European Journal of Vascular & Endovascular Surgery
Volume 32, Issue 2 , Pages 115-123, August 2006

A Prospective Analysis of Fenestrated Endovascular Grafting: Intermediate-term Outcomes

The Department of Vascular Surgery, Cleveland Clinic Foundation, Cleveland, OH, USA

Accepted 25 January 2006. published online 05 April 2006.

Article Outline

Abstract 

Purpose

To assess the intermediate-term outcomes following fenestrated grafting for juxtarenal aneurysms.

Materials and methods

A prospective trial was conducted on patients with short proximal necks, who were considered to be high-risk for open repair and unacceptable for conventional endovascular repair. Devices were designed from reconstructed CT data. Follow-up studies included CT, duplex ultrasound, and KUB and occurred at hospital discharge, 1, 6, and 12 months and annually thereafter.

Results

One hundred and nineteen patients were treated (2001–2005). Mean age and aneurysm size were 75 years and 65mm, respectively, and 82% were male. A total of 302 visceral vessels were inferior to the fabric seal (a mean of 2.5 vessels per patient), with the most common design incorporating two renal arteries and the SMA (58%). All prostheses were implanted successfully without any acute visceral artery loss. The mean follow-up was 19 months (0–42 months). One patient died within 30 days of device implantation. Kaplan–Meier estimates of survival at 1, 12, 24, and 36 months are 0.99, 0.92, 0.83 and 0.79. There were no ruptures or conversions. Pre-discharge imaging noted 11 type I and type III endoleaks. The 30-day endoleak rate was 10% (all type II). Aneurysm sac size decreased (>5mm) in 51, 79 and 77% at 6, 12 and 24 months, respectively. One patient had sac enlargement within the first year, associated with a persistent type II endoleak. In-stent stenoses occurred in 12 renal arteries and one SMA. Six renal arteries and the SMA stenosis were treated and two renal stenoses are awaiting treatment. Ten of 231 stented renal arteries occluded (three prior to discharge), one of which was recanalized. One component separation was treated with an extension at 2 years.

Conclusions

The placement of endovascular prostheses with graft material incorporating the visceral arteries is safe and appears to be effective at preventing rupture. Continued follow-up to assess the long-term benefit, aneurysm sac behavior and effect of stenting upon the visceral ostia remains critical.

Keywords: AAA, Aortic aneurysm, Endovascular graft, EVAR, Renal Stent, Juxtarenal aneurysm, Fenestrated endograft

 

Back to Article Outline

1. Introduction 

Patients presenting with abdominal aortic aneurysms (AAA) being considered for endovascular aneurysm repair are largely selected on anatomical criteria, including proximal and distal neck diameter, length, angulation and shape.1, 2 Compromised proximal neck anatomy is the most frequent rejection criteria for treatment with endovascular prosthesis. Although more robust proximal fixation systems, such as those with barbs or uncovered suprarenal stents, may provide improved device stability yet the need for a sealing zone above the aneurysm must be considered independent of the fixation. Maximizing the sealing zone with conventional prostheses requires accurate placement of the endoprosthesis into the immediate infrarenal neck. Challenging proximal necks (such as those with thrombus or surface irregularities), limitations in visualizing the visceral vessel origin, as well as technical issues associated with prosthesis deployment serve to limit the proximity of the proximal graft material with respect to the renal arteries thus, limiting the length of the intended sealing zone. Furthermore, the behavior of sealing stents within a non-parallel or irregular proximal neck is suboptimal. These issues represent the impetus for the development of devices capable of sealing above the renal arteries in the setting of compromised proximal sealing zones.

The first reports of fenestrated devices were published in 1999,3, 4 and this was followed by the development of a more versatile fenestrated device by Anderson, Lawrence-Brown and Hartley.5, 6 Additional clinical experiences have been reported.7, 8, 9 This paper updates our previously reported experiences8, 10, 11 including a substantially greater number of patients with extended follow-up.

