ADVANCES IN
`
`Vascular Surgery
`
`
`VOLUME1
`
`Editor-in Chief
`Anthony D. Whittemore, M.D.
`Associate Professor Surgery, Harvard University Medical School; Chief,
`Division of Vascular Surgery, Brigham and Women’s Hospital, Boston,
`Massachusetts
`
`Associate Editors
`M.D.
`Dennis F. Bandyk,
`Professor of Surgery, University of South Florida College of Medicine; Director,
`Vascular Surgery Division, Chief of Vascular Surgery, Tampa General Hospital,
`Tampa, Florida
`L. Cronenwett, M.D.
`Jack
`Professor of Surgery, Dartmouth College, Mary Hitchcock Medical Center,
`Hanover, New Hampshire
`NormanR.Hertzer, M.D.
`Chairman, Department of Vascular Surgery, Cleveland Clinic Foundation,
`Cleveland, Ohio
`A. White, M.D.
`Rodney
`Professor of Surgery, University of California at Los Angeles School of
`Medicine; Chief of Vascular Surgery, Harbor—University of California at Los
`Angeles Medical Center, Los Angeles, California
`
`NA Mosby
`St.Louis Baltimore Boston Chicago London Madrid Philadelphia Sydney Toronto
`
`MEDTRONIC 1013
`
`MEDTRONIC 1013
`
`

`

`N'A Mosby
`Dedicated to Publishing Excellence
`
`Sponsoring Editor: Linda M.Steiner
`Project Manager: Denise Dungey
`Project Supervisor: Maria Nevinger
`Production Editor: Laura Pelehach
`Staff Support Administrator: Barbara M. Kelly
`
`Copyright © 1993 by Mosby— Year Book,Inc.
`
`Mosby-Year Book,Inc.
`11830 Westline Industrial Drive
`St. Louis, MO 63146
`
`All rights reserved. No part of this publication may be reproduced, stored in a re-
`or transmitted, in any form or
`by any means, electronic, mechanical,
`trieval system,
`or otherwise, without prior written permission from the
`photocopying, recording,
`publisher. Printed in the United States of America.
`use is permit-
`reproducesolely for internal or
`or
`Permission to
`personal
`photocopy
`ted for libraries or other users
`registered with the Copyright Clearance Center, pro-
`vided that the base fee of $4.00 per chapter plus $.10 per page is paid directly to
`the Copyright Clearance Center, 21 Congress Street, Salem, MA 01970. This con-
`sent does not extend to other kinds of copying, such as
`copying for general distri-
`new collected works,
`or
`bution, for advertising
`promotional purposes, for creating
`or for resale.
`
`Editorial Office:
`Mosby- YearBook,Inc.
`200 North LaSalle St.
`Chicago, IL 60601
`
`International Standard Serial Number: 1069-7292
`International Standard Book Number: 0-8151-9405-6
`
`

`

`Contents
`
`Contributors
`
`Preface
`
`PARTI. Carotid Artery
`Historical Perspective of Carotid Artery Disease.
`By Jessc E.
`Thompson
`
`Disease
`
`Symptomatic Carotid Artery Disease
`D.
`Eugene Standness, Jr.
`By
`Prospective Studies
`Trial Design
`Outcome
`
`Discussion
`
`Treatment of Asymptomatic Carotid Stenosis.
`By Robert W. Hobson II
`Defining Terminology
`Considerations
`
`Surgical
`Clinical Trials
`
`Current Recommendations
`
`~
`
`Diagnostic Evaluation of Carotid Artery Disease.
`By R. Eugene Zierler
`Noninvasive Diagnostic Methods
`Indirect Tests
`Direct Tests
`Radiologic Diagnostic Methods
`Conventional Contrast
`Arteriography
`Computerized Cerebral Tomography
`Resonance
`Magnetic
`Angiography
`Clinical Applications
`
`PART Ii. Aortic Disease
`Retroperitoneal Approach for Aortic Reconstruction.
`By Calvin B. Ernst and Alexander D. Shepard
`Advantages and Disadvantages
`Indications and Contraindications
`
`vii
`
`17
`
`19
`
`20
`
`20
`
`21
`
`25
`
`26
`
`27
`
`29
`
`35
`
`39
`
`39
`39
`4]
`
`47
`47
`48
`48
`
`50
`
`59
`
`6]
`
`62
`
`62
`
`ix
`
`

