Endovascular Aneurysm Sealing (EVAS) and Chimney EVAS in the Treatment of Failed Endovascular Aneurysm Repairs

Introduction

In the era of endovascular therapy, the number of patients with failed endovascular aneurysm repair (EVAR) is increasing, with 19.2% of patients requiring additional secondary procedures in the long term. ¹ Distal endograft migration and type I endoleak are among the more serious complications of EVAR. Type I and III endoleaks may occur in up to 12% of EVAR patients ² and represent the predominant cause of late rupture and need for reintervention. ³

The majority of endoleaks can be treated with endovascular techniques; however, the best method has been an issue of debate. It is estimated that open conversion may be required in 3.7% of EVAR patients, ⁴ representing a therapeutic dilemma in older patients with significant comorbidities. Several endovascular approaches, including custom-designed fenestrated stent-grafts, infrarenal extension cuffs, extra-large stents, endoanchors, and coil embolization, have been described in conjunction with EVAR salvage maneuvers.^5,6 However, there is no “first choice” of the reintervention method to treat type I and III endoleaks.

The methods and their technical success may depend on the anatomy, experience of the vascular center, and the onset of the aortic pathology. Therefore, the use of additional endovascular techniques may still be needed to avoid open conversion in patients for whom other conventional methods would prove to be difficult or infeasible. This article describes an alternative approach for treatment of patients with failed EVAR using the Nellix Endovascular Aneurysm Sealing (EVAS) system (Endologix Inc, Irvine, CA, USA), either alone or in combination with chimney grafts.

Methods

Patients and Preoperative Imaging

Fifteen consecutive patients (mean age 79 years, range 69–93; 14 men) presented with failed prior endovascular repair at 2 institutions between March 2014 and December 2015. The mean interval from the initial implantation was 48 months (range 18–84). The median aneurysm diameter was 7.8 cm (range 6.5–11). Thirteen patients had bifurcated endografts, one had a tube endograft, and the last had a bifurcated graft plus a dual fenestrated cuff to accommodate both renal arteries and a scallop for the superior mesenteric artery (SMA). All patients were considered unsuitable for an open surgical approach because of severe comorbidities (American Society of Anesthesiologists III/IV), advanced age, or multiple previous abdominal surgeries.

Endoleak type Ia was the most common EVAR complication (n=10), followed by endoleak type IIIa (n=3; one was associated with the fenestrated cuff). Two patients had a type IIIb endoleak (Figure 1). The patient characteristics are shown in Table 1.

OPEN IN VIEWER

Figure 1.

(A) Maximum intensity projection and (B) preoperative 3-dimensional (3D) computed tomography angiography (CTA) reconstruction showing a symptomatic 7-cm abdominal aortic aneurysm with disintegration of an infrarenal main graft after endovascular aneurysm repair (EVAR) (arrows in A). (C) Intraoperative fluoroscopy showing the deployment of Nellix grafts into the failed EVAR (arrows delineate the extent of the existing stent-graft disruption). (D) Postoperative CTA and CTA 3D reconstruction showing the sealing of the aneurysm sac and the position of the Nellix grafts.

OPEN IN VIEWER

Table 1.

Characteristics and Operative Details of the 15 Patients. ^a

Characteristics
Age, y	79 (69–93)
Men	14
AAA diameter, cm	7.85 (6.5–11)
Symptomatic / ruptured AAA	6 / 2
Smoking	12
Hypertension	13
Diabetes	3
Coronary artery disease	6
COPD	3
Previous abdominal surgery	4
ASA III / IV	11 / 4
Previous EVAR
Bifurcated stent grafts	13
Fenestrated cuff and bifurcated graft	1
Tube stent-graft	1
Complications after EVAR
Endoleak type Ia	10
Endoleak type IIIa / IIIb	3 / 2
EVAS procedure
EVAS (relining with 2 Nellix grafts)	5
ChEVAS	10
1-vessel (left RA)	2
2-vessel (both RAs)	4
3-vessel (both RAs and SMA)	3
4-vessel (both RAs, SMA, and CT)	1
Operative details
Local anesthesia	5
Contrast volume, mL	230±244
Polymer fill volume, mL	56.7±47.8
Procedure time, min	171±81
ICU, d	1 (0–60)
Length of stay, d	7 (5–60)

