Abstract
Objective:
To report on the outcomes of patients undergoing an iliac branch device implantation after previous open or endovascular aorto-biliac repair, using exclusively femoral access for catheterization and delivery of the covering stent to the hypogastric artery.
Methods:
Single-center retrospective study in which all patients in whom an iliac branch device was implanted after previous open or endovascular aorto-biliac repair were identified. Patients in whom the hypogastric artery catheterization and delivery of the bridging cover stent were achieved via exclusive femoral access were included. Different techniques were used based on surgeon preference. Technical success and access-related complications, as well as iliac branch device endoleak or occlusions during follow-up, were evaluated.
Results:
From 2015 to 2021, 28 patients with a prior open or endovascular aorto-biliac repair underwent 34 iliac branch device implantations. Most (71%) had juxtarenal or thoracoabdominal aortic aneurysms, 82% had common iliac artery aneurysms, and 25% had hypogastric artery aneurysms. Bilateral iliac branch device implantations were performed in 21% of the patients, and in 26% of cases, landing in the superior gluteal artery was obtained. An “up-and-over” technique from the contralateral groin was used in 65% of the cases, and a steerable sheath in 35%. Technical success was 94%, with no complications related to access or technique to catheterize and deliver the stents in the hypogastric artery. The cohort had 20% of major complications, with 3 perioperative deaths. Kaplan–Meier estimated an iliac branch device freedom from occlusion and endoleak was 92% and 83% at 2 years.
Conclusions:
The implantation of an iliac branch device over previous aortic or open endografts involving the aortic bifurcation is feasible and safe. We suggest using a femoral approach as the primary access of choice.
Clinical Impact
In this study we present 28 patients with previous aortoiliac grafts in which iliac branch devices were performed as a subsequent step.We demonstrated the feasibility of the technique despite the difficulty of crossing a neobifurcation, with a steep angle, without complications associated with the technique. Based on our experience, we recommend transfemoral access as the first option for bypassing the hypogastric artery stent, preserving upper extremity access and its possible complications.
Introduction
Endovascular aortic repair (EVAR) has become the technique of choice for repairing abdominal aortic aneurysms (AAA) due to its minimally invasive nature. However, the rate of reintervention after EVAR remains higher than that of open AAA surgery. 1 One failure mode that may require reintervention is the loss of distal iliac seal caused by continuous enlargement of the common iliac artery (CIA), necessitating extension of the repair to establish a seal in further distal sealing zones. Several techniques have been described, the simplest being embolization of the hypogastric artery (HA) and distal extension to the external iliac artery (EIA). However, this approach may result in buttock claudication or erectile dysfunction in up to 50% of patients, 2 underscoring the importance of preserving the HA as recommended in current guidelines. 3 This becomes particularly important in complex aortic repairs using fenestrated or branched endovascular techniques, as these procedures involve coverage of longer segments of the aorta with an increased risk of spinal cord ischemia if collateral pathways, including the hypogastric circulation, are not revascularized. 4
Over 15 years ago, iliac branch devices (IBDs) were introduced and have consistently demonstrated excellent results, with long-term patency above 90% and low reintervention rates,5,6 making IBDs an ideal option for distal extension with HA preservation. In primary repairs, IBDs are usually deployed first, allowing the use of a contralateral sheath to cross the native aortic bifurcation and facilitate crossover catheterization and delivery of a bridging-covered stent (BCS) for the HA. Subsequently, the aortic bifurcation component is deployed and bridged to the IBD. When dealing with a previous endograft that has lost its distal iliac seal or a graft with ongoing aneurysmal degeneration of the iliac artery, IBD repair faces challenges associated with the acute angle of the flow divider in the neobifurcation of an open aorto-biliac graft or a bifurcated endograft. Consequently, catheterization of the HA has been performed through upper extremity access (UEA), which carries a risk of cerebral embolization and arterial or peripheral nerve injury.7,8
To mitigate these risks, a femoral approach is preferred, with limited studies focusing on the feasibility and safety of this method.9,10 This study is aimed to report on the outcomes of patients undergoing IBD implantation after prior open or endovascular aorto-biliac repair, exclusively using femoral access for catheterization and delivery of the covering stent to the HA.