`

`X
`
`Contents
`
`Preoperative Evaluation
`Operative Technique
`Positioning
`Incision
`Aortic Exposure
`Distal Exposure
`Placement
`Aortic Clamp
`Wound Closure
`Results
`
`Summary
`
`Reducing the Risk of Spinal Cord Injury During
`Thoracoabdominal Aortic Aneurysm Repair.
`By Thomas C. Naslund and Larry H. Hollier
`Spinal Cord Blood Supply
`of Ischemic Spinal Cord Injury
`Pathophysiology
`Clinical Manifestations of Spinal Cord Ischemia
`Spinal Cord Protection During Operation
`Care
`Postoperative
`Clinical Results
`
`Summary
`
`PART Ill. Endovascular Techniques
`Endovascular Repair of Abdominal Aortic Aneurysms.
`By Juan C. Parodi
`Laboratory Experience
`Clinical Experience
`Device Components
`Evaluation
`Preprocedural
`Procedural Technique
`Patient Selection
`
`Complications
`Late Results
`
`Technical Considerations
`Arterial Access
`Distal Graft Deployment
`Patent Branch Vessels
`Secondary Endovascular Procedures
`Miscellaneous Complications
`The Future of Endovascular Grafting
`
`Intravascular Stents.
`By Julio C. Palmaz, Frank J. Rivera, and Carlos Encarnacion
`Mechanical Considerations
`Hemodynamic Considerations
`
`

`

`Contents
`
`xi
`
`Anticoagulation and Stent Thrombogenicity
`Relationship Between Slow Flow and Intimal Hyperplasia
`Coated Stents
`Intraluminal Bypass
`Current Clinical Experience With Intravascular Stents
`Iliac Arteries
`
`Arteries
`
`Femoropopliteal
`Renal Arteries
`Large Veins
`Transjugular, Intrahepatic, Portosystemic Stent Shunt
`
`Intraluminal Ultrasound and the Management
`Vascular Disease.
`of Peripheral
`A. White
`M. Cavaye and Rodney
`By Douglas
`Devices
`Intraluminal Ultrasound Imaging
`Intravascular, Cross-Sectional Imaging
`Three-Dimensional Computerized Image Reconstruction
`Forward-Looking Intraluminal Ultrasound
`Intraluminal Ultrasound Imaging Techniques
`Vascular Access
`
`Optimizing Image Quality
`Image Interpretation
`Clinical Applications of Intraluminal Ultrasound
`Diagnostic Uses
`Uses
`Therapeutic
`Current Limitations and Future Directions
`
`The Role of Angioscopy in Infrainguinal Arterial
`Reconstructions.
`By William H. Pearce, Kevin D. Nolan, and James S.T. Yao
`History
`
`Equipment
`
`Angioscopy-Assisted
`Technique
`
`in Situ Bypass
`
`Angioscopically
`Technique
`as a
`
`Angioscopy
`Comments
`
`Assisted Thromboembolectomy
`
`Technique
`
`to Detect Technical Errors
`
`Disease
`PART IV. Renal Artery
`Changing Concepts in Indications for Renal Artery
`Reconstruction.
`By Richard H. Dean
`Current Concepts in
`Diagnostic Studies
`Current Indications for Intervention
`
`113
`
`114
`
`114
`
`115
`
`116
`116
`12]
`124
`125
`126
`
`137
`
`137
`139
`140
`14]
`
`141
`14]
`143
`144
`
`144
`144
`149
`
`156
`
`159
`
`159
`
`160
`
`160
`162
`
`163
`164
`
`166
`
`167
`
`169
`
`171
`
`172
`
`174
`
`

`

`xi
`
`Contents
`
`Changing Concepts in Techniques of Renal
`Revascularization.
`By Richard H. Dean
`Contemporary Technical Considerations
`Route of Incision
`Midline Incision
`Flank Incision
`Exposure Considerations
`Aortorenal Bypass
`Thromboendarterectomy
`Extra-Anatomic Bypasses
`Ex Vivo Reconstructions
`
`PART V. Status of Thrombolytic Therapy
`Thrombolytic Therapy for Arterial Disease.
`By Michael Belkin
`The Fibrinolytic System
`Fibrinogen/Fibrin
`Plasminogen/Plasmin
`Plasmin Inhibitors
`Fibrinolytic Agents
`Plasminogen Activator Inhibitors
`Intra-Arterial Infusion Techniques
`Indications and Contraindications
`
`Results
`Short-Term Results
`Long-Term Results
`Acute Limb Ischemia
`
`__
`
`Complications
`Bleeding Complications
`Intra-Arterial Complications
`Summary and Recommendations
`Symptomatic Arterial Occlusion
`Bypass Graft Occlusion
`
`;
`
`181
`
`182
`
`182
`182
`182
`183
`184
`
`184
`185
`
`185
`
`189
`
`191
`191
`192
`192
`193
`193
`195
`
`196
`
`198
`
`200
`201
`201
`205
`205
`205
`206
`207
`207
`208
`
`Thrombosis.
`
`Appropriate Therapy for Acute Lower-Extremity Venous
`213
`By Anthony J. Comerota and Samuel C. Aldridge
`214
`Thrombolytic Therapy for Acute Deep Venous Thrombosis
`For Acute Deep Venous
`Failure of Thrombolytic Therapy
`Thrombosis
`Calf Vein Thrombosis
`lliofemoral Venous Thrombosis
`Summary
`
`216
`218
`219
`222
`
`