Abbreviations: AAA, abdominal aortic aneurysm; ASA, American Society of Anesthesiologists; CG, chimney graft; chEVAS, chimney grafts in combination with endovascular aneurysm sealing; COPD, chronic obstructive pulmonary disease; CT, celiac trunk; EVAR, endovascular aortic repair; EVAS, endovascular aneurysm sealing; ICU, intensive care unit; RA, renal artery; SMA, superior mesenteric artery.

a

Continuous data are presented as the means ± standard deviation or median (range); categorical data are given as the counts.

Electrocardiographically gated, high-resolution computed tomography angiography (CTA) was performed with 1-mm slice thickness to evaluate the left subclavian artery (LSA), iliofemoral anatomy, and aortic pathology/aneurysm morphology. Chimney grafting in combination with EVAS (chEVAS) was planned in 10 patients owing to unfavorable neck anatomy, an inadequate (>10-mm) landing zone proximal to the existing stent-graft, or extension of the aneurysm pathology above the renal arteries (Figure 2). Eight of these patients were not suitable for a fenestrated endograft because of their emergent situation (6 symptomatic aneurysms and 2 contained ruptures). The procedure was explained to the patients, who all gave informed consent to the operation.

OPEN IN VIEWER

Figure 2.

(A-C) Preoperative axial computed tomography angiography (CTA), (D) 3-dimensional (3D) reconstruction, and (E) maximum intensity projection reconstructions showing the ruptured 6-cm abdominal aortic aneurysm owing to a type Ia endoleak (arrows in B, C, and E). (F, G) Postoperative axial CTA scans and (H, I) 3D reconstructions showing the sealing of the aneurysm sac and the position of the chimney and Nellix grafts.

Results

A total of 31 Nellix stent-grafts were implanted to reline 15 EVAR stent-grafts. Two Nellix grafts were implanted in 15 patients and 1 patient received 3 Nellix grafts placed with the Nellix-in-Nellix technique because of a right iliac aneurysmal segment distal to the EVAR prosthesis. The 10 chEVAS patients received 23 balloon-expandable covered stents to preserve the renovisceral vessels and extend the Nellix seal zone to the suprarenal aortic segment (Table 1). Two patients received a single chimney graft, 4 had 2 chimneys implanted, 3 patients had 3 chimneys, and 1 patient had all 4 renovisceral vessels targeted. Target vessels were 10 left renal arteries, 7 right renal arteries, 4 SMA, and 2 celiac trunks.

The patient with the fenestrated cuff presented with an 11-cm aneurysm and a type IIIa endoleak from the lower left renal fenestration. Attempts to balloon dilate the left renal stent-graft had been unsuccessful in a prior reintervention, so a chimney graft was implanted into the left renal stent-graft combined with 2 Nellix endografts to reline the fenestrated EVAR stent-grafts.

Technical success was 100%. All endoleaks were successfully sealed, and no additional intervention was required. No further endoleak after EVAS or chEVAS was recorded, and all chimney grafts were patent on intraoperative completion angiography. One endobag protrusion was seen in the patient with type Ia endoleak treated with 2 Nellix grafts as an extension. The endobag protruded into the suprarenal aortic segment without occlusion of the renovisceral ostia and remained without clinical consequences.

The mean operative time, fluoroscopy time, and contrast and polymer volumes are given in Table 1. The average renovisceral ischemia time was 12±3 minutes during the EVAS procedure. The mean blood loss during the implantation procedure was 500±380 mL (range 120–880).

One major complication, an iatrogenic renal artery injury with active bleeding, occurred in a symptomatic patient with a 7-cm aneurysm treated with 3-vessel chEVAS. The patient underwent nephrectomy and recovered after the surgery. The patient was discharged 10 days thereafter. Two patients had transient renal impairment with a peak value of 2.3 mg/dL, which decreased to the preoperative values by the time of discharge.

