Comparison of fenestrated endovascular aneurysmrepair and chimney graft techniques for pararenalaortic aneurysmHiroshi Banno, MD, PhD, Frdric Cochennec, MD, Jean Marzelle, MD, andJean-Pierre Becquemin, MD, FRCS, Crteil, France
Objective: This study compared early-term and midterm results of endovascular repair (EVAR) of juxtarenal or pararenalaortic aneurysms (PAAs) using fenestrated stent grafting (f-EVAR) and the chimney grafting (c-EVAR) technique.Methods: Consecutive patients with PAAs who underwent f-EVAR using commercially available devices and c-EVAR in atertiary vascular center from January 2006 to April 2013 were evaluated, including a retrospective scrutiny and update ofa prospectively maintained database, calculation and comparison of perioperative mortality and morbidity, overall sur-vival, reintervention-free rate, branch event-free rate, reconstructed vessel patency, and collection of data about intra-operative events, perioperative complications and reinterventions, and midterm sac behavior.Results: During the study period, 80 patients (72 men) underwent f-EVAR and 38 (34 men) underwent c-EVAR. Allf-EVAR patients were operated on electively, whereas six c-EVAR patients (15.8%; P [ .002) were operated on in anemergent setting. The preoperative PAA diameter was significantly smaller in the f-EVAR group than in the c-EVARgroup (58.6 6 8.6 mm vs 65.9 6 15.3 mm; P [ .003). The mean number of reconstructed vessels per patient was 2.4 60.7 (median, two) for the f-EVAR group and 1.6 6 0.7 (median, one) for the c-EVAR group (P < .0001). The f-EVARand c-EVAR groups did not differ in 30-day mortality (6.3% vs 7.9%; P [ .71) or in moderate to severe complications(27.5% vs 39.5%; P [ 1.0). Median follow-up duration was 14 months, (range, 0-88 months) in the f-EVAR group and12 months (range, 0-48 months) in the c-EVAR group. After 2 years, estimated survival rates (77.3% vs 71.8%),reintervention-free rates (71.4% vs 72.0%), reconstructed vessel event-free rates (90.5% vs 84.1%), and primary patency ofreconstructed vessel rates (97.1% vs 87.6%) were not statistically different. During follow-up, sac shrinkage ($5 mm) wasobserved in 43.4% of f-EVAR patients and in 30.6% of c-EVAR patients (no statistical difference).Conclusions: In this limited retrospective series, short-term and midterm results of f-EVAR and c-EVAR were not sta-tistically different. c-EVAR could be an attractive option for patients not suitable for f-EVAR. (J Vasc Surg 2014;-:1-9.)
Until recently, the gold standard for treatment of a jux-tarenal or pararenal aortic aneurysm (PAA) was open surgi-cal repair (OSR).1-3 During the last 15 years, purelyendovascular abdominal aortic aneurysm (AAA) repair(EVAR) approaches have been developed to reduce post-operative mortality and morbidity. The feasibility of fenes-trated endografting (f-EVAR) for the treatment of complexAAAs is now widely recognized, with several large seriesconfirming satisfactory early-term and midterm results.4-6
However, the applicability of the technique is limited bystrict anatomic requirements, high costs, and lengthymanufacturing lead-times.
Greenberg et al7 first described the use of parallel renalstents combined with an aortic stent graft in patients with a
the Department of Vascular Surgery, Henri Mondor Hospital, Uni-rsity Paris XII.or conflict of interest: none.rint requests: Hiroshi Banno, MD, PhD, Department of Vascular Sur-ry, Henri Mondor Hospital, University Paris XII, 51 avenue du Mar-al De Lattre de Tassigny, 94010 Crteil, France (e-mail: firstname.lastname@example.org).editors and reviewers of this article have no relevant financial relationshipsdisclose per the JVS policy that requires reviewers to decline review of anyanuscript for which they may have a conflict of interest.-5214/$36.00yright 2014 by the Society for Vascular Surgery.://dx.doi.org/10.1016/j.jvs.2014.01.036
short proximal neck to obtain a more proximal seal. Sincethen, the chimney (c-EVAR) graft technique has been advo-cated as a possible endovascular alternative to OSR ofcomplex AAAs. One advantage of this technique is its imme-diate availability. However, uncertainties about long-termpatency of reconstructed visceral vessels and the risk of typeIa endoleak (EL) through the gutters located betweenthe aorta, the stents, and the stent graft remain a concern.
To date, only a few limited retrospective series havecompared f-EVAR and c-EVAR.8-10 The objective of thisretrospective study was to compare the results of f-EVARand c-EVAR for PAAs within a single institution.
