JOURNAL OF VASCULAR SURGERY
Volume 52, Number 2
Translumbar Embolization for Type II Endoleaks After EVAR: A Multi-
center Retrospective Review
Amelia J. Simpson, Joy Garg, MD, Ralph B. Dilley, MD, David Frankel,
MD, and Nikhil Kansal, MD, Department of Vascular Surgery, University of
California, San Diego School of Medicine, La Jolla, CA
Objective: Translumbar embolization (TLE) of type II endoleaks has
been described for the treatment of enlarging aneurysms following EVAR.
This technique is reported to have a very high rate of technical success and
durability. The purpose of this study is to review our experience with TLE in
controlling type II endoleaks, in arresting the increase in aneurysm growth,
and limiting the need for subsequent intervention.
Methods: A retrospective case review was performed on 13 patients
with CT and/or angiographically confirmed type II endoleaks who under-
went TLE at two institutions. Patients were treated with combinations of
stainless-steal coils, Onyx, Cyanoacrylate (NBCA), thrombin, and/or poly
vinyl alcohol (PVA) beads. Success was defined as clear resolution of the type
II endoleak and/or an aneurysm diameter that was either stable or decreas-
ing. Failure of the technique was defined as any persistent leak, an enlarging
aneurysm sac, or the need for secondary intervention.
Results: Thirteen patients underwent TLE for type II endoleaks asso-
ciated with aneurysm enlargement. In two patients who underwent trans-
lumbar puncture, no endoleak could be identified, and no treatment was
performed. TLE was successful in only 5 (45.5%) of the remaining 11
patients. Six patients (54.5%) had unsuccessful TLE. Of the 6 treatment
failures; 2 patients required repeat interventions, 2 required open surgical
repair (1 for rupture), 1 suffered colonic ischemia requiring resection, and 1
patient has a persistent type II endoleak.
Conclusions: Our experience contrasts with previously published
studies in that fewer than half of the patients treated had successful
resolution of their endoleak with TLE alone. Although TLE is a useful
technique for the management of type II endoleaks, many patients will
require subsequent procedures. Close surveillance of patients after TLE is
imperative to identify the patients who will require additional interven-
tion. The practice gap addressed is the difficulty in obtaining adequate
results using TLE in treating endoleaks. A small number of previously
published studies showed high success rates. By identifying a lower
success rate for TLE, we hope to impress the importance of vigorous
surveillance after the procedure.
The Transition from Custom-Made to Standardized Multi-Branched
Thoracoabdominal Aortic Stent Grafts
Timothy A. Chuter, MD, Ki-Hyuk Park, MD, Jade S. Hiramoto, MD, and
Linda M. Reilly, MD, UCSF, San Francisco, CA, Daegu, Republic of Korea,
and San Francisco, CA
Objective: To compare the branch morphology and short-term out-
come of endovascular aneurysm repair using custom-made vs standardized
multi-branched thoracoabdominal stent grafts.
Methods: Custom-made stent grafts (CSG) with patient-specific cuff
locations were compared with standardized stent grafts (SSG) with uniform
cuff locations. Data on patient demographics, aortic morphology, compo-
nent use, and outcome were collected prospectively. Final branch length
(cuff to target artery orifice) and branch angle (cuff orientation to target
artery orientation) were determined using 3-D reconstruction of computed
tomography angiograms (CTA).
Results: Since January 2008, 24 patients underwent endovascular
repair using 13 CSG (10 in 2008, 3 in 2009) and 11 SSG (1 in 2008, 10 in
2009). Two SSG patients were excluded from analysis: one has yet to
undergo CTA, the other had crossed renal branches due to problems
traversing a previously reconstructed aortic arch. All stent grafts were im-
planted successfully. There were no perioperative deaths. All branches were
patent on the initial postoperative CTA. There were no statistically signifi-
cant differences between the CSG and SSG groups in terms of patient
demographics, procedural details (operation length, fluoroscopy time, con-
trast volume, blood loss), or mean branch length and angle (Table).
Conclusions: The substitution of SSG for CSG had no effect on the
complexity of the procedure, the final branch morphology, or the perioper-
Table. Comparison of branch morphology between custom-made and standardized stent grafts
Mean branch angle (degrees)
Target artery CSG (sd) SSG (sd)
CA (n 22) 14.5 (13.3) 20.4 (11.2)
SMA (n 22) 25.6 (18.6) 17.3 (17.6)
RRA (n 21) 25.3 (19.9) 25.0 (14.6)
LRA (n 21) 26.1 (23.0) 28.7 (25.0)
All (n 86) 22.8 (19.0) 22.9 (17.6)
Abstracts 529
ative outcome. The availability of an off-the-shelf SSG will broaden the
application of endovascular TAAA repair by eliminating manufacturing
delays.
