Original Articles
`
`Endoluminal Aortic Aneurysm Repair Using
`a
`Balloon-Expandable Stent-Graft Device:
`Progress Report
`
`A
`
`Juan C. Parodi, MD, Frank J. Criado, MD, Hector D. Barone, PhD, Claudio Schénholz, MD,
`and Luis A. Queral, MD, Buenos Aires, Argentina, and Baltimore, Maryland
`
`Wedescribe our
`experience with endoluminal repair of abdominal aortic aneurysms using the
`stent-graft device. Twenty-four patients underwent 25 procedures in the 27-month period
`ending December 31, 1992. Twenty-one of the patients were considered high-risk candidates
`were aortoaortic in 16
`for conventional surgical
`repair. The endoluminal stented grafts
`procedures and unilateral aortoiliac in eight. One patient underwent a second procedure
`a
`separate commoniliac artery aneurysm. Technical
`consisting of anilioiliac graft to repair
`were
`primarily related to retrograde transluminal access acrossthe iliac arteries,
`problems
`as to measurement of length. One patient died and
`tortuous aneurysms, and misjudgments
`another required secondary deployment of a distal stent at 4 months; subsequent aneurysm
`expansion mandated surgical replacement at 18 months. It is clear that this device and
`methodologywill have to undergo further refinement before the technique is acceptable for
`wider clinical application. Current experience, however, is encouraging. Aneurysm exclusion
`important therapeutic alternative over
`with an endoluminal prosthesis is likely to become an
`the next several years. (Ann Vasc Surg 1994;8:523-529.)
`
`The graft inclusion technique for intra-aneurys-
`first described in
`mal prosthetic replacement,
`1962,'? has
`justifiably become
`the
`surgical
`method of choice over
`the past 25 years.
`Its
`simplicity and safety have contributed signifi-
`to the excellent results currently attained
`cantly
`treatment of aortic
`Al-
`with surgical
`aneurysms.’
`rates are
`quite low in
`though operative mortality
`good-risk candidates, there remains a subset of
`patients with serious
`multiple comorbid condi-
`tions for whom aneurysm surgery remainsa for-
`midable undertaking. Additionally, overall periop-
`
`From the Instituto Cardiovascular de Buenos Aires, Buenos
`Aires, Argentina, and the Maryland Vascular Institute, Balti-
`more, Md.
`Presented at the Eighteenth Annual Meeting of The Periph-
`eral Vascular Surgery Society, Washington, D.C., June 6, 1993.
`Reprint requests: Frank J. Criado, MD, Maryland Vascular
`Institute, 201 E. University Parkway, Suite 650, Baltimore,
`MD 21218.
`
`erative morbidity and costs continue to be major
`concerns for aneurysm patients facing surgical
`repair.
`a
`Aneurysm exclusion by
`implanted
`graft
`transluminally would achieve the same
`therapeu-
`as the inclusion technique but would
`tic goal
`avoid the risks and costs of conventional surgery.
`treatment of high-risk pa-
`It would also permit
`not tolerate a
`major abdominal
`tients who might
`operation and aortic cross-clamping. Conceived
`with this purpose in mind, the device consists of
`a
`within the
`stent-graft combination deployed
`aneurysm via retrograde transluminal introduc-
`tion.* Clinical application began in 1990, follow-
`ing preliminary bench testing and animal experi-
`mentation.’ This report focuses on the techniques
`used and results obtained with endoluminal
`treatment of abdominal aortic aneurysms.
`
`MATERIAL AND METHODS
`In the period beginning September 1, 1990
`(first
`clinical case), and ending December31, 1992, 24
`5,3
`MEDTRONIC 1019.
`
`MEDTRONIC 1019
`
`