Materials and Methods
A retrospective analysis was conducted of data from consecutive patients who underwent IBD placement at our institution between January 2015 and December 2021.
The study protocol was considered exempt from the need for ethical approval, given the retrospective nature and the use of anonymized data.
The study included patients with previous open or endovascular aorto-biliac repair who received IBD placement into a previous graft (Figure 1A and B), with HA branch catheterization and BCS delivery via femoral access. Patients with previous open or endovascular aortic repair not involving the aortic bifurcation were excluded, as well as patients who had an IBD placed through a previous graft but had the HA branch catheterized through a UEA.

(A) Fluoroscopy image of a previous bifurcated endograft undergoing iliac branch devices (IBD). (B) Digital subtraction angiography showing a patient with a previous open Dacron aortoiliac bypass undergoing IBD.
Indications for IBD implantation encompassed the presence of an isolated CIA or HA aneurysm 35 mm or larger. In cases associated with a thoracoabdominal, juxtarenal, or infrarenal abdominal aortic aneurysm (AAA) endovascular repair, a CIA larger than 22 mm was also an indication for IBD. Smaller iliac aneurysms with rapid growth or symptomatic status were also indications of IBD. In addition, the presence of a type Ib endoleak (EL) resulting from the growth of the CIA landing zone, which was insufficient to seal with a CIA extension, also served as an indication for IBD placement.
All patients underwent preoperative computed tomography angiography (CTA) of the entire aorta and iliac arteries with a minimum slice thickness of 3 mm. A dedicated planning software (Aquarius Intuition, TeraRecon Inc, San Mateo, California) was utilized for all case planning. Two commercially available IBDs (ZBIS; Cook Medical, Bjaeverskov, Denmark and E-iliac; Artivion Inc., Kennesaw, GA, USA) were used for anatomically suitable cases, while a custom-made device (CMD) was ordered and implanted for cases that did not meet the IFU (intructions for use) criteria for the available devices.
Procedure
All procedures were performed in a hybrid operating room equipped with a fixed imaging system (Phillips Allura, Philips Healthcare, Best, Netherlands). As per current standard practice, a femoral approach was adopted for IBD placement unless anatomical restrictions (occlusion or severe atherosclerotic-calcific disease) in the contralateral femoral or iliac arteries made it infeasible.
Based on the preference of the treating surgeon, one of the following HA catheterization techniques was chosen:
Up-and-over technique: Bilateral percutaneous femoral accesses are obtained, and stiff guidewires are advanced into the thoracic aorta from both accesses. The IBD is introduced on the ipsilateral side, while a 12F × 45 cm sheath (Flexor, Cook Medical) is introduced on the contralateral side. A preloaded wire, incorporated in the commercially available IBD, is then snared to establish a through-and-through access. Using this wire, the contralateral 12F sheath is carefully advanced to the graft flow divider, and an angiogram is performed. After identifying the HA ostium, the IBD is released up to the opening of the HA branch, typically positioned 1 cm above its HA ostium. The 12F sheath is then advanced over the through-and-through wire by pulling on both ends of the wire and simultaneous pushing of the 12F sheath (push and pull technique). Once the tip of the sheath's dilator emerges into the contralateral limb, tension is released on both sides of the wire, allowing the sheath to bow, forming a curve, and, thus, avoiding excessive pressure on the graft flow divider. The sheath is then advanced over the dilator until it is lodged inside the HA side branch when possible (Figure 2). In cases of significant resistance, the sheath is left within the CIA segment of the IBD without reaching the HA side branch, and a 7F or 8F Flexor sheath may be advanced to aid in catheterization and delivery of the BCS.