`

`PART VI. Issues in Basic Science
`Hypercoagulable States in Patients with Peripheral
`Vascular Disease.
`By Magruder C. Donaldson
`Hypercoagulable States
`Congenital Hypercoagulable
`Physiologic Acquired Hypercoagulability
`Pathologic Acquired Hypercoagulability
`in Vascular Disease
`Hypercoagulability
`Overview of Thrombosis in Vascular Disease
`in Patients with Vascular Disorders
`
`States
`
`Hypercoagulability
`of
`
`Contents
`
`Xi
`
`225
`
`227
`
`227
`229
`229
`230
`
`231
`231
`
`232
`
`235
`235
`237
`
`239
`
`243
`
`Hypercoagulability
`
`Management
`Diagnosis
`Therapy
`Conclusions
`
`Index
`
`

`

`Endovascular Repair of
`Abdominal Aortic
`
`Aneurysms
`JUAN C. PARODI, M.D.
`
`
`his chapter describes the development of a form of endovascular
`
`T
`treatment for abdominal aortic aneurysms (AAAs) and other arterial
`our
`diseases. My associates and I
`hope that this information concerning
`early animal experimentation and our
`subsequentseries of patients will
`to all clinicians who
`be of value and, in some way, will be provocative
`responsible for the diagnosis and management of AAAs. Wealso are
`are
`experience to fellow investigators who are
`to lend our
`engaged
`pleased
`hope that it helps them to avoid a few of the
`in similar research, and we
`mistakes that we
`initially made.
`The diagnosis of AAA hasbeenestablished with increasing frequency
`during the past 2 decades. This observation probably is related to aging
`as well as to the extensive use of ultrasonography and
`of the population,
`computerized tomographic (CT) scanning for screening purposes. Al-
`though AAAsoccasionally may cause distal embolization to the lower
`extremities, rupture remains their most common and deadly complica-
`tion. It has been established that fewer than 30%ofall patients who sus-
`tain ruptured aneurysmssurvive this catastrophe,a figure that includes a
`mortality rate of about 50%, even among
`those wholive long enoughto
`receive urgent surgical intervention.
`to be the most
`Elective replacement with a
`synthetic graft has proved
`40 years and,
`appropriate method of preventing AAA rupture for nearly
`respected medicalcenters, it has been associated with a
`at
`postoperative
`mortality rate of less than 5%, even in high-risk patients.” Nevertheless,
`at which thereis less ex-
`surgical risk undoubtedlyis higher in hospitals
`occur
`perience with aortic reconstruction, and nonfatal complications
`is
`regularity irrespective of the setting in which the operation
`with some
`performed. Finally, the decision to recommendelective surgical manage-
`an infinite numberof considerations regarding the
`mentis influenced by
`likelihood of rupture of untreated AAAsrelatedto their size, as well as
`to the operative risk in patients who haveincidental cardiac, pulmonary,
`or renal disease.* It seems inevitable that every vascular surgeon occa-
`a
`prohibitive risk for con-
`sionally will encounter a
`patient who represents
`as
`ventional graft replacement, yet alternative forms of treatment (such
`in conjunction with induced AAA thrombosis) gen-
`axillofemoral bypass
`Advancesin Vascular Surgery,vol. 1.
`©1993, Mosby— Year Book,Inc.
`
`85
`
`

