On-label Use of Commercially-available Abdominal Endografts for Para-anastomotic Aneurysms and Pseudoaneurysms After Infrarenal AbdominalAortic Aneurysm Open Repair

Y. Tshomba, D. Mascia *, A. Kahlberg, E.M. Marone, G. Melissano, R. Chiesa

Vascular Surgery, Vita-Salute University, San Raffaele Scientific Institute, Milan, Italy

* CoE-ma1078http:

WHAT THIS PAPER ADDS

Open repair of abdominal aortic para-anastomotic aneurysms (PAAs) and pseudoaneurysms (PSAs) afterinfrarenal abdominal aortic aneurysm open repair is burdened by increased challenges and complicationsrelated to the redo procedure. Complex endovascular techniques (fenestrated/branched, chimney, and peri-scope parallel grafts) have been developed in the last decade in order to treat these lesions; however, largerseries are currently still limited, long-term follow-up is poorly known, patient customization (making thetreatment unavailable in emergency) may be required, and treatment cost is often high. As a result, the standarduse of commercially-available abdominal endovascular devices is nowadays one of the most appealing options,routinely performed whenever possible in most centers. However, in the analysis of previous series of PAAs andPSAs treated with standard commercially-available abdominal endovascular devices, the anatomical inclusioncriteria were usually not specifically defined and the series included various off-label uses of the devices. In ourstudy, we specifically analyzed the feasibility and outcomes only of strict on-label use of the devices (OnL-endovascular aortic repair [EVAR]) for the treatment of abdominal PAAs and PSAs. Our analysis confirmed bothlimited feasibility of OnL-EVAR and high rates of late complications and reinterventions.

Objectives: To analyze feasibility and outcomes of endovascular aortic repair (EVAR) with a strictly on-label use ofabdominal aortic endografts (OnL-EVAR) to treat para-anastomotic aneurysms (PAAs) and pseudoaneurysms(PSAs) after infrarenal abdominal aortic aneurysm open repair (OR).Methods: The data of all consecutive patients treated between 1999 and 2012 for non-infected abdominal PAAsand PSAs at our center were prospectively collected. All cases fit for EVAR based on the instructions for use of aseries of abdominal aortic endografts commercially available during the study period were scheduled for OnL-EVAR regardless of patients’ surgical risk. Any patients unfit for OnL-EVAR underwent OR or other complexendovascular techniques.Results: One hundred and forty-three patients were collected; 78 underwent OR and 65 endovascular repair withdifferent strategies. Coil embolization, hybrid, and chimney/periscope grafts techniques were limited to sevenpatients unfit both for OR and OnL-EVAR. Inclusion criteria for OnL-EVAR were reached in 58 patients for anoverall OnL-EVAR feasibility of approximately 40% (21% for PAAs and 55% for PSAs). In particular, OnL-EVARfeasibility was 19% in case of involvement of proximal aortic anastomosis, 71% for distal aortic anastomosis, and80% for iliac arteries. Overall, 25 aortouniiliac and 11 bifurcated implants were performed, single proximal aorticcuffs were used in 10 patients, and iliac extension in 12. Primary technical success was 98% without perioperativemortality. At a median follow-up of 67 months (range: 1e144 months), cumulative aneurysm-related mortalitywas 7%, endograft migration 7%, and reintervention was 17%. Life-table analysis showed actuarial survival andfreedom from aneurysm-related death at 1, 3, and 5 years of 100%, 98%, and 95%, and of 100%, 98%, and 95%,respectively. Freedom from aortic reintervention or open conversion at 1, 3, and 5 years was 94%, 90%, and 85%.Conclusions: Feasibility of OnL-EVAR was limited for PAAs and PSAs, with a rate that was lower than 20% in caseof involvement of proximal aortic anastomosis. Aortouniiliac configuration was the most commonly feasibleimplant and, despite strict on-label use of abdominal devices, the rate of late complications and reinterventionswas high.� 2013 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved.Article history: Received 16 February 2013, Accepted 12 September 2013, Available online 19 September 2013

rresponding author. D. Mascia, Scientific Institute San Raffaele, Vascular Surgery, Università Vita-Salute, Via Olgettina, 60, 20132 Milan, Italy.il address: mascia.daniele@hsr.it (D. Mascia).-5884/$ e see front matter � 2013 European Society for Vascular Surgery. Published by Elsevier Ltd. All rights reserved.//dx.doi.org/10.1016/j.ejvs.2013.09.013

Table

GenAge

Man

Card

658 Y. Tshomba et al.

Keywords: Abdominal aortic aneurysm repair, Aortic endografts, Endovascular treatment, EVAR, Para-anastomotic aneurysm, Pseudoaneurysm

INTRODUCTION

Abdominal aortic aneurysm (AAA) open repair is generallyconsidered as safe and durable. However, this procedure isburdened by the risk of some long term complications, suchas para-anastomotic aneurysms (PAAs) and pseudoaneur-ysms (PSAs).1e9

PAAs and PSAs are defined as aneurysm formation nearthe anastomotic line either proximal or distal to the im-plantation of a graft. According to Haimovici,7 PAAs areenclosed by uninterrupted arterial wall components, andthey may represent a residual lesion or more likely theprogression of aneurysmal disease, while PSAs are charac-terized by breaks in the arterial wall with extravasation insurrounding tissues and the formation of a fibrous capsulethat gradually expands with blood pressure.