Use of steerable sheaths: Similar to the up-and-over technique, bilateral percutaneous femoral accesses are obtained, and the IBD and steerable sheath are advanced through both CIAs, respectively. The preloaded wire is snared and the IBD is opened in a similar fashion until the HA branch is accessible. With the assistance of the through-and-through wire, 11 the sheath is steered into position and held proximal to the flow divider by keeping both wire ends taut until the BCS is positioned with or without the use of an additional 7F or 8F sheath. Alternatively, an ipsilateral approach could be employed, using a balloon to occlude the common iliac segment of the IBD, providing support to advance a sheath into the HA. 12

Fluoroscopic images showing 12F sheath advancement across the flow divider of a previous endograft in 2 different cases. (A, B) Sheath advancement (white arrow) with the dilator as a single unit, releasing wire tension and allowing a curve to form over the flow divider (white arrowhead). (C, D) Once the sheath has reached the common iliac segment of the iliac branch devices (IBD), the sheath is advanced over the dilator, reaching the hypogastric side branch (white arrow).
All patients were monitored in a specialized unit for 1 night, and a postoperative CTA was performed as part of the discharge protocol, unless contraindicated.
Data
Electronic records, preoperative and postoperative CTA, and angiograms of the procedures, as well as any reinterventions, were reviewed. Patient demographic characteristics, previous interventions, aneurysm types and sizes, side of intervention, and surgical indications were recorded.
The type of IBD, BCS, and additional stents used, as well as the presence of postoperative EL on perioperative and follow-up CTA, were analyzed on a per-implanted IBD protocol, including both sides in cases with bilateral IBD.
Technical success was defined as successful implantation of the IBD with the exclusion of the underlying aneurysm, absence of any IBD-related type I or III endoleak and preservation of antegrade flow to the HA and EIA, without damage to any of the previously placed endovascular stents, when applicable.
The primary endpoint of the study was to assess technical success and evaluate severe perioperative complications, defined according to the SVS (Society of Vascular Surgery) reporting standards for endovascular aortic repair, 13 as any complication within 30 days of the procedure that necessitates major surgical or medical intervention, may be associated with prolonged convalescence, usually accompanied by prolonged or permanent disability, or may result in death. Secondary endpoints included assessing follow-up occlusion or IBD-related EL. Aneurysmal sac stability, defined as the absence of growth greater than 5 mm, was also evaluated.
Data were presented as total numbers (n) and percentages (%) for qualitative variables, while quantitative variables were expressed as medians with interquartile ranges (IQR) denoted as (Q1 and Q3). Comparison between categorical variables was performed using Fisher exact tests, with statistical significance set at p<0.05. Kaplan–Meier curves were created to assess IBD freedom from occlusion or endoleaks.
Results
From 2015 to 2021, 234 IBDs were implanted in 184 patients. There were 28 patients who underwent 34 IBD implantations through a previous graft involving the aortic bifurcation and in whom the HA branch was catheterized through a transfemoral access, representing our cohort. In the same period, there were also 6 patients who had 7 IBD implanted through a previous graft but using an UEA, so they were excluded from this analysis.
A previous endovascular graft with a bifurcated component was present in 86% (24/28) of the patients (all of whom now presented with a type Ib EL), whereas 4 patients had a previous Dacron aorto-biliac bypass, 3 of them presenting with CIA aneurysm formation >35 mm, and 1 with a 51-mm HA aneurysm.
The majority of the cohort had a juxtarenal or thoracoabdominal aneurysm (71%), 82% (23/28) had CIA aneurysms, and 25% (7/28) had HA aneurysms. There were also 3 patients who had neither CIA nor HA aneurysms but had a type Ib EL with a CIA too short to ensure a secure seal only with distal extension. Table 1 shows baseline morbidity, characteristics of aneurysms, and previous repair. Only 1 patient presented with a symptomatic aneurysm. The patient, who already had an aneurysm treated with an aorto-biliac bypass, and who had experienced visceral aortic segment growth, presented with an acute type B aortic dissection and acute pain in a left 31 mm CIA aneurysm, so he underwent a thoracic endovascular aortic repair (TEVAR) plus a branched endovascular aortic repair (bEVAR) plus distal extension with a left IBD.