The Hybrid Procedure: Arch and Visceral/Renal Debranching Com-
bined with Endovascular Repair for Thoracic/Thoracoabdominal Aor-
tic Aneurysms
Sung W. Ham, MD, Terry J. Chong, MD, Vincent L. Rowe, MD, Mark J.
Cunningham, MD, Robbin G. Cohen, MD, and Fred A. Weaver, MD,
Surgery, University of Southern California, Los Angeles, CA
Objective: To report a single-center experience using a hybrid proce-
dure (open debranching, followed by endovascular aortic repair) for treat-
ment of thoracic/thoracoabdominal aortic aneurysms.
Methods: From 2005-2010, 48 patients (31 men; mean age, 71)
underwent a hybrid procedure for thoracic/thoracoabdominal aortic aneu-
rysms. Thirty-day, in-hospital morbidity, and mortality; and late endoleak,
graft patency, and survival were analyzed. Graft patency was assessed by CT,
angiography, or duplex ultrasound.
Results: Hybrid procedures were used to treat 24 thoracic (11 arch, 13
descending), 19 thoracoabdominal (Crawford type I-III: 3, type IV: 7, type
V: 9), and 5 para-anastomotic aortic aneurysms. The hybrid procedure
involved debranching arch vessels (47) or visceral/renal vessels (73) using
bypass grafts, followed by endovascular repair. Eighty percent of debranch-
ing and endovascular repair procedures were staged, with an average interval
of 25 days. Major 30-day and in-hospital complications occurred in 38% of
patients, and included bypass graft occlusion (4), endoleak reintervention
(2), and paraplegia (1). Thirty-day and in-hospital mortality was 4.2%,
secondary to bowel ischemia in one patient and aneurysm rupture in the
other. During a mean follow-up of 13 months, three type II endoleaks
required four interventions. Seven patients died: one of sepsis, one of
respiratory failure, and five of unknown cause. There were no documented
aneurysm-related deaths or new graft or limb occlusions. Primary bypass
graft patency was 96%. Actuarial survival was 85% at 1 year and 67% at 3
years.
Conclusions: The hybrid procedure for the treatment of thoracic/
thoracoabdominal aortic aneurysms is a competitive alternative to traditional
open repair and endovascular repair with branched stent grafts. The results
validate this approach for the short-term to midterm; however, longer
follow-up is required to appraise its durability.
Does Anticoagulation and/or Antiplatelet Treatment Influence the
Incidence Early Type II Endoleaks and Increase the Need for Post-
EVAR Reintervention?
Rafael Lemus-Rangel, MD, General Surgery, Kaiser Permanente Los Ange-
les Medical Center, Los Angeles, CA
Objective: The most common complication after endovascular aneu-
rysm repair (EVAR) is a type II endoleak. The purpose of this study is to
determine the incidence of early endoleak and the rate of post-EVAR
reinterventions in the setting of chronic anticoagulation (Coumadin) and
antiplatelet (AP) therapy.
Methods: A retrospective review of 366 patients undergoing an EVAR
over a 5-year period was performed. Patients were stratified according to the
type of oral anticoagulation therapy and AP treatment. Those treated with a
single oral antiplatelet (ASA or Plavix), dual AP, Coumadin, or combined
AP agent and Coumadin comprised the study groups. The control group
was defined as patients with healthy coagulation and platelet profiles. Fur-
ther group analysis included the incidence of endoleak and the need for
reintervention.
Results: A total of 324 men and 42 women (mean 74 8 [SD] years;
range, 48-90 years) underwent EVAR with a mean AAA size of 5.96 1.16
cm (range, 3-10 cm). Time to CT scan was 97.05 135.24 days (range,
3-1528 days). Irrespective of AP or anticoagulation profile, the overall
incidence of type II endoleak was 18% (67 of 366). The combination of AP
and Coumadin therapy resulted in 45% incidence (9 of 20, P .001), while
Mean branch length (mm)
P-value CSG (sd) SSG (sd) P-value
0.29 25.1 (15.7) 20.2 (8.1) 0.40
0.31 27.0 (12.5) 19.6 (3.0) 0.10
0.96 23.5 (11.3) 24.9 (11.6) 0.78
0.81 25.3 (8.4) 27.3 (10.6) 0.64
0.98 25.3 (12.0) 23.0 (9.1) 0.33

530 Abstracts
just Coumadin resulted in 32% incidence (12 of 38, P .008) of endoleak.