The PAAs and PSAs following AAA open repair range from0.5% to 15.0% in different series,8,9 and are both potentiallyfatal conditions as they may lead to aortic rupture,bleeding, or even infection and fistulization within theduodenum.1,5,6,8,9

Conventional treatment of para-anastomotic aneurysmsand PSAs is open surgical repair (OR) and is burdened withperioperative mortality ranging from 4.5% to 67.0%.10e13

Over the last two decades endovascular aortic repair(EVAR) has emerged as a promising alternative to OR withreduced early morbidity and mortality.14,15 However, thefeasibility of standard device-based endovascular technol-ogy with the on-label use of commercially-availableendografts (OnL-EVAR) is limited, especially for the prox-imity of renal arteries to the failed proximal aortic anasto-mosis. As a result, branched endovascular technology,16,17

chimney/periscope (CHIMPS),18,19 and “hybrid” tech-niques20,21 have been developed over the last few years in

1. Clinical manifestations and preoperative risk factors of on-lab

der Male(y) <55

55e6970e79>80

ifestations AsymptomaticExpansionCompression (GU tract)Compression (GI tract)Contained ruptureFree rupture

iac statusa

0 Asymptomatic, with nelectrocardiogram1 Asymptomatic but wmyocardial infarction byoccult myocardial infarctelectrocardiogram, or fixdipyridamole thallium or

order to treat abdominal aortic lesions near to renal andvisceral arteries; however, indications and long-term resultsare debated and costs are high.

The aim of this study is to investigate the feasibility rateand outcomes of OnL-EVAR for PAAs and PSAs after priorAAA open repair in a single-center setting.

METHODS

We performed a retrospective analysis of prospectivelycollected data of all consecutive patients who underwentopen or endovascular repair for PAAs or PSAs (aortic oriliac) following previous OR for atherosclerotic AAA. Allcases with suspected graft infection and/or aorto-entericfistula were excluded from the study. All cases availablefor OnL-EVAR based on the manufacturer’s instructions foruse (IFU) were scheduled for EVAR regardless of the surgicalrisk of the patient. Any patients unfit for OnL-EVAR weretreated with conventional OR whenever possible; casesunfit both for OR and OnL-EVAR were treated with alter-native strategies, such as coil embolization, CHIMPS, and“hybrid” techniques. No cases of custom-made fenestratedor branched endografts were used in the study period totreat PAAs or PSAs.

Data were collected from clinical and radiological records,and entered onto a computerized database. Demographics,clinical manifestations, and preoperative risk factors of pa-tients who underwent OnL-EVAR (summarized in Table 1)were reported according to the Society for Vascular SurgerySuggested Reporting Standards22 and to the Society forVascular Surgery/American Association for Vascular Surgerymedical comorbidity grading system.23

All patients underwent a preoperative contrast-enhancedcomputed tomography angiography (CTA). Preoperative

el endovascular aortic repair (OnL-EVAR) study group.

OnL-EVAR (%)39 (67)4 (7)8 (14)31 (53)15 (26)26 (45)24 (42)2 (3)1 (2)3 (5)2 (3)

ormal 3 (5)

ith either remotehistory (6 months),ion byed defect onsimilar scan

11 (19)

Table 1-continued

OnL-EVAR (%)2 Any one of the following: stable angina, noangina but significant reversible perfusion defecton dipyridamole thallium scan, significant silentischemia (1% of time) on Holter monitoring,ejection fraction 25e45%, controlled ectopy orasymptomatic arrhythmia, or history ofcongestive heart failure that is now wellcompensated

19 (32)

3 Any one of the following: unstable angina,symptomatic or poorly controlled ectopy/arrhythmia (chronic/recurrent), poorlycompensated or recurrent congestive heartfailure, ejection fraction <25%, myocardialinfarction within 6 months

25 (43)

Renal statusb

Stage 1 (GFR �90)(kidney damage with normal or [GFR)

11 (19)

Stage 2 (GFR 60e89)(kidney damage with mild YGFR)

29 (49)

Stage 3 (GFR 30e59)(moderate YGFR)

10 (18)

Stage 4 (GFR 15e29)(severe YGFR)

5 (9)

Stage 5 (GFR <15 of dialysis)(kidney failure)

3 (5)

Pulmonary status0 Asymptomatic, normal chest X-ray film, PFTs20%

4 (7)

1 Mild dyspnea on exertion or mild X-rayparenchymal changes, PFTs 65e80%

8 (14)

2 Between 1 and 3 20 (34)3 VC < 1.85 L, FEV1 <1.2 L or <35%,pCO2 > 45 mmHg, pulmonary hypertension

26 (45)

Hypertension0 (Diastolic <90 mmHg) 2 (3)1 (Easily controlled, single drug) 12 (20)2 (Requires two drugs) 38 (65)3 (>Two drugs or uncontrolled) 6 (10)

Note. GU tract ¼ genitourinary tract; GI tract ¼ gastrointestinal tract; GFR ¼ estimated glomerular filtration rate; PFTs ¼ pulmonaryfunction tests; VC ¼ vital capacity; FEV1 ¼ forced expiratory volume in 1 second; pCO2 ¼ partial pressure of CO2.a Major cardiac complication rates.b Stages of chronic kidney disease.