`

`86
`
`J.C. Parodi
`
`erally have been abandoned, despite preliminary reports of their initial
`success.*
`In 1976, we
`a
`to
`plan for the endovascular treatment
`develop
`began
`of AAAs that was based upon the fundamental principles of aortic re-
`placement. Ourfirst two prototypes may be described as follows:
`1. A metallic self-expandable mesh with a
`cov-
`“zig zag” configuration,
`a thin fabric graft that was
`ered by
`compressed and introduced in-
`side a sheath. This prototype
`was
`employed experimentally in nor-
`mal canine aortas, but we foundthat it was difficult to
`deploy be-
`cause the elastic properties of the metallic cage were difficult to
`standardize. If too much radial pressure were
`applied, damageto the
`aortic wall eventually could cause its rupture. Conversely, the appli-
`cation of too little pressure permitted subsequent migration of the
`graft, with leakage of blood betweenit and the host aorta.
`A Silastic bag with a
`cylindric lumen, fitted within a Dacron
`graft.
`was introduced into simulated AAAsin
`dogs, and then was
`This bag
`distendedbythe injection of silicone into the bag. Unfortunately, this
`method was associated with prompt thrombosisof the aorta in all ex-
`perimental animals.
`We
`eventually abandoned both of these prototypes because of their
`complications, but we reinitiated our
`project when balloon-expandable
`stents becameavailable in 1988. Our current approachis predicated upon
`place of sutures to secure the prox-
`the conceptthat stents may be used in
`imal and distal ends of a fabric graft extending the length of the AAA
`to fulfill the necessary re-
`(Fig 1). Balloon-expandable stents appeared
`
` FIGURE 1.
`
`Schematic illustration of the endovascular treatment of abdominal aortic aneu-
`rysms.
`
`

`

`Endovascular Repair of Abdominal Aortic Aneurysms
`
`87
`
`ana,
`POTTINTTLITTTTee
`
`\
`
`yyy
`WA
`VAAL
`
`AULA
`LALA
`
`
` —
`
`
`
`
`
`FIGURE 2.
`The elements comprising
`
`our endovascular device.
`
`quirements for this purpose becausetheir positioning could be calculated
`was
`precisely, the radial force necessary for their deployment
`predictable,
`and their diameter could be selected to matchthat of the aortaitself.
`The sheath that we
`employ currently contains the following elements
`(Fig 2):
`or double-balloon catheter of appropriate diameter.
`1. A single-
`2. Mounted overthe catheter, a
`stent similar to that
`balloon-expandable
`& Johnson, Warren, NJ), but adapted
`introduced by Palmaz (Johnson
`for larger diameters.
`3. A knitted, crimped, thin-walled synthetic graft made of Dacron yarn
`in strength to other commercialgrafts that already
`that is comparable
`are available.
`Oncethegraft has been introduced transfemorally to the appropriate
`level of the aorta, either the proximal balloon alone or both the proximal
`in
`and the distal balloons are inflated in orderto fix the graft firmly
`place.
`If our method proves to be safe and durable, it would avoid the compli-
`cations of general anesthesia, intra-abdominal surgical dissection, blood
`transfusion, prolonged aortic clamping, and sexual dysfunction related
`to disturbance of the pelvic autonomic nervesthat may
`beassociated with
`treatment.
`traditional surgical
`
`

`

`88
`
`J.C. Parodi
`
`LABORATORY EXPERIENCE
`weper-
`In order to determine the feasibility of our
`proposed method,
`formed a number of animal experiments that have been reported else-
`©
`a
`In summary, we constructed anartificial AAA
`where.®
`by resecting
`it with a fusiform Dacron
`segmentof the infrarenal aorta and replacing
`graft (Fig 3). This model reproduced the important features of human
`AAAs; that is, endothelium was foundto coverthe surface of the graft
`at
`the “neck” of the experimental aneurysm, and mural thrombus occupied
`were
`much of the cavity of the dilated segment. Both of these changes
`encountered regularly within a month after the experimental
`AAA had
`been implanted.
`were
`After 4 weeks, these dogs again
`placed under general anesthe-
`was ad-
`was
`sia and the common femoral artery
`exposed. A guidewire
`a
`vanced into the suprarenal aorta, and a 14-French sheath containing
`Palmaziliac stent and a Dacron graft mounted on a balloon catheter was
`overthe wire. After the sheath had been removed, the endovascu-
`placed
`AAA andthen wasse-
`was
`positioned within the experimental
`lar graft
`4 and 5). In one group
`cured by deploying either one or two stents (Figs
`we used a second
`a
`we used only
`proximalstent, and in another,
`of dogs,
`stent at the distal end of the exclusion graft.
`in most cases to evaluate different
`ona total of 43
`We
`dogs,
`operated
`
`
`
`
`
`FIGURE 3.
`An exampleof the “artificial aneurysm” constructed using
`our animalstudies.
`
`a Dacron conduit in
`
`

`

`
`
`FIGURE 4.
`A representative aortogram 4 weeksafter the construction of an “artificial aneu-
`rysm”in a
`dog.
`
`
`
`
`
`
`FIGURE5.
`Anotheraortogram following the implantation of a stented aortic graft through
`the right femoral artery in the same dogillustrated in Figure 4.
`
`

