Introduction

Since the advent of endovascular abdominal aortic aneurysm repair (EVAR) for abdominal aortic aneurysms (AAA), the rate of technical proficiency has reached impressive levels, with modern reports reaching 99%–100% of endografts successfully placed during initial treatment. However, despite technical success and exceedingly low perioperative mortality, the need for reintervention continues to be the crux of EVAR and is most often related to endograft-specific complications. Although complications may be universal to all devices, a significant proportion of devices are being used outside their instructions for use (IFU) guidelines, leading to concern for device-specific complications when used in this manner. Being aware of and knowing how to mitigate or to prevent these complications will help provide better outcomes for patients undergoing EVAR.

Although patient selection is often critical in achieving good outcomes, procedural/device-related issues can occur. Technical complications directly associated with endograft placement during EVAR include: device delivery and deployment issues; access-related complications; lack of fixation or poor sealing of the graft leading to endoleak or aneurysm expansion; stent fractures, graft material breakdown, or component separations. Despite these issues, EVAR has become the standard of care for the surgical treatment of AAA, and multiple devices share these common limitations.

Zenith endografts (Cook Medical Inc., Bloomington, Indiana), one of the most implanted endografts in the United States today, consist of a woven polyester and stainless steel main body. It is a modular, long-bodied, three-piece, self-expanding system with an active suprarenal fixation. It comes with an integrated sheath for delivery and its push-pull method of deployment allows accurate repositioning until the top cap is released. It is approved for 15-mm infrarenal necks and ≤60 degree angulation. Because it is one of the earliest and most used endografts, there is a relatively abundant amount of data on outcomes with use of this device. Here we review the complications that may arise with EVAR results associated with the use of Cook’s Zenith endografts for the treatment of AAA.

Endoleak

Early results using the Zenith AAA endovascular graft, reported in 2004 by Greenberg and colleagues , demonstrated a safe and effective treatment for AAA in both groups of patients, either deemed too high risk for open repair or those considered safe for open repair. Despite technical success being near 99% or higher for both groups, the rate of endoleak in the perioperative period was upward of 17%, the majority of perioperative endoleak being type II. On follow-up, the endoleak rate diminished to 7.4% and 5.4% at 12- and 24-months, respectively. No patient had persistent Type I or III endoleak by 24 months. This was counter-balanced by the standard risk group (for open repair) undergoing more secondary interventions, most of which were for endoleaks. Follow-up to this in 2008 for the same group concluded the long-term durability of the graft to be acceptable in both groups. However, the rate of endoleak continued to be of concern. The rate of late endoleak development (first appearing 6 months after repair and surveillance) was reported between 12% and 14%, most of which were Type II. While generally considered benign, 80% of the endoleaks attributed to causing aneurysm sac enlargement were Type II. Moreover, the majority of secondary interventions were a result of these Type II endoleaks during the study period. Notably, in either group, no Type I or III endoleak was identified at the 5-year follow-up period.

Most recently, Verzini and colleagues have reported the 14-year outcomes of patients treated with Zenith endograft for AAA in Italy. They treated 610 patients over a longer than 10-year period. Overall endoleak of any type at any time was 22.9%, 74% of which were Type II endoleaks. Endoleak of any type was attributed to 9 of the 11 conversions to open repair and 56 of the 80 patients requiring reintervention. While endoleak contributed to a large proportion of the reinterventions for these grafts, the results compare favorably in comparison with other reports, both remote and recent.

Access Site Complications

Insertion of devices for deployment, most commonly the common femoral arteries as the access vessels for EVAR and TEVAR, but also brachial artery access for branched and fenestrated endograft placement, requires either direct vessel cut-down or percutaneous access. Complications during this portion of the procedure is one of the most common group of adverse events in most reports. As the case difficulty and anatomic complexity increases, so does the incidence of these complications.

Complications include acute bleeding during the procedure caused by sheath-size mismatch or vessel laceration, rupture or avulsion of the vessel, hematoma (developing during or after the procedure), thrombosis around the sheath, and dissection or embolization of the vessel distal to the access point. This may require open repair of the vessel in some cases.