`

`524 Parodiet al.
`
`were treated with the endoluminal
`patients
`method. Patient ages ranged from 66 to 83 years.
`There were 23 males and one female. At the time
`patients had a
`of intervention, two
`symptomatic
`aneurysm and 22 were
`asymptomatic. No rup-
`tured aneurysm was treated during the study
`were considered high
`period. Twenty-one patients
`risk by currently accepted criteria.*? Aneurysm
`ranged from 5.1 to 14.0 cm in greatest trans-
`size
`verse diameter. Aneurysms
`were all infrarenal.
`Preoperative evaluation included CT and aortog-
`in all cases.
`raphy
`Treatment consisted of transluminal retrograde
`insertion of a thin-walled Dacron
`graft sutured to
`a
`com-
`stent. The stent-graft
`balloon-expandable
`bination was mounted on a balloon catheter of
`appropriate size and assembled into an introducer
`sheath for transluminal delivery within the aneu-
`in terms of balloon
`rysm (Fig. 1). The device,
`was tailored to the
`diameter and graft length,
`characteristics and size of the lesion in each
`individual patient. Cranial and caudal balloon-
`
`
`
`aSESS
`NS
`
`eeeWUT
`
`
`
`TTT |
`
`Of,
`Lavery
`
`B
`
`Fig. 1. The thin-walled Dacron graft is sutured to the
`stent as shown. The stent-graft combination is
`then
`mounted on a balloon catheter and the graft
`folded
`around the stent and the catheter. Introduction into the
`delivery sheath completes preparation of the device for
`endoluminal deployment.
`
`Annals of
`Vascular Surgery
`
`expandable stainless steel stents were
`to
`deployed
`A
`in its intra-aortic position.
`attach the prosthesis
`distal stent was not used in the first four proce-
`dures of this series. Access was
`by femoral cut-
`in the common femoral
`down and arteriotomy
`artery (Fig. 2). Aneurysms involving the aortic
`arteries were
`bifurcation and common iliac
`treated by insertion of a
`tapered aortoiliac unilat-
`a
`eral graft and contralateral iliac exclusion using
`or
`Re-
`transcatheter technique
`surgical ligation.
`vascularization of the opposite lower extremity
`a crossover femorofemoral bypass.
`A
`required
`procedures involved a
`aor-
`total of 16
`straight
`cases an aortoiliac
`toaortic graft (Fig. 3); in eight
`was inserted,
`seven of
`(unilateral) prosthesis
`a concomitant femorofemoral by-
`which required
`pass (Fig. 4). Chronic iliac occlusion in one case
`obviated the need for crossover
`One
`grafting.
`patient underwent aortoaortic grafting first, and
`was
`then a similar ilioiliac graft
`2
`days
`placed
`aneu-
`later for a
`common iliac artery
`separate
`rysm, makinga total of 25 endoluminal grafting
`procedures in 24
`patients.
`were all performed underfluo-
`The procedures
`roscopic, arteriographic guidance. Intraoperative
`measurements of aortic and aneurysm diameter
`by the use of a
`were aided
`and length
`specially
`designed arteriographic catheter containing radi-
`at known intervals. Computer
`opaque markings
`
` Fig. 2. Exposure of the common femoral artery by surgical
`
`across
`cutdown. A
`guidewire is introduced transluminally
`at the wire intro-
`the aneurysm. The arteriotomy begins
`duction point (A), The ensheathed assembled device is
`inserted over the wire (B).
`
`

`

`Vol. 8, No. 6
`1994
`
`Balloon-expandable stent-graft device for AAA repair
`
`525
`
` Fig. 3. Completed aortoaortic endoluminalgraft.
`
` 1
`
`}
`Fig. 4. Tapered endoluminal!aortoiliac unilateral graft. Crossover grafting and contralateraliliac
`or
`exclusion, either transluminally
`by surgical ligation (inset), complete the procedure.
`
`