`

`90
`
`J.C. Parodi
`
`variety of balloons and re-
`models for experimental aneurysm orto test a
`we
`In our last group of dogs,
`performed 6-month eval-
`placementgrafts.
`scans in some cases
`uations, obtaining color-Doppler ultrasound (duplex)
`and complete pathologic studies in all animals. Using both optical and
`were
`we foundthat both endsof the stented grafts
`electronic microscopy,
`covered by endothelium, and that the shaft of the graft that was in con-
`tact with the mural thrombusof the experimental AAA was covered by
`a
`fibrin-platelet barrier.
`Once weweresatisfied that a Dacron
`graft could be delivered through
`place by balloon-expandable stents, we ob-
`a catheter andfixed firmly
`in
`a
`tained the permission of ourinstitutional ethical committee to
`perform
`pilot clinical study with a small group of patients who were informed
`were followed
`fully regarding the nature of our work and subsequently
`very carefully.”
`
`CLINICAL EXPERIENCE
`
`DEVICE COMPONENTS
`The device used in ourclinical workis illustrated in Figure 2. A super-
`stiff guidewire with a diameter of 0.038 in. (Amplatz, Meditech, Boston)
`tortuousiliac arteries, keeps
`facilitates introduction of the device through
`to that of the aorta, and, therefore, tends to
`the axis of the device parallel
`prevent disruption of laminated thrombus contained within the AAA.
`stentis a
`cylindric tube with longitudinalslots
`The balloon-expandable
`a diamond shape when expanded. This design, initially de-
`that adopt
`°
`expand from a diameterof only
`permits the stent to
`scribed by Palmaz,®
`5 mm whencollapsed
`to a diameter of more than 30 mm when expanded.
`Thestent is 3.5 mm in length, and its metallic componentrepresents only
`20% of its surface area after deployment.
`graft has a wall thickness of 0.2 mm andis fab-
`The knitted Dacron
`ricated using strong yarns with tensile and bursting strengths compara-
`ble to those of commercially available grafts. The compliance of the shaft
`of the graft (15%) also is similar to standard, knitted Dacron
`grafts, but
`the compliance of the segment overlapping the stent is 45% in order to
`a transitional zone
`to its expansion. Between these two segments,
`adapt
`in an
`attempt to avoid abrupt
`is interposed
`of intermediate compliance
`in the diameterof the graft that eventually could produce dele-
`changes
`terious friction between the stent and the shaftof the graft itself. The di-
`is tailored to fit the individual patient,
`ameter and length of each graft
`most often are 18 to 20 mm in diameter
`we have employed
`but the grafts
`and 8 to 12 cm in
`length.
`a thin gold wire is knitted to each graft.
`Because of its radiopacity,
`This permits the radiographicidentification of both ends andthesides of
`in order for us to correct torsion during implantation. The bal-
`the graft
`with a
`loon catheters have a shaft constructed of polyvinylchloride
`9-French diameter, and they contain either two or three lumina, depend-
`ing upon the numberof low-profile, nonelastomeric polyethylene bal-
`loons affixed to them. If two balloons are used, the caudal one has a cy-
`lindric configuration leading into a
`in the catheter shaft
`30-degree angle
`immediately below it in order to accommodatethe typical angle of origin
`
`