Patients and inclusion criteria. The study includedall consecutive patients treated by f-EVAR or c-EVARwith commercially available stent grafts for PAA, fromJanuary 2006 to April 2013, in a tertiary vascular unit(Henri Mondor Hospital, Crteil, France). PAAs includedshort-necked infrarenal, juxtarenal, and PAAs consideredunsuitable for conventional EVAR. Exclusion criteriawere bailout chimney graft procedures for unexpectedcovering of one or both renal arteries during standardEVAR, f-EVAR with a physician-modified stent graft, andbranched or combined (ie, fenestrated and branched orfenestrated and chimney) stent grafts. Patients with failure
JOURNAL OF VASCULAR SURGERY2 Banno et al --- 2014
of cannulation through fenestration(s), followed by re-construction of visceral branch(es), by bailout chimneytechnique were included in the f-EVAR cohort.
In our institution, OSR or EVAR is scrutinized for allpatients with PAAs during a weekly multidisciplinarymeeting of vascular surgeons, interventional radiologists,and anesthesiologists. All patients were deemed at highrisk for OSR. The choice between f-EVAR and c-EVARwas determined by clinical and anatomic characteristics.In the current series, indications for f-EVAR met manufac-turers inclusion and exclusion criteria. Patients with anaortic angle of $45 were excluded. Most c-EVAR proce-dures were performed for a PAA associated with at leastone of the following conditions:
d Anatomical contraindications to f-EVAR, includingany or all of small or tortuous iliac arteries, severe iliacocclusive disease, proximal migration, type Ia EL afterprevious EVAR, or an angulated aorta;
d A patient with an aneurysm that was considered life-threatening (diameter >60 mm, rapidly expanding orpainful aneurysm) and who could not wait for a fenes-trated stent graft to be manufactured.
In the c-EVAR group, we limited reconstructions toone or two target vessels if possible. Patients who requiredreconstruction of more than two target vessels were consid-ered to be treated by another parallel technique, such assandwich technique.
Preoperative assessment and device sizing. All pa-tients underwent a high-resolution computed tomogra-phy (CT) scan preoperatively and before discharge.Procedure planning and device sizing were defined usingAquarius WS (TeraRecon Inc, Mateo, Calif), a dedicatedthree-dimensional vascular imaging workstation, withcenterline luminal reconstructions.
For f-EVAR cases, device designs suggested by theimplanting physicians complied with the manufacturersguidelines and were systematically reviewed and approvedby the planning center of the relevant device manufacturer.The sealing zone was defined as parallel neck of 15 mm inlength, whether or not it included the orifice of a visceralartery; thus, we used the aorta at the level of the fenestra-tion as a sealing zone if that segment was healthy.
To be eligible for c-EVAR, patients had to present asatisfactory (>15 mm in length) proximal landing zoneof healthy aorta below the proximal edge of the main de-vice. When visceral branches were covered, we recon-structed them by chimney technique.
Definitions. The proximal extent of the PAA was clas-sified according to American College of Cardiology/Amer-ican Heart Association 2005 practice guidelines for themanagement of patients with peripheral arterial disease.11
Preoperative coronary artery disease was defined as anabnormal result on a stress test or coronary angiogramand a history of myocardial infarction or open or percuta-neous coronary artery revascularization. Congestive heart
failure was clinically defined by past hospital admissionfor acute exacerbation or an ejection fraction of 160 mmol/L. Renal impairment wasdefined by a 50% increase in the creatinine value comparedwith baseline levels.
The definition of a target vessel was any vessel recon-structed using a bare-metal stent or a covered stent or ves-sel(s) failing to be reconstructed despite planning. Scallopsin f-EVAR were not considered indicative of target vessels.Target vessels reconstructed by a bailout chimney grafttechnique in the f-EVAR group were considered as a failureof revascularization and excluded from target vessel anal-ysis. Target vessel patency was defined as absence of targetvessels stenosis or occlusion requiring reintervention dur-ing follow-up. Reconstructed vessel events included targetvessel occlusion, any reintervention needed, or aneurysmsac growth induced by a branch-related EL (type III inf-EVAR or type I gutter in c-EVAR).
Device implantation. Technical aspects of f-EVARusing commercially available devices have been detailed inprevious reports.5,12,13 We have previously reported ourtechnique of chimney stent grafting.14,15 In short, whenonly one target artery was reconstructed, cannulation andcovered stent deployment were performed through a leftaxillary or a left brachial approach. At the beginning of ourexperience, when two or three target arteries were treated,we used two axillary/brachial arteries approaches to ensureaccess to at least two target arteries until the final angio-gram. As we gained