Dual AP therapy had a 12% rate (3 of 25, P .427). Single AP therapy
resulted in a 16% incidence (34 of 225, P .29). Patients with a normal
coagulation and platelet profile had a 13% incidence (11 of 82) of endoleak.
There was a 9.6% need for postoperative reinterventions for all comers. For
patients on a combination of AP and Coumadin, there was a 35% need for
postoperative intervention (P .001) compared with 16% if just on Cou-
madin alone (P .044). This decreased to 13% for those on dual AP agents
(P .06) and 7% for both single AP (P .396) or no agents.
Conclusions: Combination AP and Coumadin therapy has the highest
risk for the development of endoleak. Coumadin alone was also found to be
a significant risk factor. Dual AP therapy, single AP therapy, and no agents
were equivalent. In addition, the combination of AP and Coumadin resulted
in a high incidence of post-EVAR reintervention. These patients should be
monitored closely after endovascular aortic aneurysm repair.
Candidate Compounds for Drug-Device Hybrid Therapies in the Man-
agement of Abdominal Aortic Aneurysm (AAA) Disease
Yasunori Iida, MD, Geoffrey M. Schultz, MD, Monica M. Dua, MD,
Noriyuki Miyama, MD, Susan Rea Peterson, PhD, Shola Sulaimon, DVM,
MS, PhD, and Ronald L. Dalman, MD, Stanford University School of
Medicine, Stanford, CA, and Medtronic, Inc, Santa Rosa, CA
Objective: AAA is a common and potentially lethal condition. Endo-
vascular aneurysm repair (EVAR) reduces morbidity but is less durable than
open repair secondary to continued remodeling of the diseased aorta. Local
delivery of compounds inhibiting degeneration of the neck and promoting
sac regression may stabilize endografts and improve durability. Our objec-
tive was to determine if HMG-CoA reductase inhibitors (statins) and
angiotensin receptor blockers (ARBs) are appropriate candidates for drug-
device hybrid therapies based on their efficacy in experimental AAA attenu-
ation.
Methods: Apo E/ male mice were infused with angiotensin (Ang)
II (1.0 mg/kg/min) via minipumps. Statin (n 6; 40 mg/kg/d), ARB1
(n 6; 50 mg/kg/d), and ARB2 (n 6; 10 mg/kg/d) were administered
via drinking water (statin) or chow (ARBs) and compared with control mice
treated with Ang II alone (n 6). Aortic diameter was serially determined
with ultrasonography. Mice were sacrificed 28 days after pump implantation.
Immunohistochemistry was performed to evaluate aortic elastin preserva-
tion, medial SMC maintenance, and mural inflammation and neovascular-
ization. Patterns of gene expression were determined via qPCR. Serum drug
concentrations were quantified on the day of sacrifice.
Results: Multiple dissections (n 4) and ruptures (n 3) occurred in
control animals. None were observed in statin- or ARB-treated mice. Statin
(1.30 0.19 mm), ARB1 (1.17 0.12 mm), and ARB2 (1.13 0.05 mm) Maureen M. Tedesco, MD, Jeffrey A. Norton, MD, Robin M. Cisco, MD,
treatment significantly attenuated aortic dilatation compared with control
(1.69 0.46 mm; P .05; Fig). Histologic, biomarker, and serologic
analyses were not complete at the time of submission.
Conclusions: Statins and ARBs inhibit experimental AAA progression.
Further histologic and biomarker analysis is forthcoming and will aid in the
determination of mechanisms of statin and ARB aneurysm inhibition. Local
delivery of these compounds may limit endograft failure via inhibition of
aortic remodeling following AAA exclusion.
Fig. Comparison of baseline and terminal aortic diameters, *P
.05 Control vs all other groups.
JOURNAL OF VASCULAR SURGERY
August 2010
Techniques and Results of Portal Vein/Superior Mesenteric Vein
Reconstruction Using Femoral and Saphenous Vein During
Pancreaticoduodenectomy
Dae Y. Lee, MD, Erica L. Mitchell, MD, Mark A. Jones, MD, Gregory J.
Landry, MD, Timothy K. Liem, MD, Brett C. Sheppard, MD, Kevin G.