Baseline Characteristics.
Abbreviations: CIA, common iliac artery; COPD, chronic obstructive pulmonary disease; fEVAR, fenestrated endovascular aortic repair; IQR, interquartile ranges.
One patient with an open aorto-biliac repair and posteriorly a fEVAR without a bifurcated component.
Of the 28 patients, 79% (22/28) had unilateral and 21% (6/28) had bilateral IBD implantation. Of the 34 IBDs implanted in these patients, 59% (20/34) were placed on the right side. No patient underwent HA embolization.
The ZBIS IBD was used in 79% (27/34) of cases. E-iliac IBD and a CMD IBD were used in 12% (4/34) and 9% (3/34) of the time, respectively.
Table 2 summarizes the characteristics of the implanted devices. In 24% (8/34), an f-bEVAR with a bifurcated component was implanted in the same procedure as the IBD, while in 44% (15/34), the IBD was implanted via a previously implanted f-bEVAR with a bifurcated component. In 29% (10/34), the IBD was implanted over a previous EVAR, and in 3%, over a previous old Dacron graft without any additional endograft.
Implanted Devices Characteristics.
Abbreviations: BES, balloon-expandable stent; CMD, custom-made device; EIA, external iliac artery; HA, hypogastric artery; IBD, iliac branch device; SES, self-expanding stents.
The majority of the HA BCS were balloon-expandable covered stents, Advanta V12 (Atrium Medical Corporation, Merrimack, NH, USA) in 35% (12/34), VBX (W. L. Gore, Flagstaff, AZ) in 32% (11/34), particularly frequently after 2020 and in 1 case there was a Begraft (Bentley InnoMed, Hechingen, Germany) used. There were 7 cases (21%) done with a self-expandable BCS (Viabahn; W. L. Gore, Flagstaff, AZ), all landing in the superior gluteal artery. In 29% (10/34) of the cases, an uncovered balloon-expandable stent (BES) was used to reinforce the overlap between the IBD HA cuff and a self-expandable covered stent (in 8 cases) or to reinforce a short overlap among a BCS and the HA cuff (2 cases).
EIA relining with bare self-expanding stents (SES) was utilized in 53% of the IBDs, 94% (17/18) of times with a Wallstent (12–16 mm; Boston Scientific Corporation, Marlborough, Massachusetts), and 4% with a Protégé self-expandable stent (Medtronic/Covidien, Minneapolis, Minnesota). In 5 cases, a BES was used to rectify tortuosity within the CIA and in 2 cases with a narrow iliac bifurcation in a kissing stent configuration between the HA mating stent and the EIA.
In 64% (22/34) of the cases, we used the up-and-over technique with the use of a 12F × 45 cm Flexor sheath (Cook Medical) in all but 1 case (10F Flexor sheath). In 35% (12/34), we used a steerable sheath to facilitate catheterization and stent delivery in the HA. In 8 cases, the 10F Fustar sheath (Lifetech Scientific, Shenzhen, China) was used, and in 3 cases, the 12F Destino Twist (Oscor Inc, Palm Harbor, FL, USA) was used. In 1 case, we completed catheterization and HA BCS placement through ipsilateral access with the aid of an 8,5F Destino Twist steerable sheath (Figure 3A–C). The steerable sheath approach was used 67% of the time in the last 2 years of the study inclusion time (2020–2021).

(A) Fluoroscopic image showing a right iliac branch devices (IBD) through a steerable sheath. (B) Fluoroscopic image showing a right IBD through an up-and-over technique. (C) Fluoroscopic image showing a left IBD through an ipsilateral access with a steerable sheath and additional support given by a balloon (white arrowhead).