`

`Endovascular Repair of Abdominal Aortic Aneurysms
`
`91
`
`takenby theiliac arteries at the level of the aortic bifurcation. We gener-
`use
`to an inflated diameter of
`just two balloon diameters that adapt
`ally
`18 to 30 mm.
`Finally, the Teflon sheath has a diameter of 21-French and
`a valve at the proximal end.
`
`PREPROCEDURALEVALUATION
`Our evaluation of patients who may be potential candidates for endovas-
`cular treatmentis identical to that for those who receive conventionalsur-
`gical management. A CTscanof the abdominalaorta is performed
`at 1-cm
`intervals, usually with additional “cuts” at the proximal and distal ex-
`tent of the AAA. The CTscanallowsus to estimate the distance between
`the renal arteries and the aortic bifurcation, to calculate the diameter of
`the aorta at the upper and lower ends of the AAA, andto identify aortic
`or
`angulation that may influence our
`approach. In addition, we
`tortuosity
`obtain a
`biplanar abdominal aortogram in every patient because it pro-
`vides further detail concerning the visceral branches, the patency of the
`inferior mesenteric and lumbararteries, and the course and diameter of
`a
`the iliac arteries. Angiography is
`pigtail catheter with
`performed using
`gold calibrations at 2-cm intervals, and a
`radiopaque ruler also is posi-
`tioned vertically behind the patient.
`The aortogram is interpreted from a
`perspective focused on the cen-
`screen because this precaution minimizes distor-
`ter of the fluoroscopy
`tion of the image. Most AAAscontain laminated thrombusthat typically
`is distributed in a characteristic fashion. Thrombus usually involves the
`entire length of the AAA, but ordinarily
`is most dense in the distal aorta
`and tends to occlude the ostia of the lumbar and inferior mesenteric ar-
`an
`teries at that level. This feature produces
`enlarged flow lumen in the
`proximal AAA, together with a lumen of nearly normal diameter in its
`distal third. Therefore, unless the angiographic features are
`supplemented
`by CT data, the distal neck of the AAA may
`be considered adequate to
`secure a stented graft when, in fact, it is not. If we encounter substantial
`differences in the information concerning length and diameter that we
`we review our data
`have collected using CT scanning and aortography,
`to determine whether an obvious error has been committed. Minor dif-
`ferences are not uncommon, however, and we have found that the CT
`scan sometimes overestimates the intraluminal diameterof the aorta
`a
`by
`factor of 1 to 4 mm. Onceall dimensions have been determined,
`weare
`able to design the implantation device to be used in a
`specific patient.
`
`PROCEDURAL TECHNIQUE
`as
`though conventional AAA resec-
`The patient is prepared and draped
`tion were to be performed. Undereither local anesthesiaor, as we
`prefer,
`epidural anesthesia, the common femoral artery is exposed through
`a
`standard groin incision ipsilateralto the iliac artery that is straighter and
`less diseased
`angiographically. Sodium heparin (5,000 IU) is adminis-
`tered intravenously, and a number 18 Cournand needleis inserted ceph-
`alad into the common femoral artery. A soft-tip (0.038-in.) guidewire is
`advancedthroughthe needleintothe distal thoracic aorta, and a 5-French
`pigtail angiographic catheter is introduced over it. Once the catheteris
`positioned in the visceral segment of the abdominalaorta, the wireis re-
`
`

`

`92
`
`~_—i*J;.C. Parodi
`
`moved and preprocedural angiographyis performed. Thepigtail catheter
`has radiopaque calibrations at 20-mm intervals to obtain measurements
`a radiopaque
`from the computer-assisted angiogram, and weagain place
`to the axis of the aorta. After com-
`ruler behindthe patientthat is parallel
`paring our measurements to those that were collected previously, the ap-
`propriate endoluminal device may be constructed.
`stent by about twothirds, andis at-
`The graft overlaps the proximal
`tached to it using braided, synthetic suture material. We now
`place two
`sutures on both sides of the graft separated by 180 degrees because we
`one suture was used on
`once encountered graft dislodgment when only
`over the balloon(s), the graft is
`each side. After mounting the stent(s)
`folded and the entire assembly is introduced into a 21-French Teflon
`sheath.The guidewire is reintroduced,the pigtail catheter is removed, and
`the original wire is replaced with a
`super-stiff (Amplatz) wire. Another
`intravenous dose (5,000 IU) of sodium heparinis administered, and the
`sheath containing the device is advanced over the new wireto the level
`of the proximal neck of the AAA (Fig 6). The sheath then is removed,
`leaving the graft, stent(s), and balloon(s) in the aortic lumen (Fig 7). In
`orderto preventdistal migration during inflation of the proximalballoon,
`weoften initiate an intravenousinfusionof nitroglycerin to maintain the
`mean systemic blood pressure below 100 mm Hg. As soon as the blood
`pressureis stable at an appropriate level, the proximal balloonis inflated
`
`\\
`
`
`
`ei
`
`
`
`
`
`FIGURE 6.
`Introduction of the sheath for our clinical device using
`
`a femoral arteriotomy.
`
`

`

`Endovascular Repair of Abdominal Aortic Aneurysms
`
`93
`
`aR
`
`E
`
`
`
`
`
`
`{087
`<>
`
`oH:
`
`
`SAK
`
`
`
`
`FIGURE7.
`After removal of the sheath, the balloon is positioned for deploymentof the
`stent.
`
`
`
`FIGURE8.
`Anillustration of the three components of our clinical procedure: proximal
`stent deployment, distention of the graft, and placementof the distal stent.
`
`
`
`