European Journal of Vascular and Endovascular Surgery Volume 46 Issue 6 p. 657e666 December/2013 659

planning to assess landing zones length, diameter andangulation, aortic bifurcation diameter, iliofemoral access,overall implant length, and presence of thrombus and cal-cifications was performed with analogic systems or a varietyof post-processing software until 2008, and then from 2009

Table 2. Instructions for use (IFUs) and frequency of employment of s

Stent-graft Brand n (%) cases in this series

Excluder Gore 3 (5)Zenith Flex Cook 32 (56)Zenith LP Cook 3 (5)Talent Medtronic 10 (17)Endurant Medtronic 7 (12)Vanguard Boston Scientific 3 (5)

OsiriX (Pixmeo, Bernex, Switzerland) using multiplanarreconstruction tools was employed.

All endovascular procedures were performed in theoperating room equipped for open surgical conversion usinga portable C-arm (Moonray Simad Medical Technology,

tent-grafts used in this series.

IFUAortic neck length (mm) Aortic neck angulation15 60�

15 60�

15 60�

10 60�

10/15 60�/75�

15 d

Table 3. Stent-graft used for para-anastomotic aneurysm (PAA)and pseudoaneurysm (PSA) treatment in the on-labelendovascular aortic repair (OnL-EVAR) group and type of surgicalreconstruction performed in the open surgical repair (OR) group.

PAA PSAOnL-EVAR (n [ 58)Bifurcated 1 10Aortouniliac 3 22Aortic cuff 5 5Iliac cuff 2 10OR (n [ 78)Aorto-aortic 19 14Aorto-bifemoral 21 24

660 Y. Tshomba et al.

Mirandola (Mo), Italy, or Ziehm-Vision RFD Ziehm Imaging,Nuremberg, Germany)).

Patients were approached in most cases through a uni- orbilateral groin incision. Extraperitoneal iliac access wasrequired in two cases (3%) for severely diseased iliac axis.

Adjunctive brachial or radial access was usually per-formed in case of involvement of proximal anastomosis forcatheterization of the lower renal artery in order to attemptrescue endovascular maneuvers in case of its unintentionalcovering.

Iopamidol 370 mg/ml (Iopamiro 370; Bracco, Milano,Italy) was used in all patients with serum creatinine<0.9 mg/dL, while Iodixanol 320 mgI/ml (Visipaque, GEHealthcare, Buckinghamshire, UK) was used in all patientswith serum creatinine >1 mg/dL, but renal status <2.18

Intravenous heparin (70 IU/kg) was administered afterintroducer insertion.

Implanted abdominal aortic commercially-available off-the-shelf endografts were

� Excluder (W.L. Gore & Associates, Flagstaff, AZ, USA);� Zenith (Cook Inc, Bloomington, IN, USA);

Figure 1. On the left is shown a preoperative computed tomographpseudoaneurysm following aorto-bi-iliac reconstruction; of note, bilatethe endovascular repair using an aortouniiliac stent graft with contral

� Talent abdominal and Endurant (Medtronic, Minneapolis,MN, USA);

� Vanguard (Boston Scientific/Scimed, Natick,Massachusetts).

Planned endograft oversizing was strictly determinedaccording to the manufacturer’s IFU. IFU and volume of usein the study period of all devices are reported in Table 2.The use of endografts with suprarenal fixation systems waslargely preferred in cases of proximal PAA and PSA(Table 2).

Endografts were always delivered over a stiff 0.03500

guidewire (Amplatz Super Stiff, Boston Scientific, Miami, FL,USA; Lunderquist Extra Stiff, Cook) with ballooning accord-ing to manufacturer’s IFU.

Completion angiography after stiff guidewire removalwas always performed to assess patency of the renal ar-teries and to discover endoleak or other technicalconcerns.

Patients were followed-up with CTA, duplex ultrasound,and clinical examination at 1 month and annually thereafter.

All-cause late mortality and complications were identifiedthrough a review of office charts and telephoneassessment.

Continuous variables were reported as means and SDs,and dichotomous variables were reported as percentages.The KaplaneMeier method was used to estimate survival.Analyses were run using IBM SPSS Statistics 18 software(SPSS, Chicago, IL, USA).

RESULTS

In our center 4,539 AAA open repairs (between 1993 and2012) and 835 EVAR (between 1999 and 2012) were per-formed overall. From 1999, 143 consecutive patientstreated for PAA or PSA were found. Proximal anastomosis

y scan with three-dimensional (3D) reconstruction of a multipleral hypogastric occlusion. On the right a postoperative 3D image ofateral occlusion and femoro-femoral bypass grafting.

Figure 2. Pre-operative computed tomography images and three-dimensional reconstruction of a ruptured proximal aortic pseudoa-neurysm after aorto-aortic reconstruction in a patient with bilateral double renal arteries.

European Journal of Vascular and Endovascular Surgery Volume 46 Issue 6 p. 657e666 December/2013 661

was involved in 95 (58%) lesions, aortic bifurcation in 38(23%) lesions, and iliac arteries in 32 (19%) lesions for atotal of 165 lesions in 143 patients. The index procedurewas performed in our center in 61% of cases. The mediandelay from index to redo procedure was 8 years (range: 0e18). Seventy-eight patients underwent OR (Table 3) and 65

Table 4. Intraoperative and early postoperative complications ofpatients who underwent on-label endovascular aortic repair(OnL-EVAR).

