229
`radiologists, and 60 cardiologists. Clinical experience at all
`these new centers is difficult to evaluate. The technical
`success rate, however, as reported by clinical monitors is
`approximately 95%.
`
`SUMMARY
`The Ancure endograft was the first commercially pro-
`duced graft to undergo trials in the United States; FDA
`
`approval was granted in September of 1999. Some
`improvements in the attachment and delivery systems
`were necessary along the way, which resulted in excellent
`long-term results to date. This technology remains fairly
`novel, despite giant steps forward during the last decade.
`Further improvements and innovations are to be expected,
`delivering on the promise to make treatment of AAA a
`fairly effective and minimally invasive reality.
`
`I thank Maria Decker, MD, from Guidant Cardiac and
`Vascular Division for collecting data and providing the latest
`information available for this update.
`
`REFERENCES
`1. Makaroun M, Zajko A, Orons P, et al. The experience of an academic
`medical center with endovascular treatment of abdominal aortic
`aneurysms. Am J Surg 1998;176:198-202.
`2. Makaroun M, Zajko A, Sugimoto H, et al. Fate of endoleaks after
`endoluminal repair of abdominal aortic aneurysms with the EVT
`device. Eur J Vasc Endovasc Surg 1999;18:185-90.
`3. Amesur N, Zajko A, Orons P, et al. Embolotherapy of persistent
`endoleaks following endovascular repair of abdominal aortic aneurysm
`with the Ancure-Endovascular Technologies endograft. J Vasc Interv
`Radiol 1999;10:1175-82.
`4. Amesur NB, Zajko AB, Orons PD, et al. Endovascular treatment of
`iliac limb stenoses or occlusion in 31 patients treated with the Ancure
`endograft. J Vasc Interv Radiol 2000;11:421-8.
`5. Makaroun M, Zajko A. Endoleaks following endovascular aortic
`aneurysm repair: clinical significance and treatment modalities.
`Perspect Vasc Surg. In press.
`6. Matsumura JS, Chaikof EL. Continued expansion of aortic necks after
`endovascular repair of abdominal aortic aneurysms. J Vasc Surg
`1998;28:422-30.
`7. Zarins CK, White RA, Fogarty T. Aneurysm rupture after endovascu-
`lar repair using the AneuRx stent graft. J Vasc Surg 2000;31:960-70.
`
`Submitted Jun 12, 2000; accepted Aug 18, 2000.
`
`