One patient died in-hospital 2 months postoperatively (6.6% mortality). The 82-year-old patient with ruptured aneurysm and preexisting chronic obstructive pulmonary disease required prolonged ventilator time and succumbed to multiple organ failure.

All patients received postoperative CTA showing chimney patency and no endoleak. Over a median follow-up of 8 months (range 3–24), 3 patients (all >80 years old) died at 3, 7, and 7 months after the surgery because of non-aneurysm-related causes (stroke, cardiac failure, and unknown). After 3, 6, and 12 months, 13, 11, and 5 patients had CTA, respectively. No endoleaks were observed, and a stable or shrinking (≥5-mm, n=6) sac diameter was detected in all patients. All Nellix and chimney stent-grafts in the surviving patients remained patent. No late aneurysm-related deaths or reinterventions occurred.

Discussion

Open surgical conversion of failed EVAR is a challenging procedure. It may pose increased perioperative mortality and morbidity because of suprarenal clamping and the risk of renovisceral aortic segment injury during stent-graft explantation, particularly in the presence of suprarenal fixation. ⁸

The development of type I endoleak after EVAR has an incidence of 5% to 25% and represents ongoing risk of aneurysm rupture, strongly indicating reintervention. ⁹ In the presence of a suitable infrarenal neck, an additional proximal stent-graft (ie, cuff) can be used. Other options, such as coil embolization and endoanchors, can be performed. However, the few publications concerning embolization of type Ia endoleak report that >1 embolization session ¹⁰ may be required or even conversion in 2.4% of the cases. ¹¹ Despite high technical success and encouraging outcomes of the endoanchor system, ¹² key anatomical limitations, including significant thrombus, calcification, and plaque >2 mm, may compromise fixation of the anchors in the aortic tissue, thereby preventing successful application of this device. If the proximal neck is inadequate, fenestrated EVAR offers an established technique with excellent early and midterm outcomes in the repair of type I endoleak after previous EVAR. ¹³ However, this technique is often not performed in acute situations and may be limited because of anatomical shortcomings.^14,15

The Nellix EVAS system has been introduced as a new and effective technique to treat abdominal aortic aneurysm. An increasing number of studies indicate that EVAS is safe and effective, with promising short- and midterm results.^16–22 EVAS had been used to repair type I/III endoleaks after failed EVAR in isolated cases^23,24 and post-EVAR complications in cases collected from several centers, ²⁵ but the experience to date is still limited. The present 15-patient series suggests that the use of EVAS may offer a safe and effective alternative treatment after failed EVAR.

In this study, we used the EVAS technology because of its compliant endobags and ability to conform to anatomical and prosthetic structures. Moreover, an advantage of chEVAS is the possibility to extend the sealing into a healthier aortic segment, providing an endovascular option to land suprarenally or even above the celiac trunk in previous failed EVAR reconstruction.

Special attention should be paid during the prefill step and pressure monitoring to avoid endobag protrusion, which occurred in one of our patients but without compromising the renal arteries. Some technical modifications of the EVAS steps, including addition of 20 mL of contrast in the prefill with saline, may be useful to visualize the endobags and achieve a more controlled endobag filling.

The use of parallel grafts with conventional aortic stent-grafts (chimney technique) has shown positive midterm outcome in the treatment of patients with failed prior EVAR and type Ia endoleak.^26–28 A low incidence of persistent gutter endoleak has been reported using the conventional chimney technique, though we did not observe this event in the current series. A potential benefit of the chEVAS technique is the filling of the spaces between the parallel grafts by the endobags, preventing the development of gutters. This concept has been clearly reported by other groups performing chEVAS.^29–34 We also observed stable structural integrity of the Nellix graft and the conventional graft devices. No disconnections or type III endoleaks occurred during the brief follow-up period.

Although chEVAS can be a very demanding procedure, especially in emergent situations after failed EVAR, its availability represents a great advantage as an off-the-shelf solution compared with the fenestrated grafts. Therefore, this technique can be used when other therapeutic options are classified as difficult or not suitable.

Conclusion

The present preliminary experience demonstrates that the use of EVAS/chEVAS is feasible for treatment of failed EVAR. This technique may be used as bailout or an alternative treatment when other established methods are infeasible or not available.