OnL-EVARTotal procedures 58Procedural time (mins) 65 (range 38e93)Blood loss (mL) 220 (range 50e350)ICU stay (h) 6 (range 0e35)Length of stay (days) 2 (range 1e4)Technical success 57 (98%)Type I endoleak 3 (5%)Type II endoleak 1 (1.7%)Perioperative mortality 0Respiratory failure 4 (7%)Renal failurea 6 (10%)AMI 3 (5%)Ileus 0Gluteal claudication 2 (3%)Hemorrhagic complications 0

Note. ICU ¼ intensive care unit; AMI ¼ acute myocardialinfarction.a Renal failure was doubling of preoperative serum creatinine.

endovascular repair with different strategies. Off-labelcomplex endovascular techniques were used in sevencasesdall PAAs involving the proximal aortic anastomosis(periscope or chimney grafts in six cases, and coil emboli-zation through a bare metal aortic stent in one case). In-clusion criteria for OnL-EVAR were reached in 58 patients

Figure 3. KaplaneMeier curve shows overall survival after on-labelendovascular aortic repair (OnL-EVAR) of para-anastomotic aneu-rysm and pseudoaneurysm repair.

Figure 4. KaplaneMeier curve shows freedom from aneurysm-related death after on-label endovascular aortic repair (OnL-EVAR) for all stent grafts that were used.

662 Y. Tshomba et al.

(Table 3), for an overall OnL-EVAR feasibility of 40% (21%for PAAs and 55% for PSAs). In particular, OnL-EVAR feasi-bility was 19% in the case of involvement of proximal aorticanastomosis, 71% for distal aortic anastomosis, and 80% foriliac arteries. The OnL-EVAR groups’ proximal anastomosiswas involved in 18 (26%) lesions, aortic bifurcation in 27(38%) lesions, and iliac arteries in 23 (36%) lesions for atotal of 70 lesions in 58 patients. Aortouniliac configurationwas used in 25 cases (Fig. 1), bifurcated in 11 cases (Fig. 2),aortic cuff in 10 cases, and iliac cuff in 12 cases (Table 3).

Four polar renal arteries were intentionally overstented,and no cases of unintentional main renal arteries over-stenting were reported. In most iliac PSAs or PAAs, anintentional overstenting of a hypogastric artery was per-formed, mainly after previous endovascular hypogastric

Figure 5. Freedom from aortic reintervention or open conversionafter on-label endovascular aortic repair (OnL-EVAR) for all stentgrafts used.

artery occlusion with an Amplatzer Vascular Plug (AGAMedical, Plymouth, MN, USA) in order to prevent type IIendoleak.

A type I endoleak was detected upon intraoperativeangiography immediately after endograft deployment inthree cases (5%) and it was corrected in two cases afterintraoperative endograft ballooning. A secondary procedureof cuff re-lining with a bare metal stent (Jotec E-XL,Hechingen, Germany) was performed on the patient withthe uncorrected type I endoleak after endovascular exclu-sion of a proximal symptomatic PAA by means of an aorticcuff. For persistent ruptured pseudoaneurysm perfusion,the patient underwent open conversion with cuff removaland aortic repair with a Dacron tube graft after 5 days. Inone case (2%), a type II endoleak was intraoperativelydetected but resolved spontaneously at 30 days (as seen onCTA). No perioperative mortality was recorded, and earlyresults and morbidity are summarized in Table 4. Inparticular, after OnL-EVAR, transient doubling of serumcreatinine never requiring dialysis was observed in six cases(10%) and perioperative bowel ischemia in three cases (5%).Bowel ischemia was suspected in any case of abdominalpain, peritoneal irritation, nausea, vomiting, and diarrheaassociated with positive laboratory tests (persistent arterialblood pH <7.34; plasma lactate level >2.1 mmol/L; whiteblood cell count >12,000/mL). In all the cases the clinicalsuspicion of bowel ischemia was confirmed at early post-operative colonoscopy and was medically resolved.

At a median follow-up of 67 months (range: 1e144months), the cumulative all-cause and aneurysm-relatedmortality was 15% (nine patients) and 7% (four patients),respectively. Life-table analysis showed actuarial survivaland freedom from aneurysm-related death at 1, 3, and 5years of 100%, 98%, and 95%, and of 100%, 98%, and 95%,respectively (Figs. 3 and 4). Freedom from aortic reinter-vention or open conversion at 1, 3, and 5 years was 94%,90%, and 85%, respectively (Fig. 5).

Late migration was observed in four single proximalaortic cuffs among 10 implanted, and in all cases a supra-renal fixation was present. The overall cumulative latereintervention rate was 17%, with a rate of 40% after theuse of proximal aortic cuff, 27% after the use of bifurcatedendografts, 8% after the use of aortouniiliac endografts and8% after the use of iliac extensions. Five patients weretreated for secondary type I endoleak, four for aorto-iliacmalperfusion, and one for secondary type III endoleak assummarized in Table 5. In particular, four late open surgicalprocedures were performed: two patients underwentemergency surgical iliofemoral thrombectomy and stenting;one patient underwent an extra-anatomic axillo-bifemoralbypass grafting; and one patient underwent cuff removaland juxtarenal aorto-bifemoral bypass graft through a redo-laparotomy (Fig. 6).