Back to Article Outline

2. Materials and Methods 

Patients were prospectively enrolled in a physician-sponsored investigational device exemption protocol beginning in August 2001. For the purpose of this publication, patient accrual terminated in June 2005 with follow-up through August of 2005. All patients were deemed physiologically high-risk for open surgical repair, and anatomically unsuitable for traditional infrarenal endovascular grafts. Informed consent, approved by our Institutional Review Board, was obtained for all research subjects.

The pre-operative assessment of the aortic morphology centered on high-resolution spiral CT-scans incorporating the distal descending thoracic aorta through the profunda femoral artery. Arteriography was utilized sparingly, in general reserved for patients with concomitant occlusive disease of the visceral or pelvic vasculature. The Zenith™ device (Cook Inc., Bloomington, Indiana) formed the foundation of the fenestrated graft. The material construct and delivery system are similar to the Zenith™ which has been previously described.10, 12, 13 Device planning required accurate information regarding anatomic measurements including the ostial diameter of each visceral vessel, their relative distances from a fixed landmark (typically the superior mesenteric artery), and their radial orientation. Fenestrations were constructed to match the ostial diameter and maximize the sealing zone. Small fenestrations had a diameter between 6 and 8mm, and were located at a minimum of 15mm distal to the top of the fabric top. The ostia for the small fenestrations were placed between stent struts of the aortic device, to allow unimpeded access into the visceral artery, and intended for use in conjunction with a balloon expandable stent within the target vessel. Large fenestrations had greater diameters (8–12mm) with the potential of a strut crossing the fenestration, and thus not intended for use with additional stents. Scallop fenestrations were hemi-ovals 6–12mm in height and a 10mm diameter located at the most proximal portion of the fabric.

2.1. Implantation 

Patients were anticoagulated with heparin (100units/kg) and activated clotting times were maintained at greater than 300s using additional heparin. A stiff wire was placed into the ascending aorta through the intended main body delivery side, while the contralateral side was double punctured, or a large diameter valved sheath (Check Flo, Cook Inc.) was placed allowing multiple sheaths (2–3) to be placed within punctures made to the external valve. A multisidehole flush catheter was placed through one of the sheaths to visualize the renal arteries intermittently using small bursts (5–7cm3 at 30cm3/s) of non-ionic contrast. The proximal tubular fenestrated endograft component was oriented using gold markers located on the device body and around each fenestration. Partial deployment of this component was followed by minute adjustments to properly orient the fenestrations with their respective ostia (Fig. 1). Access to main body was achieved from both contralateral access ports, using selective catheterization techniques. Visceral vessel access was then established through the fenestrations, following which sheaths or guides were placed into the desired arteries. The posterior wire constraining the device to a lesser diameter was then removed allowing the graft to fully expand and the top cap was released. Balloon expandable stents were chosen with a minimum of 17mm in length and mounted upon a balloon sized to the diameter of the intended visceral vessel. The stents were deployed such that approximately 2/3 of the stent was within the visceral vessel, and just less than 1/3 extended into the aorta (Fig. 2). The aortic portion of the stent was then flared with a 10mm balloon and selectively flared further using a compliant latex balloon. The second (bifurcated) component was implanted ensuring a minimum of two stents of overlap with the tubular component. The remainder of the deployment did not differ from previously described publications.8

  • View full-size image.
  • Fig. 1. 

    The aortic portion of the stent was initially flared with a 10mm balloon and subsequently with a latex compliant balloon. This had the effect of both aligning and maximizing the area of fenestration to wall contact.

  • View full-size image.
  • Fig. 2. 

    Thirty endoleaks were visualized on the pre-discharge CT scan including seven type I and four type III endoleaks. The rest were type II endoleaks. Our practice of aggressively treating proximal type I and type III endoleaks in the post-operative accounts for the low incidence of late type I endoleaks.