`

`526
`
`Parodietal.
`
`assistance was utilized in these calculations. Epi-
`dural anesthesia supplemented with intravenous
`sedation was
`in 23 procedures, and
`employed
`local anesthesia in two.
`Average length of hospi-
`was 3 days. Linear
`tal stay
`in each
`follow-up
`patient included clinical reassessment, color-flow
`at 6-month intervals.
`duplex, and CT
`scanning
`was
`at least once
`Aortography
`performed
`the follow-up period.
`
`during
`
`RESULTS
`
`One death occurred in the series; this was in a
`man with an 8 cm abdominal aortic
`72-year-old
`aneurysm that was very tortuous. Technical dif-
`ficulties with proper positioning and deployment
`of the cranial stent led to
`transluminal
`multiple
`reentries that caused distal embolization of aneu-
`to severe dis-
`rysm contents, leading ultimately
`seminated intravascular coagulation and fatal in-
`tracerebral hemorrhage.
`Intraprocedural complications have generally
`as
`to the length and
`involved misjudgments
`three-dimensional configuration of the aneurysm
`operations there was
`and its upper neck. In two
`inadequate proximal bloodtight sealing because
`or
`of the
`of incomplete
`improper deployment
`in an
`antegrade “‘leak”’ into
`upper stent, resulting
`insertion of a
`the aneurysm sac. This
`required
`stent within the proximal
`second graft-covered
`stent in one case; in the other, luminal reentry
`was left with a
`proved impossible and the patient
`pulsatile aneurysm. Conventional surgical repair
`has not been undertaken because of prohibitive
`associated medical risks. As of the time of this
`the aneurysm remains
`writing (August 1993),
`pulsatile but has not
`enlarged appreciably. The
`other type of intraoperative complication involved
`one
`patient with a
`retrograde “leak” at the distal
`end caused by failure of the stent to
`completely
`seal the graft-aorta interface. Correction was also
`by deployment of a second cloth-covered stent.
`Postoperative complications have occurred in
`two
`patients. Secondary reflux from the distal,
`unstented end of the graft with recurrent, pal-
`pable pulsatility of the aneurysm was discovered
`in one
`4 months following the initial
`patient
`intervention. Secondary deployment of a distal
`stent seemed to correct the situation. However,
`at 18 months with an
`the patient presented again
`con-
`expanding pulsatile aneurysm that required
`ventional surgical repair. The cause of the prob-
`lem had beenthe insertion of a
`graft that was too
`short to
`begin with and secondary deployment of
`a distal stent that did not attain complete exclu-
`
`Annals of
`Vascular Surgery
`
`sion of the aneurysm with consequent continued
`In one other pa-
`expansion in its distal portion.
`tient a
`large groin wound hematoma developed
`that resolved with conservative management. All
`of the patients continue to be followedserially,
`as to the signs of pulsatility
`or aneu-
`primarily
`rysm expansion, endoluminal graft dilatation,
`and signs of intra-aneurysm “leaks.”
`
`DISCUSSION
`
`Endovascular grafting for aneurysmal exclusion
`a
`would have the potential of avoiding
`major
`abdominal operation involving significant blood
`loss and aortic
`cross-clamping. Furthermore, the
`risks of nerve
`injury causing sexual dysfunction
`and the occurrence of graft-enteric complications
`It was first envi-
`would be obviated altogether.
`one of us
`in 1977; various
`sioned by
`(J.C.P.)
`attachment methods and devices were conceived
`It was the
`but none
`proved satisfactory.
`develop-
`ment of balloon-expandable metallic stents that
`the critical component for the endovas-
`provided
`cular repair technique.
`Several designs and inventions have been pat-
`ented.*** The first in vivo demonstration of the
`was
`endoluminal methodology
`reported by Balko
`et al.” who excluded experimental aneurysms in
`a
`polyurethane graft with a Nitinol or
`sheep using
`stainless steel frame. These acute
`experiments
`proved the feasibility and soundness of the con-
`cept. More recent
`experiments in
`dogs by Mirich
`et al.’° involved the implantation of wire stents
`a porous nylon graft.
`covered by
`The design of an endoluminalrepair device and
`must
`take into consideration the
`technique
`unique anatomic and pathologic characteristics of
`abdominal aortic aneurysms. More to the point,
`thrombus and at
`the presence of intraluminal
`times marked tortuosity and elongation of the
`aorta may pose difficulties and risks when using
`an intraluminal method. Precise three-dimen-
`sional sizing of the aneurysm and its neck is
`paramountin tailoring the appropriate device and
`implantation technique in a
`case. This point
`given
`emphatically underlined in our
`was
`experience by
`the one
`fatality resulting from misjudging the
`length of the aneurysm andits neck, whichled to
`reentries and exchanges that
`dislodged
`multiple
`the thrombus and caused downstream emboliza-
`tion. Current and future advances in CT and MR
`to becomecritical compo-
`arteriography are likely
`nents for proper and safe execution of endolumi-
`nal repair techniques. Intravascular ultrasound
`a role in the future.
`imaging may also play
`
`

`

`Vol. 8, No. 6
`1994
`
` Fig. 5. Iliac pull-down
`
`maneuver
`by extensive circumfer-
`ential dissection of common femoral and external iliac
`arteries.
`
`Balloon-expandable stent-graft device for AAA repair
`
`527
`
`Retrograde transluminal access
`through the
`femoral-iliac arterial system is a critical step and
`a
`limitation. Current
`technology has
`frequent
`permitted reduction of
`the entire ensheathed
`stent-graft device to a minimum diameter of 18 F.
`This still requires iliac arteries of considerable
`caliber and free of significant atherosclerosis. Fo-
`cal stenotic lesions, when present, should un-
`in preparation for the perfor-
`dergo angioplasty
`in such a
`way as
`mance of the procedure, ideally
`to allow enough time (2 weeks or
`for
`more)
`healing of the angioplasty site. The rigidity
`of the
`a
`current device represents
`problem when
`dealing
`with tortuous iliac arteries. It has been found that
`can be overcome and the artery
`iliac tortuosity
`it
`is
`essentially free of
`straightened, provided
`atherosclerosis. The “iliac pull-down maneuver’
`(Fig. 5) has been a
`simple and effective aid in
`several cases. In other situations
`more
`involving
`severe
`angulations and iliac arterial disease, we
`have resorted to another solution consisting of
`the attachment of a
`to
`temporary prosthetic graft
`the commoniliac artery exposed through
`a lim-
`ited retroperitoneal incision. This graft
`is then
`passed down under the inguinal ligament and
`used as the conduit for introduction and deploy-
`ment of the endoluminal device (Fig. 6). Percu-
`
` Fig. 6. Retroperitoneal exposure of the commoniliac artery and anastomosis of a 10 mm Dacron
`
`tube used for transluminal delivery of the device. Following implantation the Dacroniliac graft
`is cut and oversewn almost flush with theiliac artery.
`
`