DISCUSSION

Open repair of PAAs and PSAs represents a challengingprocedure not only because re-dissection through scarred

Table 5. Cause of secondary failure and type of treatment of patients who underwent reintervention and/or conversion during follow-up.

Patient Treatment option of PAA/PSA Cause of secondary failure Treatment of secondary failure1 Proximal cuff Distal migration and type I EL Axillobifemoral bypass2 Proximal cuff Distal migration and type I EL Cuff removal and aortic repair3 Bifurcated Neck enlargement and proximal type I EL Proximal cuff4 Proximal cuff Distal migration and aortoiliac occlusion Aortouniiliac þ crossover5 Proximal cuff Distal migration and aortoiliac occlusion Aortouniiliac þ crossover6 Aortouni-iliac Iliac branch occlusion Iliac recanalization7 Bifurcated Iliac branch occlusion Iliac recanalization8 Aortouniiliac Modular de connection and type III EL Aortouniiliac relining9 Bifurcated Neck enlargement and proximal type I EL Aortouniiliac þ crossover10 Iliac extension Distal migration and type I EL Bifurcated implant

Note. PAA ¼ para-anastomotic aneurysm; PSA ¼ pseudoaneurysm; OR ¼ open surgical repair; EL ¼ endoleak.

European Journal of Vascular and Endovascular Surgery Volume 46 Issue 6 p. 657e666 December/2013 663

tissues with an usual large amount of inflammation iscommonly required,24e26 but also for specific issues relatedto the limited infrarenal neck length available in redoprocedure.

Standard technique for infrarenal AAA OR requires thejuxtarenal positioning of the aortic clamp with the anasto-mosis performed as near as possible to the lower renalartery. The rationale of this strategy is to perform theanastomosis on the aortic tissue in the strongest andhealthiest state possible.8 The usual thickness of theabdominal aortic clamps we used (Medicon,Wertheim “AT”clamp 24.5 cm [9 3/4

00]) ranged from 3 to 5 mm. This meansthat when a PAA and a PSA involves the proximal anasto-mosis, the infrarenal aortic neck is a few millimeters longerthan the clamp thickness. For this reason, the aortic neck isnot usually enough to allow a safe infrarenal redo-clampingand redo-anastomosis and often an inter- or suprarenal

Figure 6. Preoperative computed tomography scan with three-dimenspreviously treated for a proximal para-anastomotic aneurysm; of notedoaneurysm reperfusion (yellow arrow).

aortic clamping is required with increased risk of renal andvisceral complications.27,28

EVAR has emerged over the last two decades as a validand safe alternative to OR for AAA,29,30 and it would beparticularly appealing in cases of PAAs and PSAs, wherethere is the potential to avoid retroperitoneal re-dissectionand suprarenal clamping.

Also, in EVAR cases, the infrarenal neck length is a criticalissue and one of the main inclusion criteria for on-labelendograft use together with the neck angulation. The me-dian infrarenal neck length required by the commercially-available endografts used in our center in the studyperiod ranges from 10 to 15 mm (Table 2). This is the clearreason why, if proximal anastomosis during index openrepair were performed with aortic clamp in juxtarenal po-sition, the infrarenal neck length would not usually beenough for OnL-EVAR.

ional reconstruction of an aortic cuff distal migration in a patientin axial sections and in the multiplanar reconstruction the pseu-

664 Y. Tshomba et al.

In analyzing the results of previous series of PAA and PSAEVAR performed with standard commercially-available de-vices, the anatomical inclusion criteria were not clearlydefined, and various off-label uses of the devices wereincluded, such as thoracic endografts used in abdominal aortaor abdominal devices deployed in neck length shorter than IFUrecommendations.14,31e33 Consequently, and not surprisingly,Ten Bosch et al.31 have recently reported disappointing out-comes at a median follow up of 41 months with rates ofreintervention and conversion of 26.9% and 6.9%, respec-tively. Van Herwaarden et al.32 also reported a rate of con-version of 14.3% (all patients treated with tube grafts) at amedian follow up of 12 months, and Sachdev et al.14 reportedan endoleak rate of 11% within 1 month of procedure.14

In our study, we analyzed a series of patients who un-derwent PAA and PSA EVAR only through the strict on-labeluse of commercially-available abdominal devices. Short andangulated necks not matching the IFU of devices wereexcluded from our study. Our analysis confirmed limitedfeasibility of OnL-EVAR and, interestingly, reported unex-pectedly high rates of late complications and reinterven-tions, despite the strict on-label use of the devices. Theincreased risk of endograft migration from an “in-graft”polyester landing zone has already been reported31 andthis, together with the poor quality of the aortic wall andthe increased risk of neck enlargement that may be hy-pothesized in patients developing a PAA or a PSA, mayexplain these poor results. Neither device-related pro-pensity of late complications nor specific protection fromdevice migration of suprarenal fixation emerged in our se-ries, perhaps because of the small size of the sub-groups.However, poor outcomes with the use of single proximalaortic cuffs with especially high risk of migration (40%) werereported. In our series, the highest rate of on-label implantswas with aortouniiliac endografts. This approach wasneeded in most cases of previous aorto-bi-iliac or bifemoralreconstruction because in the index procedure a very shortmain graft body (2e3 cm) would not allow the deploymenteither of a cuff or of bifurcated endografts. In the cases ofPAA or PSA involving distal anastomosis in patients with aprevious tube-graft repair, an aortouniiliac implant wasfrequently needed for the tube graft diameter (wecommonly use a 14e16-mm diameter graft), which doesunusually allow for the correct opening and expansion oftwo iliac limbs of a bifurcated endograft. Some new,commercially-available bifurcated endografts allow for agood opening of both the iliac limbs, even within distalaortic diameter of 15 mm, and this could enlarge the in-dications to bifurcated endografts in the next series.