2.2. Patient follow-up 

Post-operative evaluations were conducted at hospital discharge, 1, 6, and 12 months and yearly thereafter. Helical CT scans, duplex ultrasound (with the exception of the discharge time point), creatinine assessment, and flat plate radiography were obtained. Secondary interventions were performed in the setting of a suspected type I or III endoleak, selectively for type II endoleaks, compromised visceral vessel flow, or aneurysmal growth. When appropriate, outcome analyses were conducted in accordance with the reporting standards for endovascular aneurysm repair.14 Elevation of serum creatinine was considered significant if the serum value increased to greater than 2mg/dl (180μM) or a rise of >30% from baseline levels was noted. Duplex determined end systolic and diastolic velocities were collected for each visceral vessel treated at each follow-up visit.

Back to Article Outline

3. Results 

A total of 119 patients were treated. There were 98 men and 21 women with a mean age of 75±7 years. The pre-operative risk factors are listed in Table 1. The mean diameter of the proximal neck was 26±3.5mm (range, 17–34mm). The mean proximal neck length was 8±4mm (range, 3–18mm). The proximal neck length was <10mm in 70 patients, and between 10 and 18mm in 49 patients, all of which had morphologic factors implying compromised sealing or fixation. The shape of the neck was considered to have an inverted funnel (n=23), funnel (n=23), irregular (n=31), or parallel (n=42) shape. Mean maximum transverse diameter (MTD) of the AAA was 65±11mm (range, 46–102mm). Regional anesthesia was used in 97 cases and general anesthesia in 22 cases. A total of 302 visceral vessels (Fig. 3) were incorporated in the prosthesis design. Table 2 demonstrates the distribution of renal, superior mesenteric and celiac arteries treated.

Table 1. Pre-operative co-morbidities or risk factors for the 119 patient cohort
Patients (n)%
Congestive heart failure1714
Arterial hypertension9479
CABG/coronary stenting5849
Previous MI5345
Renal insufficiency (creatinaemia>120μmol/l)3126
Diabetes2219
COPD7462
Previous stroke2924
  • View full-size image.
  • Fig. 3. 

    Mean aneurysm size decreased from a 65mm pre-op. to 50mm at 3 years. Looked at from a different perspective 51% of patients had a decrease in aneurysm size of 5mm or more at 6 months and this increased to almost 80% at 3 years.

Table 2. This table depicts the distribution of vessels incorporated into the fenestrated devices used in this series
Mesenteric fenestrationsNumber of renal fenestrations
123
NONE7360
SMA5691
SMA+CELIAC010

The vast majority of patients (69%) had devices involving two main renal arteries and the SMA. A single patient had a large accessory renal that was preserved with a third renal fenestration, and two patients had renal arteries that were markedly disparate, requiring only single fenestrations. Only one patient required incorporation of all four visceral vessels.

3.1. Procedure 

All prostheses were successfully implanted. A conduit to the common iliac artery (10mm diameter polyester vascular graft) was planned and performed in six patients. Technical success was achieved in all cases although in one case a single renal vessel was not successfully catheterized and stented. This vessel was patent at the completion of the procedure (angiographically), however, it occluded prior to the pre-discharge CT scan. It was located within a tortuous portion of a short proximal neck. Procedure time averaged 227±76min with a mean of 56±22min of fluoroscopy time. The mean volume of contrast utilized was 179±53ml. The mean blood loss was 739ml.

The use of a large balloon expandable stent (Palmaz stent, Cordis, Great Lakes, NJ) immediately below the renal arteries was required in 23 patients to supplement the seal, and a single patient was treated with a proximal aortic cuff for a type I endoleak intraoperatively. Similarly, balloon expandable stents were also deployed at the junction of the fenestrated component and the bifurcated component in 14 patients and extension grafts in two patients to treat modular joint leaks intraoperatively (Fig. 4). Additional self-expanding stents were implanted into seven iliac limbs to alleviate kinking or unacceptable tortuosity following endograft deployment. One patient required a femoral endarterectomy and prosthetic patch closure on the side of the main delivery system insertion.