In 26% (9/34) of cases, the distal landing zone was the superior gluteal artery. In 3 cases presenting with CIA aneurysm and ectasia of HA, it was necessary to land in a healthy target vessel; in the other 6 cases, it was to exclude HA aneurysms. Five cases required concomitant embolization of gluteal branches to seal, 4 times with the use of Nester coils (Cook Medical, Bjaeverskov, Denmark) and 1 time with the use of an Amplatzer Vascular Plug (AVP2; Abbott Vascular, Redwood City). In 3 cases, only 1 branch from the main HA was embolized, while in 2 cases, 2 branches were embolized with coils. The median (IQR) diameter at the landing zone in this group was 7 mm (6.5–8) vs 10 mm9, 11 in the group with landing in the main HA trunk (p=0.001).
Technical success was achieved in 94% of the IBD (32/34), with the exception of 2 early occlusions, one due to a compression of proximal end of the BCS by the bridging iliac limb and another thrombosis due to an acute angle in the EIA, which successfully underwent percutaneous thrombectomy and relining. There were no cases of graft migration or other complications associated with the up-and-over or steerable sheath technique.
Table 3 shows these results.
Early and Follow-up Results.
Abbreviations: AAA, abdominal aortic aneurysms; EL, endoleak; HA, hypogastric artery; IBD, iliac branch device.
Five (18%) patients had a major perioperative complication, including 3 patients with perioperative mortality, 1 patient with severe chronic obstructive pulmonary disease (COPD) who underwent a fenestrated EVAR (fEVAR) + IBD, developed postoperative severe pneumonia and eventually refused further therapy, 1 patient with liver cirrhosis and pulmonary sarcoidosis developed severe pneumonia with intraalveolar hemorrhage; and 1 patient died secondary to multiple complications including paraplegia after a bEVAR, 1 patient had an access site hematoma on the side of the IBD after failed percutaneous closure that required open revision, and 1 patient presented with a thrombosed iliac limb, who after thrombectomy and implantation of a left IBD, ended with occlusion of a previously placed IBD on the right side.
Thirteen (46%) patients had an early postoperative EL, not related to the IBD. Seventy-six percent (10/13) were type II EL, 5 underwent IMA (inferior mesenteric artery) coiling during follow-up, and the rest were conservatively managed. Median (IQR) length of stay in the intensive care unit (ICU) and overall (OA) stay were 1 (1–3) and 8 (5–11) days, respectively, mainly triggered by the simultaneous f-bEVAR.
Follow-up was achieved for 75% (21/28) of the cohort with a median (IQR) of 1.8 years (1.2–2.4).
At follow-up, there were 12 non-IBD-related aortic reinterventions in 9 (32%) patients who completed follow-up.
There were also 2 type Ic EL from a HA stent both 1.8 years after the procedure and successfully intervened, one with relining in the distal main HA trunk, the other with postdilatation of the previously placed balloon-expandable BCS. There were 2 patients with a complete left iliac limb occlusion, including CIA, HA, and EIA parts of the IBD, one 2 years after the procedure, who underwent a crossover femoro-femoral bypass, and another 4 years after the procedure (the same patient who had a type Ic EL from the HA BCS).
Kaplan–Meier curve estimated freedom from occlusion of 92% 2 years (standard error [SE] <10%), while 2 years freedom from EL was 83%, but with over 10% of SE (Figure 4).

Kaplan–Meier curves showing freedom from IBD occlusion and endoleak. Arrow indicates when SE >10%. IBD, iliac branch devices; EL, endoleak; SE, standard error.
At follow-up, 67% of patients experienced regression or stability of the aortic aneurysmal sac as well as 100% of the iliac aneurysms, with a median (IQR) diameter growth of 4 (2–8) mm for the aorta and a decrease of 2 (0.3–5) mm for the iliac aneurysms. Of the patients who experienced growth in the aneurysmal sac, 64% had an f-bEVAR, in comparison to 36% with EVAR (p>0.05).
Discussion
We present the results of a patient series treated with IBDs following previous endovascular aortic repair (EVAR) or open repair exclusively using transfemoral access.