Instructions for use of most commercially-availableendografts have changed over the last decade, and inclu-sion criteria for OnL-EVAR in our study match the anatomicalcriteria and refer to the commercially-available devices at thetime of procedure, so that current technological improve-ments and indications were missed by our analysis. Even thelack of sophisticated CTA before 2008 and of the post-processing software, which are currently available, mayhave biased our study, with a reduced accuracy of

preoperative planning before 2008. Furthermore, additionaltools are now available that improve primary technical suc-cess of EVAR with standard devices in the case of short andangulated neck, such as the use of giant stents or endoan-chors, without the need for chimney techniques or fenes-trated endografts.34 Palmaz XL stent (Cordis, Miami Lakes, FL,USA) is a one-piece laser-cut stainless steel slotted tube,available in 40 and 50 mm lengths that can be expandedbetween 14 and 25 mm, and has been proposed to supportEVAR in difficult proximal necks.35 Self-expandable stents arealso now available on the market for endograft re-lining,combining the outward radial force of self expandable stentdesign with open cells. In particular, flexibility of these de-vices may help endograft remodeling in tortuous shortlanding zones.36 Endostaples or endoanchors are alsodesigned to replicate the function of an interrupted aorticsuture, and have been proposed and tested to performendograft fixation in angulated infrarenal aortic neck.37 Inthis setting the use of single cuff both to exclude PSAs or PAAsand to fix prior endografts may be disappointing. In thissetting the problem is that the prior endograftsmay persist tomigrate, leading to disconnection and huge type III endo-leaks. Securement of the endografts to the aortic wall, as wellas to the proximal cuff, has the potential to improve long-term results and should be investigated in the future.

In the meantime, fenestrated or branched endograftsimplantation has been proven to be safe and feasible inproximal PAAs and juxtarenal aneurysms, with several largeseries confirming satisfactory intermediate-term out-comes.38e40 However, these techniques requiremanufacturing delay of custom endografts, making themunsuitable for an emergency setting and are rather expen-sive.14,15,31,33 CHIMPS techniques also offer an appealingalternative based on already existing materials that areusually readily available, although mid- and long-term re-sults are yet to be evaluated.19,41,42

Regarding OnL-EVAR in the case of PAAs and PSAsinvolving iliac anastomosis, the rate of feasibility with stan-dard abdominal devices in our series was satisfactory,reaching 80%, mainly through the use of iliac limbs or ex-tensions. Only one late complication was reported in a pa-tient who had successfully undergone a secondaryendovascular procedure. However, this high rate of feasibilitywould have been strongly reduced in this subgroup of pa-tients if sparing of hypogastric arteries had been alwaysplanned. Overstenting of hypogastric arteries has been pre-viously reported to be related to risk of buttock claudicationand bowel ischemia,43e45 and in our series, three cases ofbowel ischemia, all medically resolved, were reported afteroverstenting of one hypogastric artery. It is well known howsafe and effective specifically-designed branched endog-rafts46 are in both aneurysm exclusion and saving hypogastricartery patency with reasonable costs and logistics.

CONCLUSION

Despite reduced anatomical feasibility, the use ofcommercially-available abdominal endovascular devices to

European Journal of Vascular and Endovascular Surgery Volume 46 Issue 6 p. 657e666 December/2013 665

treat proximal PAAs and PSAs after AAA open repair is avery appealing option nowadays, routinely performedwhenever possible in most centers. A deeper understandingof early and late outcomes, and possible complications ofthese procedures is therefore mandatory, and based onearly and late results reported in the present study, acareful follow up would be necessary for the high risk of latecomplications and reinterventions, even after the strict on-label use of abdominal devices in these patients.

CONFLICT OF INTEREST

None.

FUNDING

None.

REFERENCES

1 Lorentzen JE, Nielsen OM, Arendrup H, Kimose HH, Bille S,Andersen J, et al. Vascular graft infection: an analysis of sixty-two graft infections in 2411 consecutively implanted syntheticgrafts. Surgery 1985;98:81e6.

2 Calligaro KD, Veith, Yuan JG, Gargiulo NJ, Dougherty MJ.Intraabdominal aortic graft infection: complete or partial graftpreservation in patients at very high risk. J Vasc Surg 2003;38:1199e205.

3 Ricco J-B. InterGard silver bifurcated graft: features and resultsof a multicenter clinical study. J Vasc Surg 2006;44:339e46.

4 Hallett Jr JW, Marshall DM, Petterson TM, Gray DT, Bower TC,Cherry Jr KJ, et al. Graft-related complications after abdominalaortic aneurysm repair: reassurance from a 36-year population-based experience. J Vasc Surg 1997;25:277e86.