Billingsley, MD, and Gregory L. Moneta, MD, Division of Vascular Surgery,
Dept of Surgery, Oregon Health and Science University, Portland, OR and
Portland, OR
Objectives: Patients with pancreatic tumors may have portal vein (PV)
and/or superior mesenteric vein (SMV) invasion. In such cases, lower
extremity veins can provide an autogenous conduit for PV/SMV recon-
struction. Little data exist, however, describing the technique of PV/SMV
reconstruction, the patency of such reconstructions, and the morbidity of
using lower extremity veins for PV/SMV reconstruction during pancreati-
coduodenectomy.
Methods: Thirty-four patients (mean age, 62.6 years) underwent
PV/SMV reconstruction during pancreaticoduodenectomy using lower
extremity vein. Saphenous vein was preferred for patching and femoral vein
for replacement. We analyzed preoperative imaging, reconstruction patency,
vein harvest morbidity, and late mortality.
Results: All 34 patients had preoperative CT imaging and/or
endoscopic ultrasound (EUS) imaging. Of the 34 patients, 14 had
evidence of PV/SMV invasion on CT or EUS, 14 did not, and 6 studies
were indeterminate. Twenty-five patients had follow-up imaging, and 22
(88%) had patent reconstructions. Fifteen patients had PV/SMV replace-
ment using femoral vein. Of these 15 patients, 7 had minor postoperative
lower extremity edema that resolved over time, 5 had wound complica-
tions from the femoral vein harvest site, 3 of which required minor
operative procedures for treatment. Fifteen patients had PV/SMV patch-
ing with greater saphenous vein, none had postoperative wound prob-
lems, and one had minimal postoperative lower extremity edema. Four
patients had PV/SMV patching using femoral vein, none had postoper-
ative wound problems, and one had minimal postoperative lower extrem-
ity edema. Compared with patients undergoing pancreaticoduodenec-
tomy without PV/SMV reconstruction, Kaplan-Meier analysis showed
there was no difference in late mortality.
Conclusions: Preoperative imaging may fail to detect PV/SMV in-
volvement in patients undergoing pancreaticoduodenectomy. PV/SMV
reconstruction with leg vein provides good patency with minimal postoper-
ative lower extremity complications and no increase in late mortality. The
lower extremities should be routinely included in the operative field of
patients undergoing pancreaticoduodenectomy.
Pancreatic Mass Resection and Revascularization
Tae K. Song, MD, and E. John Harris, Jr, MD, Surgery, Stanford University
Medical Center, Stanford, CA
Objective: Resection of the pancreatic mass is the only curative option
for pancreatic cancer. Tumors that invade the nearby vasculature are deemed
unresectable. By utilizing revascularization and bypass techniques, we re-
sected pancreatic lesions with vascular invasion to improve survival in this
patient population.
Methods: Patients who presented to a high-volume, tertiary referral
center with unresectable pancreatic lesions due to vascular invasion were
evaluated by the surgical oncology and vascular teams. If both teams
found revascularization and vascular repair options were available to the
patient, a pancreatic resection was performed by the surgical oncology
team with revascularization and repair performed by vascular surgery.
Patients were followed-up by both surgical teams while in the hospital
and as outpatients.
Results: Seventeen patients (23.5% females) underwent pancreatic
resection with revascularization or vascular repair between August 2005 and
May 2009. The average age was 60.5. Average pancreatic mass size was 4.6
cm. The most common vessels requiring repair or revascularization were the
superior mesenteric and portal veins. (Two patients required SMA revascu-
larization, and one patient required right hepatic artery repair.) A pylorus-
sparing pancreaticoduodenectomy was done in 71% of patients; the remain-
der underwent distal pancreatectomy and splenectomy. All operations were
performed open. Vessels were reconstructed with cyropreserved vein (18%)
or autologous vein graft. There were no postoperative deaths. Two patients
(12%) returned to the operating room within 24 hours for a complication
related to the vascular repair, both of which were treated successfully.
Adenocarcinoma comprised 65% of lesions, and the remaining were neu-
roendocrine tumors. At this time, 82% of the patients are alive (follow-up
range, 4-48 months).
Conclusions: Planned en bloc pancreatic resection with direct revas-
cularization and/or vascular repair for traditionally unresectable lesions is a
feasible option in a high-volume center. This requires preoperative planning
and a multidisciplinary effort. Long-term studies are required to determine
the impact on patient survival, which initially appears favorable compared
with historic controls.