  • View full-size image.
  • Fig. 4. 

    Kaplan–Meier estimate of the survival function for all-cause mortality, with point-wise 95% confidence limits; the ticks mark the censoring times. The survival function gives, at each possible time from implantation, the probability of surviving beyond that time.

3.2. Early follow-up (<30 days) 

One patient died within 30 days (day 7) of the initial procedure. This patient had bilateral common iliac aneurysms and an attempted hypogastric bypass was aborted due to difficult pelvic access (in a morbidly obese male). The patient initially fared well; however, he developed an ileus and suffered an aspiration pneumonia, which further compromised his chronic obstructive pulmonary disease. He ultimately died of sepsis and multisystem organ failure.

3.2.1. Endoleak analysis 

The post-procedural CT scan depicted a proximal type I endoleak in six patients, a distal type I endoleak in one patient, a type III endoleak in four patients and a type II endoleak in 19 patients (Fig. 1). Four of six patients with proximal type I endoleaks underwent post-operative secondary procedures prior to hospital discharge. Three were treated with a giant Palmaz stent placed in the immediate infrarenal segment. The first patient also required the placement of a balloon expandable stent, which was placed within a right renal artery that was not involved in the initial fenestrated design. Another patient had a Jomed covered stent placed into a left renal artery that was stented in the initial fenestrated procedure, yet had a persistent leak through the fenestration into the aneurysm sac. The fourth patient did not have anatomy amenable to additional stenting and, therefore, underwent glue embolisation of the leak within the sac. All four treated endoleaks resolved and none have recurred to date. Two patients with proximal type I endoleaks were not treated prior to hospital discharge and both resolved by the 1 month CT scan. One sac remains excluded through 1 year of follow-up, while the other sac demonstrated recurrence of the leak at the 12-month imaging study. The patient then underwent a secondary procedure involving placement of a proximal graft extension and restenting of the left renal artery. The endoleak resolved and has not recurred at the 24-month imaging study. A distal type I endoleak at the left iliac limb was treated with coil embolisation of the hypogastric artery and an extension limb to the external iliac artery. Although frequently leakage was visualized at the junction of the tubular fenestrated and bifurcated components during completion angiography, only four of the untreated joint leaks remained after intraoperative treatment to the pre-discharge CT scan. Three of these were treated prior to hospital discharge. One with an aortic extension cuff placed at the overlap segment, the second a giant Palmaz stent, and the third patient with a converter device followed by a a femoral–femoral bypass when the endoleak failed to resolve following the placement of two giant Palmaz stents. All of these leaks were not present on later follow-up (6 months, 2 years and 1 month). A fourth type III endoleak was not treated, resolved within 1 month and has not recurred through 1-year follow-up. No type II endoleaks were treated in the immediate post-operative period. An additional six secondary procedures were performed in the early post-operative period including a brachial artery repair, repair of a femoral artery pseudo aneurysm, a brachial artery thrombectomy, a femoral–distal bypass and hallux amputation for a distal embolic events, and evacuation of a groin hematoma.

Three stented renal arteries occluded prior to the pre-discharge CT scan. In the first patient with a severely angulated neck, the inability to gain wire access to the renal artery precluded stenting of the vessel. In the second patient a dissection occurred during placement of the renal stent within a severely diseased renal artery. Although recognized and immediately treated with more distal stent placement, the vessel thrombosed. The third patient had an uncomplicated procedure but post-operatively was noted to have a rising lactate and underwent angiography, which revealed a patent SMA and occluded right renal artery. A fourth patient had patent renals at the pre-discharge CT scan, yet occluded a renal artery when he represented 6 weeks after implantation with a ruptured thoracic aortic aneurysm and profound hypotension. The thoracic rupture was successfully treated with another endoprosthesis.