5 Conrad MF, Crawford RS, Pedraza JD, Brewster DC,Lamuraglia GM, Corey M, et al. Long-term durability ofopen abdominal aortic aneurysm repair. J Vasc Surg 2007;46:669e75.

6 Biancari F, Ylönen K, Anttila V, Juvonen J, Romsi P, Satta J, et al.Durability of open repair of infrarenal abdominal aortic aneu-rysm: a 15-year follow-up study. J Vasc Surg 2002;35:87e93.

7 Haimovici’s vascular surgery. 5th ed., 45; 2004. p. 763e76.8 Szilagy DE, Smith RF, Elliott JP, Hageman JH, Dall’Olmo CA.Anastomotic aneurysm after vascular reconstruction:problems of incidence, etiology, and treatment. Surgery1975;78:800e16.

9 Edwards JM, Teefey SA, Zierler RE, Kohler TR. Intraabdominalparaanastomotic aneurysms after aortic bypass grafting. J VascSurg 1992;15:344e50.

10 Matsumura JS, Pearce WH, Cabellon A, McCarthy 3rd WJ,Yao JS. Reoperative aortic surgery. Cardiovasc Surg 1999;7:614e21.

11 Treiman GS, Weaver FA, Cossman DV, Foran RF, Cohen JL,Levin PM, et al. Anastomotic false aneurysms of the abdominalaorta and the iliac arteries. J Vasc Surg 1988;8:268e73.

12 Hagino RT, Taylor SM, Fujitani RM, Mills JL. Proximal anasto-motic failure following infrarenal aortic reconstruction: latedevelopment of true aneurysms, pseudoaneurysms, andocclusive disease. Ann Vasc Surg 1993;7:8e13.

13 Mulder EJ, van Bockel JH, Maas J, van den Akker PJ, Hermans J.Morbidity and mortality of reconstructive surgery of nonin-fected false aneurysms detected long after aortic prostheticreconstruction. Arch Surg 1998;133:45e9.

14 Sachdev U, Baril DT, Morrissey NJ, Silverberg D, Jacobs TS,Carroccio A, et al. Endovascular repair of para-anastomoticaortic aneurysms. J Vasc Surg 2007;46:636e41.

15 Gawenda M, Zachringer M, Brunkwall J. Open versus endo-vascular repair of para-anastomotic aneurysms in patients whowere morphological candidates for endovascular treatment.J Endovasc Ther 2003;10:745e51.

16 Austermann M, Donas KP, Panuccio G, Troisi N, Torsello G.Pararenal and thoracoabdominal aortic aneurysm repair withfenestrated and branched endografts: lessons learned andfuture directions. J Endovasc Ther 2011;18:157e60.

17 Hosokawa H, Iwase T, Sato M, Yoshida Y, Ueno K, Tamaki S,et al. Successful endovascular repair of juxtarenal and supra-renal aortic aneurysms with a branched stent graft. J Vasc Surg2001;33:1087e92.

18 Pecoraro F, Pfammatter T, Mayer D, Frauenfelder T,Papadimitriou D, Hechelhammer L, et al. Multiple periscopeand chimney grafts to treat ruptured thoracoabdominal andpararenal aortic aneurysms. J Endovasc Ther 2011;18:642e9.

19 Moulakakis KG, Mylonas SN, Avgerinos E, Papapetrou A,Kakisis JD, Brountzos EN, et al. The chimney graft technique forpreserving visceral vessels during endovascular treatment ofaortic pathologies. J Vasc Surg 2012;55:1497e503.

20 Böckler D, Schumacher H, Klemm K, Riemensperger M,Geisbüsch P, Kotelis D, et al. Hybrid procedures as a combinedendovascular and open approach for pararenal and thor-acoabdominal aortic pathologies. Langenbecks Arch Surg2007;392:715e23.

21 Chiesa R, Tshomba Y, Melissano G, Marone EM, Bertoglio L,Setacci F, et al. Hybrid approach to thoracoabdominal aorticaneurysms in patients with prior aortic surgery. J Vasc Surg2007;45:1128e35.

22 Johnston KW, Rutherford RB, Tilson MD, Shah DM, Hollier L,Stanley JC. Suggested standards for reporting on arterial an-eurysms. J Vasc Surg 1991;13:452e8.

23 Chaikof EL, Fillinger MF, Matsumura JS, Rutherford RB,White GH, Blankensteijn JD, et al. Identifying and grading fac-tors that modify the outcome of endovascular aortic aneurysmrepair. J Vasc Surg 2002;35:1061e6.

24 Tshomba Y, Bertoglio L, Marone EM, Melissano G, Chiesa R.Visceral aortic patch aneurysm after thoracoabdominal aorticrepair: conventional vs hybrid treatment. J Vasc Surg 2008;48:1083e91.

25 Tshomba Y, Melissano G, Civilini E, Setacci F, Chiesa R. Fate ofthe visceral aortic patch after thoracoabdominal aortic repair.Eur J Vasc Endovasc Surg 2005;29:383e9.

26 Melissano G, Di Mario C, Tshomba Y, Civilini E, Gimelli G,Nicoletti R, et al. Endovascular treatment of a noninfectedanastomotic juxtarenal aortic aneurysm. Tex Heart Inst J2000;27:408e11.