`

`528
`
`Parodi et al.
`
`in the near future
`taneous access is not
`likely
`with a
`balloon-expandable device of this type.
`Spring-loaded prostheses would have an advan-
`In anycase, it is our view that
`tage in this
`regard.
`percutaneous insertion per se does not constitute
`a
`critically important goal of these developments.
`is the avoidance of a
`The focus, by definition,
`major transabdominal operation and aortic cross-
`clamping.
`re-
`Oneother critical aspect of the technique
`andits
`lates to the nature of the graft required
`to the aortic wall. Knitted thin-
`attachment
`walled Dacron tubes were used
`in this
`exclusively
`experience. Although bench testing and tensile
`measurements have been reassuring,
`it
`strength
`remains to be seen whether progressive graft
`dilatation will be a concern. In
`theory, such would
`be unlikely and probably inconsequential because
`of the intra-aortic endoluminal position of the
`prosthesis.
`The integrity of the graft attachment with
`stents has not been a
`properly placed
`problem,
`or
`We
`either acutely
`during the follow-up period.
`remain confident that long-term graft anchoring
`is adequate. Migra-
`afforded by this technique
`tions have been related to
`improper positioning
`of the stents, highlighting again
`and
`deployment
`the importanceof pre- and intraoperative imaging
`and measurements of the aneurysm andits neck.
`A minimum
`length of upper neck is required for
`proper stenting and anchoring; this is
`in
`probably
`the range of 2.5 to 3.0 cm.
`Precise visualization of the renal arteries dur-
`is paramount. More often than
`ing implantation
`not
`the cranial stent
`is deployed immediately
`below therenal artery ostia. In three patients the
`superior, uncovered portion of the stent was im-
`across one or both renal artery ostia
`planted
`without any apparent compromise of renal blood
`or
`the
`flow either acutely
`follow-up period.
`during
`in
`in
`This is certainly
`keeping with experience
`other arterial segments where the Palmaz stent
`across an
`can be expanded
`important branch
`orifice (internal iliac) without occluding flow.
`Unlike the proximal ‘‘anchoring” stent,
`to achieve a
`distal stent is
`bloodtight
`designed
`seal of the space between the (inner) graft and
`the (outer) aortic wall, thus preventing retrograde
`flow outside the graft and continued aneurysm
`expansion. A minimum length of nonaneurysmal
`2 cm is
`to
`aorta of approximately
`safely
`required
`the stent. Unlike the surgical scenario,
`“seat’’
`to the bifur-
`where one can suture the tube graft
`cation itself or even the iliac artery ostia,
`the
`endoluminal method requires clear-cut sparing of
`
`the
`
`Annals of
`Vascular Surgery
`
`the bifurcation and distalmost aorta from aneu-
`rysmal involvement. If this is not the case, and
`when the aneurysm is frankly aortoiliac, we have
`resorted to direct deployment distally into one
`a
`tapered prosthesis and a
`iliac artery using
`stent. The contralateraliliac
`smaller distal sealing
`artery must be excluded via transcatheter balloon
`or
`surgical ligation of the external
`deployment
`and internaliliac arteries. Crossover femorofemo-
`A bifurcated
`ral grafting completes the procedure.
`endoluminal graft would be the best approach,
`in an
`is
`early devel-
`but the technique
`currently
`opmental stage.
`The fate of the inferior mesenteric and lumbar
`arteries following endoluminalgraft implantation
`is not clear at this time. Continued patency of
`these vessels might prevent complete thrombosis
`of the periprosthetic intra-aneurysmal lumen and
`in blood flow to some
`degree. This is
`result
`as to cause
`to be of sufficient magnitude
`unlikely
`and further expansion, but
`aneurysm pulsatility
`the issue may require specific clarification in the
`future.
`Several other methods of endoluminal aneu-
`are
`being developed throughout the
`rysm repair
`world. Variations of the stent-graft concept in-
`clude self-expanding (spring-loaded) devices and
`the use of alternate attachment methods with
`and without stents. Stents to encompass the
`length of the aneurysm (cloth-covered,
`tightly
`are also being
`knit, and other modifications)
`tried. It is clear that further development of the
`device itself and the techniques of deployment
`will be required before widespread clinical appli-
`cation is warranted. Specifically, miniaturization,
`as well as
`improved trackability and
`flexibility,
`stent
`will
`better methods of bipolar
`deployment
`be forthcoming and contribute to
`simplification,
`increased safety, and wider applicability of the
`technique. Precise anatomic characterization of
`the aneurysm and
`nonaneurysmal aorta, espe-
`at the upper neck, will also be required and
`cially
`to further refinements of
`contribute enormously
`Endovascular grafting with
`this methodology.
`this type of stent-graft combination has also been
`to obliterate arteriovenous fis-
`used successfully
`a translumi-
`tulas and acute dissections through
`nal approach (Fig. 7).
`to
`Althoughitis still too
`predict what role
`early
`these techniques will have in the managementof
`in the future and what num-
`aneurysm patients
`it
`ber of patients will be suitable candidates,
`appears safe to state that this and other endolu-
`minal methods will likely prove useful in some
`“Opposition will be mounted”’* but as in
`cases.
`
`