`

`94
`
`J.C. Parodi
`
`
`
`FIGURE 9.
`A preprocedural aortogram (patient number10 in our
`
`series).
`
`
`
`for 1 minute with whatever volumeis necessary to attain the properdi-
`a
`ameter for that particular patient. To apply
`to
`perfectly cylindric shape
`the stent, we
`occasionally inflate the balloon again at both ends of the
`stent. None of our
`patients has experienced discomfort or
`significant
`blood pressure changes during balloon inflation. In one of them,
`we
`adopted the shape of the stent to an
`irregular aneurysm neck by making
`ofthe stent.
`repeated inflations under low pressure alongthe entire length
`we inflate the balloon
`stent has been deployed,
`After the proximal
`to distend it under low pressure (Fig 8). Pro-
`along the shaft of the graft
`vided all previous measurements were correct, the distal radiopaquecal-
`ibrations on the graft should be flush to the aortic bifurcation at the con-
`
` FIGURE10.
`
`Another aortogram following endovascular grafting in the same
`trated in Figure 9.
`
`patientillus-
`
`

`

`Endovascular Repair of Abdominal Aortic Aneurysms
`
`95
`
`clusion of the procedure (Figs 9 and 10). A completion aortogram then is
`obtained. If it demonstrates reflux around the distal end of the graft into
`the aneurysm sac, a secondstentis placedat that level. Special balloons
`have been designed specifically for this purpose because the distal neck
`a stan-
`of the AAA
`usually is short and would not accommodateeasily
`dard valvuloplasty balloon. These special balloonsare
`cylindric, and their
`catheters are
`near the balloon end ofthe shaft in
`angulated by 30
`degrees
`adaptto the angle at the origin of the commoniliac artery. When-
`orderto
`ever endovascular grafting is performed in patients who already havesus-
`tained previous lower-extremity embolization, we
`use two
`electively
`stents mounted on a double-balloon catheter. After deploying the proxi-
`malstent and then deflating its balloon, the distal stent is
`engaged byin-
`flating the second balloon underslight tension to straighten thegraft.
`The arteriotomyis closed after flushing theiliac artery to vent any
`debris that might have been loosened during the procedure. The patient
`is transferred to an intensive care unit for the next 24 hours, and usually
`on a
`2 additional days
`regular nursing floor before being dis-
`spends
`charged from the hospital. Antibiotic coverage has been limitedto a sin-
`1 g intravenously,
`30 minutes before the
`gle dose of cephalothin (Keflin),
`endoluminal procedure.
`
`PATIENT SELECTION
`patients, because it was
`Ourclinical experience has been limited to 13
`our intention to
`our results in considerable detail before recom-
`analyze
`mending endoluminal treatment of AAAs on a widescale. For this rea-
`son, we
`initially selected only those candidates who had serious comor-
`a
`bidities implying
`high surgical risk with conventional resection and
`we
`a few volunteers
`graft replacement. Although
`subsequently accepted
`who did not represent prohibitive surgical risks, two anatomic criteria
`must be met under any circumstances:
`andthe distal necks of the AAA mustbe atleast 2
`1. Both the proximal
`cm in length.
`2. At least one of the iliac arteries must be patent and
`sufficiently
`to access the device. A short segmentof iliac stenosis, how-
`straight
`ever, can be corrected by percutaneous transluminal angioplasty
`(PTA) in conjunction with the endoluminalgraft procedure.
`Wedecided notto treat patients who had patent inferior mesenteric
`or lumbar arteries in our
`pilot study, but, with these exceptions about
`half of all the candidates we evaluated met our other anatomiccriteria.
`are
`Wepresently
`investigating the feasibility of endoluminal placement
`a modification of our endoluminal
`of aortoiliac bifurcation grafts using
`approach through both commonfemoralarteries. If this refinement proves
`to be successful, we estimate that more than 90% of patients with an AAA
`eventually could be treated without a traditional operation.
`Twelveof our
`13 patients had AAAs, and 1 had a subclavian
`original
`arteriovenousfistula. There were 12 men and 1 womaninourseries. They
`ranged in age from 62 to 83 years (mean, 71 years), and their follow-up
`periods extend from 3 to 23 months (mean, 10
`months). The results in
`ourfirst 5 patients have been reported previously.”
`
`