27 Coselli JS. Strategies for renal and visceral protection in thor-acoabdominal aortic surgery. J Thorac Cardiovasc Surg2010;140(Suppl. 6):S147e9.

28 Chong T, Nguyen L, Owens CD, Conte MS, Belkin M. Suprarenalaortic cross-clamp position: a reappraisal of its effects onoutcomes for open abdominal aortic aneurysm repair. J VascSurg 2009;49:873e80.

29 EVAR Trial Participants. Endovascular aneurysm repair versusopen repair in patients with abdominal aortic aneurysm (EVARtrial 1): randomised controlled trial. Lancet 2005;365:2179e86.

30 Prinssen M, Verhoeven EL, Buth J, Cuypers PW, vanSambeek MR, Balm R, et alDutch Randomized EndovascularAneurysm Management (DREAM) Trial Group. A randomized

666 Y. Tshomba et al.

trial comparing conventional and endovascular repair ofabdominal aortic aneurysms. N Engl J Med 2004;351:1607e18.

31 Ten Bosch JA, Waasdorp EJ, de Vries JP, Moll FL, Teijink JA, vanHerwaarden JA. The durability of endovascular repair of para-anastomotic aneurysms after previous open aortic recon-struction. J Vasc Surg 2011;54:1571e8.

32 Van Herwaarden JA, Waasdorp EJ, Bendermacher BL, van denBerg JC, Teijink JA, Moll FL. Endovascular repair of para-anastomotic aneurysms after previous aortic prostheticreconstruction. Ann Vasc Surg 2004;18:280e6.

33 Piffaretti G, Tozzi M, Lomazzi C, Rivolta N, Caronno R, Castelli P.Endovascular treatment for para-anastomotic abdominal aorticand iliac aneurysms following aortic surgery. J Cardiovasc Surg(Torino) 2007;48:711e7.

34 Ghouri M, Krajcer Z. Endoluminal abdominal aorticaneurysm repair. The latest advances in prevention of distalendograft migration and type 1 endoleak. Tex Heart Inst J2010;37:19e24.

35 Byrne J, Mehta M, Dominguez I, Paty PS, Roddy SP, Feustel P,et al. Does Palmaz XL stent deployment for type 1 endoleakduring elective or emergency endovascular aneurysm repairpredict poor outcome? A multivariate analysis of 1470 pa-tients. Ann Vasc Surg 2013;27:401e11.

36 Chisci E, Ventoruzzo G, Alamanni N, Bellandi G, Michelagnoli S.Transrenal E-XL stenting to resolve or prevent type Iaendoleak in the case of severe neck angulation during endo-vascular abdominal aortic aneurysm repair. J Vasc Surg2013;57:1383e6.

37 Avci M, Vos JA, Kolvenbach RR, Verhoeven EL, Perdikides T,Resch TA, et al. The use of endoanchors in repair EVAR cases toimprove proximal endograft fixation. J Cardiovasc Surg (Torino)2012;53:419e26.

38 Verhoeven EL, Vourliotakis G, Bos WT, Tielliu IF, Zeebregts CJ,Prins TR, et al. Fenestrated stent grafting for short-necked and

juxtarenal abdominal aortic aneurysm: an 8-year single-centreexperience. Eur J Vasc Endovasc Surg 2010;39:529e36.

39 Beck AW, Bos WT, Vourliotakis G, Zeebregts CJ, Tielliu IF,Verhoeven EL. Fenestrated and branched endograft repair ofjuxtarenal aneurysms after previous open aortic reconstruc-tion. J Vasc Surg 2009;49:1387e94.

40 Scurr JR, Brennan JA, Gilling-Smith GL, Harris PL,Vallabhaneni SR, McWilliams RG. Fenestrated endovascularrepair for juxtarenal aortic aneurysm. Br J Surg 2008;95:326e32.

41 Bruen KJ, Feezor RJ, Daniels MJ, Beck AW, Lee WA. Endovas-cular chimney technique versus open repair of juxtarenal andsuprarenal aneurysms. J Vasc Surg 2011;53:895e904.

42 Lee JT, Greenberg JI, Dalman RL. Early experience with thesnorkel technique for juxtarenal aneurysms. J Vasc Surg2012;55:935e46.

43 Mehta M, Veith FJ, Ohki T, Cynamon J, Goldstein K, Suggs WD,et al. Unilateral and bilateral hypogastric artery interruptionduring aortoiliac aneurysm repair in 154 patients: a relativelyinnocuous procedure. J Vasc Surg 2001;33:S27e32.

44 Lee CW, Kaufman JA, Fan C, Geller SC, Brewster DC,Cambria RP, et al. Clinical outcome of internal iliacartery occlusions during endovascular treatment of aor-toiliac aneurysmal diseases. J Vasc Interv Radiol 2000;11:567e71.

45 Karch LA, Hodgson KJ, Mattos MA, Bohannon WT, Ramsey DE,McLafferty RB. Adverse consequences of internal iliac arteryocclusion during endovascular repair of abdominal aortic an-eurysms. J Vasc Surg 2000;32:676e83.

46 Parlani G, Verzini F, De Rango P, Brambilla D, Coscarella C,Ferrer C, et al. Long-term results of iliac aneurysmrepair with iliac branched endograft: a 5-year experienceon 100 consecutive cases. Eur J Vasc Endovasc Surg2012;43:287e92.