When using the IBD, after deploying the iliac main body, a preloaded wire is utilized to achieve a through-and-through access from both femoral arteries. This access is crucial, as it stabilizes the sheath used to catheterize and deliver the covered mating stents to the HA. Typically, the through-and-through wire must be kept taut, exerting downward pressure on the aortic bifurcation with the risk of device migration in cases after EVAR.
In patients with prior open repair involving the aortic bifurcation, the flow divider generally results in a narrow and sharply angled bifurcation, making access from the contralateral groin difficult. Similarly, in patients with prior EVAR, the flow divider poses problems for contralateral crossing, with the added concern of graft dislocation if excessive downward tension is placed on the neobifurcation.
Reintervention with IBD in patients with previous EVAR has been described; however, most studies have favored a UEA to catheterize the HA. Bisdas et al 5 presented a series of 18 patients with IBD installation over previous EVAR using upper extremity access exclusively; similarly, Spath et al 14 recently reported a multicenter experience with 75 patients undergoing 88 IBD implantations over previous EVAR or f-bEVAR, 93% of which had UEA.
Although UEA has the advantage of a straighter access route, it carries the potential risk of cerebral embolization, arterial or peripheral nerve injury and may increase irradiation to the operator.7,15 So it appears preferable to avoid UEA if possible. Presumably, some of the factors that promote the use of UEA despite these risks are the possibility of access failure or concern about graft migration or dislocation when accessing from contralateral femoral access.
Nevertheless, there have been case series demonstrating the feasibility of IBD implantation over previous grafts using femoral approach. Tenorio et al 9 presented a series of 11 patients who had an IBD implanted with the Gore device after previous EVAR using an up-and-over technique, with 1 patient requiring brachial access due to technical failure. Mesnard et al 10 recently reported another series of 28 patients undergoing IBD with a ZBIS (Cook) after previous EVAR, 14 of them with exclusively femoral access, without complications related to the up-and-over technique.
In comparison to these previous experiences, we employed different devices (Cook and Artivion) and different approaches. The up-and-over technique was utilized in most cases (61%), complemented in the rest by the use of steerable sheaths. Steerable sheaths have been available for at least a decade, and previous reports have demonstrated their use to perform an IBD with prior EVAR with transfemoral access.16–18 They are also part of our essential tools for complex aortic repair. 19 Several brands with different characteristics are available, and it is recommended to always use the IBD preloaded wire to stabilize the sheath and facilitate the HA mating stent placement.
As in the aforementioned studies, we did not encounter any complications related to the technique used for BCS delivery to the HA. In our cohort, BECS were most frequently used (70%), with the VBX stent being particularly favored from 2020 due to its good trackability and stent-retention.
The described cohort had a near 20% complication rate, driven mainly by 3 perioperative deaths. However, this series comprises a highly complex group of patients, as over 70% of them had thoracoabdominal or juxtarenal aortic aneurysms and underwent complex aortic repair with f-bEVAR, which are known to carry a higher complication rate compared to infrarenal AAA, also evidenced by 37% of patients undergoing non-IBD aortic reinterventions. Along the same lines, aortic aneurysmal sac stability was relatively low (61%); however, 67% (6/9) of the patients with reinterventions have less than 1 year of follow-up after the last procedure, making it difficult to judge the actual aneurysm size after intervention.
We found, as previously published,20,21 a low rate of occlusions or EL within the IBD at follow-up.
We acknowledge several limitations of our study, including the retrospective nature of this series, a small cohort, and the different techniques used depending on the preferences of each surgeon, all of which may impact the results.
Despite a limited follow-up in our series, we could reaffirm the feasibility and excellent results, even when faced with difficult technical challenges such as the presence of a previous bifurcated open surgical or endovascular graft.
Conclusion
The implantation of IBD over previous aortic or open endografts involving the aortic bifurcation avoiding UEA is feasible and was not associated with complications due to the technique in this series. Based on our results, we suggest the utilization of a femoral approach as the primary access of choice.
Footnotes
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: T.K.—consultant, proctoring, IP, royalties, research, and travel grants with Cook Medical.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