`

`Vol. 8, No. 6
`1994
`
`Balloon-expandable stent-graft device for AAA repair
`
`529
`
` Fig. 7. Cloth-covered stent used to obliterate arteriove-
`
`nous fistulas and other lesions.
`
`other areas of surgery,
`the march toward less
`is inevitable. Vascular sur-
`invasive
`techniques
`geons would be ill advised not to follow these
`developments with curiosity and receptivity, for
`to
`change the treatment
`they have the potential
`of aneurysms and other vascular lesions dramati-
`in the twenty-first century.
`cally
`
`CONCLUSION
`
`Aneurysm repair by endoaortic transluminal
`is a
`grafting is
`clearly feasible. Potentially it
`promising alternative for high-risk patients who
`would experience significant morbidity and mor-
`tality with conventional surgical repair. Further
`are needed with the device itself
`developments
`In
`and the techniques of deployment.
`particular,
`trackability and miniaturization of the device,
`transluminal access
`through the iliac arteries,
`intra-aortic attachment, and precise sizing and
`anatomic characterization of aneurysms are areas
`that should be the focus of current and future
`to have a
`are
`research. These techniques
`likely
`on the way we
`manage aneurysm
`major impact
`patients.
`
`REFERENCES
`1. Javid H, Julian OC, Dye WS,et al. Complications of abdomi-
`nal aortic grafts. Arch Surg 1962;85:142.
`2. Creech O. Endoaneurysmorrhaphy and treatment of aortic
`aneurysm. Ann Surg 1966;164:935.
`3. Hollier LH, Taylor LM, Ochsner J. Recommendedindications
`for operative treatment of abdominal aortic aneurysms.
`J Vasc Surg 1992;15:1046.
`4. Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal
`graft implantation for abdominal aortic aneurysms. Ann Vasc
`Surg 1991;5:491.
`5. Yeager RA. Basic data related to cardiac testing and cardiac
`risk associated with vascular surgery. Ann Vasc Surg 1990;
`4:193.
`6. Choudhury MH. Method for performing aneurysm repair.
`U.S. Patent No. 4, 140, 126, 1979.
`an intralu-
`7. Kornberg E. Device and method for performing
`minal abdominal aortic aneurysm repair. U.S. Patent No. 4,
`617, 932, 1986.
`8. Lazarus HM.Intraluminal graft device: System and method.
`U.S. Patent No. 4, 787, 899, 1988.
`9. Balko A, Piasecki GJ, Shah DM,et al. Transfemoral place-
`mentof intraluminal polyurethane prosthesis for abdominal
`aneurysm. J Surg Res 1986;40:305.
`10. Mirich D, Wright KC, Wallace S, et al. Percutaneously placed
`endovascular grafts for aortic aneurysms: Feasibility study.
`Radiology 1989;170:1033.
`11, Bergan JJ. Commentary. In Parodi JC, Palmaz JC, Barone HD.
`Transfemoral intraluminal graft implantation for abdominal
`aortic aneurysms. Ann Vasc Surg 1991;5:491.
`
`