3.3. Late follow-up (>30 days) 

The mean follow-up was 19 months (range 0–48 months). Sixteen patients died during the follow-up period, one within 30 days, seven within the first year, and eight patients after 12 months of follow-up. Kaplan–Meier estimates of survival at 1, 12, 24, and 36 months are 0.99, 0.92, 0.83 and 0.79, respectively (Fig. 2). There were no late aneurysm related deaths and no conversions to open procedures.

The 30-day endoleak rate was 10% (all type II endoleaks). Sac shrinkage (>5mm) was noted in 51% (40/78) of patients at 6 months, in 79% (41/52) at 12 months and in 77% (23/30) of patients at 2 years. Mean aneurysm size decreased from 65mm post-op. to 50mm at 3 years (Fig. 3). One patient demonstrated a 5mm growth of his aneurysm at 1 year. This patient had a type II endoleak at 1 month which was not seen at 6 months but which was again visualized at the 12-month follow-up visit. The patient refused treatment. Other type II endoleaks were treated selectively in four patients at 1, 1, 1 and 6 months, with glue embolisation of the IMA and sac.

3.4. Renal function 

Forty-five patients (42%) had pre-operative creatinine >1.2mg/dl. Thirty patients (25%) developed at least a transient increase in the serum creatinine of >30% over the course of the follow-up period. Five of the six renal artery stenoses treated to date were associated with a transient rise in serum creatinine that universally improved following the secondary intervention. All of the 10 patients with renal artery occlusions demonstrated an increase in serum creatinine, four of which ultimately required transient or permanent dialysis. One of these patients with a right renal artery occlusion diagnosed at the 1 month imaging study was recanalised and restented but the patient remains on dialysis.

A fifth patient who did not have a renal artery occlusion but had chronic renal insufficiency pre-operatively developed acute renal failure post-operatively and expired prior to his 6-month review.

3.5. Visceral vessel patency 

In addition to the renal artery occlusions (n=10) and stenoses (n=12) previously described (Fig. 4), a single SMA stenosis manifested as persistent post-prandial abdominal pain within 30 days following the procedure. Although no lesion was visualized by angiography, intravascular ultrasonography performed demonstrated fabric material partially obstructing the SMA origin. This was treated with angiography and stenting of the SMA. Duplex ultrasonography at 6 months demonstrated no evidence of restenosis and the patient is well. Duplex ultrasonography depicted elevated velocities at the origin of the SMA in a further two patients at 1 and 2 years post-procedure. Both patients are completely asymptomatic and in one patient angiography for a persistent type II endoleak did not visualize a stenosis of the SMA. The second patient continues to be observed.

Back to Article Outline

4. Discussion 

Fenestrated endovascular grafts are designed to extend the range of use of endovascular prosthesis by moving the proximal sealing zone from the infrarenal segment to the juxta and supra-renal aorta thereby providing clinicians with the ability to treat short infrarenal necks, properly align sealing zones in the setting of angulated anatomy or complex neck morphology. The fenestrations are necessary to incorporate visceral vessels that encroach on the desired sealing zone. Despite this ability, a segment of relatively normal aorta that constitutes the proximal neck is required, unlike a sister device which utilized covered visceral artery stents to treat true suprarenal or thoracoabdominal aneurysms.15, 16

Successful results with a fenestrated graft require appropriate patient selection, proper device design, and technical expertise with endovascular grafting as well as visceral vessel cannulation and stenting. The absence of any one of these factors may result in significant visceral vessel loss, aneurysms unprotected from rupture risk, or the need for open surgical conversion. For this reason, it is imperative that high-resolution imaging studies be obtained and carefully evaluated for acceptability. In addition to conventional endovascular device sizing requirements, fenestrated device design requires accurate distance calculations to be made between the visceral vessels as well as the correct location of the visceral vessel ostia from the aortic circumference. With these data, one can design a prosthesis that will incorporate the required vessels. When patient enrollment began in 2001 it was our practice to perform both an axial CT and a pre-operative diagnostic angiogram to gain sufficient information regarding access vessel size and tortuosity, neck angulation and the relative positions of the renal and other visceral vessels to confidently and accurately plan and size a custom device. Since mid 2003 with the more widespread availability of 3D workstations, availing modalities such as 3D reconstructions, centerline of flow measurement capability, as well as the use of multilpanar reconstructions, the need for pre-operative diagnostic angiography has been supplanted. Nearly, all planning is based on high-resolution CT using a 3D workstation.