Planning and evaluating access sites for device delivery is vital to minimizing these problems. Moreover, small common or external iliac arteries may be difficult to traverse with the large sheaths needed for device delivery. The caliber of the vessels, location and extent of calcification at any point in the vessel, and the tortuosity of the vessel are all characteristics of the access vessels that should be analyzed preoperatively.

Limb Kinking and Occlusion

Limb kinking and occlusion are more commonly seen with endografts following EVAR (2.3% versus 0.2%) compared with open repair and have been associated with a significant increase in endoleak, graft thrombosis, and graft migration. Moreover, graft kinking and occlusion can cause acute limb ischemia. When patients present with symptoms of occlusion, they are often subject to multiple interventions, including thrombolysis, angioplasty and stenting, or ultimately surgical bypass to restore limb perfusion. The rates of limb kinking and limb occlusion, however, are evolving with new devices and delivery systems.

Zenith Spiral-Z iliac limbs are composed of a continuous spiral nitinol stent, between two Z-stents at each end, which delivers kink resistance and flexibility. The rates of limb occlusion with Zenith devices ranges from 2.3% to 3.4%. Concerns of significantly increased rates of limb occlusion with the Zenith LP device, report of upward of 35% , have been dulled by subsequent studies demonstrating minimal limb and access complications when used with Spiral-Z limbs, as mentioned earlier.

Device Migration

Device migration is one of the major causes of secondary intervention after endovascular aneurysm repair. Migration is a time-dependent complication that requires close surveillance and consideration of serial assessments of the endograft. In general, migration of the endograft of 10 mm or more or any displacement requiring a secondary intervention should be considered a serious consequence. Most commonly, proximal aortic aneurysmal degeneration and dilatation or poor patient selection is attributed to device migration. Feared consequences of aneurysm expansion and ultimate rupture lead providers to aggressive intervention in most cases.

Early analysis of the Zenith endograft demonstrated promising results, with 98% freedom of migration (defined as >10 mm endograft movement or any migration related to a clinical event) at 4 years, compared with the AneuRx device, which was reported at 72%. This durability has been further demonstrated by Greenberg, who reported no clinically relevant migrations of >10 mm at 5 years; of 19 patients who were found to have migration distance of 5 to 10 mm, none required secondary intervention or had associated adverse events.

Separation of Components

Endograft component separation is a consequence of modular endograft design or with more complex branched and fenestrated devices. In general, inadequate sizing or overlap may lead to separation or may be related to device-specific integrity. Most of these complications require intervention as a Type III endoleak and may require balloon angioplasty, additional stent placement for bridging the separation, or surgical revision. Component separation with the Zenith graft has been reported to be as low as <0.01% (3 of 736 endografts by Greenberg and colleagues).

Additional Technical Considerations

The top cap on the Zenith endograft device delivery system is a key component of deployment. The top stent is restrained by a trigger wire mechanism that is the limiting factor in deployment location and accuracy. Once the top cap is released, the barbed suprarenal fixation stent is released. The top cap must be advanced proximally into the aorta to deploy fully. If there is difficulty retrieving the top cap, the user should rotate the grey pusher to reorient the smooth edge to aid in advancement. In some cases, it may be difficult to release the trigger wire attached to the top cap. In those instances, inspection of the delivery system under fluoroscopy is mandated to ensure that the system has not been rotated excessively and needs to be untwisted. If difficulty still exists, then the top cap pin vise should be loosened and the top cap pulled down onto the grey pusher. Once the pin vise is retightened, attempts should be made to pull the trigger wire.

Conclusion

EVAR has revolutionized the treatment of AAA. Despite significant improvement in perioperative morbidity and mortality, EVAR is not flawless. Knowing the complications associated with vessel access, device delivery, short- and long-term endograft issues such as endoleak, migration, and component separation, or limb kinking and occlusion, will help providers to maintain the immediate and long-term success of EVAR for the treatment of AAA. The Cook Zenith endograft remains a safe and reliable option in the armamentarium of the vascular surgeon.