`

`96
`
`—_—s*CJ:C.. Parodi
`
`
`
`
`
`a 10-mm Dacron graft to facilitate
`
`FIGURE 11.
`A temporary iliac conduit constructed using
`the introduction of a 21-French sheath.
`were considered to be high-risk candidates for
`Ten of the 13 patients
`standard AAAresection and refused conventional operations, and the re-
`maining 3 patients volunteered to be included in our
`protocol. Risk fac-
`tors included recent
`myocardial infarction, hemorrhagic brain infarction,
`and pulmonary edema(each in 1 patient); another patient had chronic
`renal failure. Six patients had chronic pulmonary insufficiency, and an-
`other 6 had severe coronary artery disease. All of the patients and their
`our method andits per-
`close relatives were informed fully concerning
`ceived risks, and all gave written consent.
`performedin ourcardiac laboratory underei-
`were
`The procedures
`
`i>
`wv
`‘|
`
`co
`
`G3
`
`
`
`FIGURE 12.
`An endovascular stent covered by
`
`a Dacrongraft.
`
`

`

`Endovascular Repair of Abdominal Aortic Aneurysms
`
`97
`
` FIGURE 13.
`
`A preprocedural angiogram demonstrating
`a
`caused by
`gunshot wound.
`
`a subclavian arteriovenousfistula
`
`or local anesthesia. Three of these procedures
`ther epidural (12 patients)
`were
`supplemented by PTA oftheiliac artery and another two
`by seg-
`mental graft replacement of common femoral aneurysms. Ourinitial at-
`tempt to introduce the sheath failed in 2 patients, but, in each case, the
`was
`employed successfully. Transfemoral access
`contralateraliliac artery
`
` FIGURE 14.
`
`Another angiogram following endovascular grafting in the same
`trated in Figure 13.
`
`patient illus-
`
`

`

`98
`
`J.C. Parodi
`
`in only 1 patient. In this case, we made
`from either side was
`impossible
`an anastomosis between the
`a short, extraperitoneal incision to perform
`a conduit
`commoniliac artery and a straight Dacron graft that provided
`for the sheath (Fig 11). After the stented aortic graft had been implanted,
`was removed, with the exception of a short segment that
`the iliac graft
`was oversewn nearits iliac anastomosis.
`was referred for treatment with
`simply
`Oneinteresting 62-year-old patient
`the mistaken diagnosis of a subclavian aneurysm. In fact, he had sustained
`an ar-
`gunshot wound 2 years earlier and subsequently had developed
`a
`pulsatile subclavian mass, a
`a
`prominent
`teriovenousfistula manifest by
`subcutaneous venouspattern, and hyperdynamic cardiomegaly.Athis re-
`a sheath in the ipsilateral axillary
`quest, we inserted our device through
`artery in order to implanta short endoluminal Dacrongraft using a 10-mm
`balloon deployed directly at the site of the fistula itself (Fig 12). All ob-
`vious manifestations of the previousfistula (thrill, bruit, and tachycar-
`was inserted, and an immediatearterio-
`dia) resolved as soon as the graft
`gram confirmedthe closure of the fistula. Another arteriogram obtained 1
`a
`patent subclavian artery without any appar-
`month later demonstrated
`ent abnormalities (Figs 13 and 14).
`
`COMPLICATIONS
`andall of our patients
`There were no deaths related to our
`procedures,
`our current follow-up period.
`have remained alive throughout
`in 5 of our 13 pa-
`We have encountered six nonfatal complications
`tients:
`Inguinal hematoma. This complication occurred only once, probably
`1.
`
` FIGURE 15.
`
`An aortogram demonstrating dilation of the distal abdominal aorta 18 months
`after endovascular implantation of a Dacron graft having inadequate length.
`
`

`

`Endovascular Repair of Abdominal Aortic Aneurysms
`
`99
`
`2.
`
` FIGURE 16.
`
`A computed tomographic scan after deploymentof a distal stent that engaged
`only the mural thrombusin the same
`patientillustrated in Figure 15.
`wasrelated to the fact that we did not reverse systemic hepariniza-
`tion at the conclusionof the first procedure in ourseries, and has not
`recurred since we have begun routine administration of postproce-
`dural protaminesulfate.
`Malpositioning of the proximal stent. This complication occurred in
`the second
`patient in our series and representeda technical error on
`the part of the interventionalist. This oversight could be corrected
`only by standard AAAresection and graft replacement, and it was
`during this operation that we
`recognized that surgical removal of an
`implanted stent is quite difficult (see later). Nevertheless, this partic-
`ular patient had an uneventful postoperative course.
`3. Leakage through the proximal stent. As indicated earlier, we
`initially
`secured our
`graft to its stent with only two sutures positioned 180
`degrees apart. In such a case, oneof the sutures loosened and permit-
`on that side. After identifying this
`ted the graft to migrate distally
`we
`problem during the procedure,
`successfully deployed another
`stented Dacrongraft within the lumen ofthe displaced graft. This pa-
`tient had no further complications.
`4. Miscalculation of graft length. In an