Despite a proper device design, the implantation procedure can be challenging. Complicating anatomic factors include visceral vessel stenoses, iliac tortuosity or calcifications that diminish the rotational freedom of the device, and proximal neck angulation. It is critical that clinicians understand the difference between infrarenal and fenestrated devices. The former devices require accurate longitudinal positioning during deployment. Error in achieving the desired location, although undesirable, can be rectified with the use of extension grafts. When implanting a fenestrated device, the longitudinal position must be accurate to within a few millimeters in addition there is a need for precise rotational orientation (to allow access into the visceral vessels). It is the rotational positioning, which is thwarted by severe tortuosity or occlusive disease, which limits torque that may be applied to the proximal graft by rotating the delivery system. This scenario occurred once early in our experience and resulted in our inability to cannulate a single renal artery ultimately causing in renal artery thrombosis. Additional delivery concerns relate to the potential for extensive manipulation of the device within the proximal neck, particularly if atheroma is present, may increase the risk of embolisation.

The need for protection of the renal ostia from graft material management with stents extending through the fenestration into the aorta and degree of flare has been the subject of some discussion. Two types of stents were utilized during the course of the study. These included 0.035 and 0.014in. systems. The larger system (17mm Double Strut, EV3, Plymouth, MN) was used in the absence of renal disease (stenosis) and with single renal arteries. This stent was associated with adequate flaring properties as tested in out laboratory. If significant renal occlusive disease was present or should two renal stents be planned within one fenestration as a result of multiple renal ostia, or an early main renal artery bifurcation, the use of lower profile system was desired and a stent with greater radial force was delivered (Herculink, Guidant Inc., Menlo Park, California). The later stents were not flared beyond the desired renal artery diameter. Worsening renal function was noted in almost 25% of patients and the loss of a number of renal arteries are a significant concern. The majority of renal occlusions occurred within the first month of follow-up, and the remainder within the first year of follow-up. Pre-operative device planning and delivery issues were already discussed. Yet the majority of the renal complications did not have a defined etiology. The early appearance of renal dysfunction is likely associated with either contrast nephropathy or atheroembolization, particularly in the setting of a neck lined with atheroma. The increased manipulation of the device within the pararenal aorta is a potential cause. The need for extensive manipulation likely decreases with the experience of the physicians, but all must remain conscientious of this potential. In spite of the noted renal impairment, no patient without baseline renal insufficiency (GFR<60ml/min) (Fig. 5) required hemodialysis.11 We have struggled with the application of renal and visceral vessel duplex follow-up. Theoretically, ultrasound evaluations may assist with the subclinical detection of stenoses, yet a significant number of false positive duplex studies were encountered. Current duplex guidelines rely upon velocity measurements and ratios of such to determine the severity of stenosis. The presence of an aortic stentgraft, in addition to stents intentionally placed to extend into the aortic lumen, create areas of turbulence, and thus elevated velocities. Therefore, these duplex studies reporting systolic velocity ratios must be interpreted in the context of the clinical situation, the procedural images and the follow-up CT scans. A number of patients have been imaged angiographically based upon elevated velocities (some with a renal aortic ratio as high as 13) and noted to have no aspect of stenosis on imaging studies or with hemodynamic assessment (pressure gradient assessment).

  • View full-size image.
  • Fig. 5. 

    A total of 22 renal stenoses and occlusions occurred in the course of follow-up including 12 renal artery stenoses and 10 renal artery occlusions. Overall five patients required either intermittent or permanent dialysis and this includes four patients with renal artery occlusions. All patients that required dialysis had preoperative renal insufficiency (GFR<60ml/min.)

Despite the technical features allowing the more proximal placement of graft material in the visceral aortic segment, a number of proximal endoleaks were still noted. This was found to be especially prevalent in the setting of neck angulation. When detected these intraoperative endoleaks were all treated during the initial implantation with the superimposition of a Palmaz stent (Cordis Inc.). This stent was deployed below the renal artery stents and provides additional radial force in the immediate infrarenal aorta. It also has the effect of straightening excessive angulation to some degree. Post-operatively, proximal type I endoleaks were aggressively addressed with Palmaz stents or (in a single case) with aortic extension cuffs to increase radial force. All such leaks have resolved and none have recurred to date. The overall low incidence of late type I leaks, supports the concept that the extended sealing zone adequately excludes the aneurysm, even in the setting of the large number of very short proximal necks treated in this series. This is further reinforced by relatively rapid the sac shrinkage observed. Almost 80% of the patients have demonstrated a decrease in aneurysm sac size by at least 5mm ensuring that the natural history of the aneurysmal disease is reversed.

Migration with a device that incorporates branches is unacceptable. The current reporting standards for endovascular aneurysm repair require device movement of 10mm or more, or the presence of a clinical event caused by migration. This definition was revised for our applications.17 Yet it appears that fenestrated devices that migrate will result in massive renal or visceral complications prior to reaching the requisite distance of 10mm. In this series, a single patient was noted to have suffered device migration on the basis of a required secondary renal intervention. The proximal aspect of the device moved only 3mm caudally, yet the renal stent (which was initially placed into the middle portion of the fenestration) was crushed causing a near occlusion. The patient was treated with supplemental renal stenting. This observation was previously reported11 and bespeaks to the importance of the fixation systems. Furthermore, this prompted the placement of the renal stents into the most caudal aspect of the fenestration, to allow for slight downward movement of the device as the hooks of the proximal fixation system engage the aortic wall. Component separation occurred in only one patient at 2 years, and this was treated with an extension cuff. Extensive overlap between components is encouraged to reduce the risk of this uncommon complication. The multiple component interfaces creates concern with respect to long-term durability. Given the complex nature of the device, meticulous imaging follow-up is mandatory such that component separations, device migration, and stent fracture can be detected and treated if necessary. Despite the minimal occurrence of such events in this series, the need for late follow-up of this patient cohort is underscored to allow for more definite conclusions regarding the durability of this repair.

Our results demonstrated that in the hands of an experienced endovascular team, patients can be safely treated despite proximal morphology previously contraindicating an endovascular repair. Early post-operative mortality was exceptionally low, secondary interventions were relatively infrequent, while aneurysm exclusion and sac shrinkage was common. Attention to renal sequellae will be critical, as well as judicious use of post-operative imaging studies to assess the status of the sac. Our practice of aggressively treating proximal type I and modular type III endoleaks seems justified, but has been argued by some to be unnecessary. The infrequency of migration, late visceral loss in the setting of favorable sac behavior indicates that our belief that a stable sealing and fixation site can be created within the pararenal aorta is accurate through intermediate follow-up.

Back to Article Outline

Conflict of Interest Statement 

The author receives (or has received) research funding from Cook Incorporated and royalties for patents licensed to Cook Incorporated.

Back to Article Outline

References 

  1. Greenberg R, Fairman R, Srivastava S, Criado F, Green R. Endovascular grafting in patients with short proximal necks: an analysis of short-term results. Cardiovasc Surg. 2000;8(5):350–354
  2. Green RM. Patient selection for endovascular abdominal aortic aneurysm repair. J Am Coll Surg. 2002;194(Suppl 1):S67–S73
  3. Faruqi RM, Chuter TA, Reilly LM, Sawhney R, Wall S, Canto C, et al. Endovascular repair of abdominal aortic aneurysm using a pararenal fenestrated stent-graft. J Endovasc Surg. 1999;6(4):354–358
  4. Browne TF, Hartley D, Purchas S, Rosenberg M, van Schie G, Lawrence-Brown M. A fenestrated covered suprarenal aortic stent. Eur J Vasc Endovasc Surg. 1999;18(5):445–449
  5. Stanley BM, Semmens JB, Lawrence-Brown MM, Goodman MA, Hartley DE. Fenestration in endovascular grafts for aortic aneurysm repair: new horizons for preserving blood flow in branch vessels. J Endovasc Ther. 2001;8(1):16–24
  6. Anderson JL, Berce M, Hartley DE. Endoluminal aortic grafting with renal and superior mesenteric artery incorporation by graft fenestration. J Endovasc Ther. 2001;8(1):3–15
  7. Kinney EV, Kaebnick HW, Mitchell RA, Jung MT. Repair of mycotic paravisceral aneurysm with a fenestrated stent-graft. J Endovasc Ther. 2000;7(3):192–197
  8. Greenberg RK, Haulon S, O'Neill S, Lyden S, Ouriel K. Primary endovascular repair of juxtarenal aneurysms with fenestrated endovascular grafting. Eur J Vasc Endovasc Surg. 2004;27(5):484–491
  9. Verhoeven EL, Prins TR, Tielliu IF, van den Dungen JJ, Zeebregts CJ, Hulsebos RG, et al. Treatment of short-necked infrarenal aortic aneurysms with fenestrated stent-grafts: short-term results. Eur J Vasc Endovasc Surg. 2004;27(5):477–483
  10. Greenberg RK, Haulon S, Lyden SP, Srivastava SD, Turc A, Eagleton MJ, et al. Endovascular management of juxtarenal aneurysms with fenestrated endovascular grafting. J Vasc Surg. 2004;39(2):279–287
  11. Haddad F, Greenberg RK, Walker E, Nally J, O'Neill S, Kolin G, et al. Fenestrated endovascular grafting: the renal side of the story. J Vasc Surg. 2005;41(2):181–190
  12. Lawrence-Brown M, Hartley D, MacSweeney S, Kelsey P, Ives F, Holden A, et al. The Perth endoluminal bifurcated graft system—development and early experience. Cardiovasc Surg. 1996;4(6):706–712
  13. Greenberg RK, Chuter TA, Sternbergh WC, Fearnot NE. AAA endovascular graft: intermediate-term results of the US multicenter trial. J Vasc Surg. 2004;39(6):1209–1218
  14. Chaikof EL, Blankensteijn JD, Harris PL, White GH, Zarins CK, Bernhard VM, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg. 2002;35(5):1048–1060
  15. Greenberg R, West K, Foster J, Skender D, Pfaff K, Young D et al. Beyond the aortic bifurcation: branched grafting for thoracoabdominal and aortoiliac aneurysms. J Vasc Surg; in press.
  16. Chuter TA, Gordon RL, Reilly LM, Pak LK, Messina LM. Multi-branched stent-graft for type III thoracoabdominal aortic aneurysm. J Vasc Interv Radiol. 2001;12(3):391–392
  17. Greenberg RK, Turc A, Haulon S, Srivastava SD, Sarac TP, O'Hara PJ, et al. Stent-graft migration: a reappraisal of analysis methods and proposed revised definition. J Endovasc Ther. 2004;11(4):353–363

 Presented at the Annual Meeting of the European Society for Vascular Surgery on 17th September 2005 in Helsinki, Finland.

PII: S1078-5884(06)00088-8

doi:10.1016/j.ejvs.2006.01.015

European Journal of Vascular & Endovascular Surgery
Volume 32, Issue 2 , Pages 115-123, August 2006

Article Tools

* Image Usage & Resolution