Explantation of fenestrated endograft complicated by complete junctional dissociation and symptomatic type IIIa endoleak

Abstract

In this report, we present a 67-year-old man who was transferred from an outside facility with complaints of severe back pain. Computed tomography imaging demonstrated a type IIIa endoleak with complete junctional dissociation of his prior fenestrated endograft, between the visceral and the bifurcated components. Given the perpendicular configuration of the dissociated stent grafts, no endovascular salvage was feasible, and open explantation was indicated. His preoperative evaluation was significant for a positive cardiac stress test warranting coronary artery stenting. The patient underwent subsequent open graft explantation and aortic repair, and was ultimately discharged home after an uncomplicated hospital course.

The optimal approach to abdominal aortic aneurysm (AAA) repair (ie, open vs endovascular) remains a topic of active discourse. Since the introduction of endovascular aneurysm repair (EVAR), treatment patterns have shifted toward less invasive management. However, long-term data from the EVAR-1 and DREAM trials indicate that, although EVAR offers lower perioperative morbidity, open repair may confer superior long-term survival.^1,2 These findings, along with the requirement for lifelong imaging surveillance after EVAR, have renewed interest in open repair, especially among younger patients. We present a late complication of endovascular repair in a patient who underwent the procedure in his 50s. The patient provided consent for publication.

Case report

A 67-year-old man presented to an outside facility with severe, acute-onset back pain. Abdominal plain films revealed an unexpected finding of misoriented abdominal endografts (Fig 1), prompting transfer to our institution for a higher level of care. On arrival, he was hemodynamically stable with moderate back pain. Further history revealed a prior history of a juxtarenal AAA treated with a Zenith Fenestrated AAA Endovascular Graft (ZFEN; Cook Medical) and a stent to the left accessory renal artery, placed 10 years earlier. Per operative report documentation, the initial AAA measured 5.8 cm, and the patient was asymptomatic. The aneurysm was first identified on an aortic duplex, which reported a size of 6.3 cm. The indication for the duplex is not documented. He had since been lost to follow-up without interval surveillance imaging.

Fig 1.

Abdominal plain-film demonstrating dissociation of fenestrated endograft components.

Computed tomography confirmed marked graft malorientation consistent with a type IIIa endoleak (Fig 2). On center-line imaging, the proximal seal zone measured 15 mm proximal to the accessory left renal artery. The right and main left renal arteries were at the same level proximal to this, and the superior mesenteric artery was 2 mm higher than the renal vessels. His comorbidities included controlled hypertension and tobacco use. Endovascular repair would have been considered as a first-line option,³ but the dissociated graft’s perpendicular configuration precluded a safe endovascular approach. He was placed on impulse control therapy given his symptomatic endoleak. Preoperative evaluation included laboratory studies, echocardiography, and a dobutamine stress test (given his reports of exertional dyspnea), which was positive for inducible ischemia. The patient subsequently underwent preoperative coronary artery stenting to the left anterior descending artery for 99% stenosis, and was maintained on aspirin and cangrelor drip. The timing interval from presentation to coronary stenting was 3 days, and from coronary stenting to aortic surgery was an additional 9 days.

Fig 2.

Three-dimensional reconstitution of computed tomography angiogram.

Prior to surgery, epidural catheter placement was deferred due to the patient’s ongoing cangrelor infusion, which was discontinued approximately 45 minutes before the incision. The patient was brought to the operating room and underwent general anesthesia. A midline laparotomy was performed, and dissection was carried down to the retroperitoneum. The duodenum was mobilized in standard fashion. The left renal vein was ligated and divided (per surgeon preference), allowing for dissection of the juxtarenal aorta. All renal arteries were carefully dissected and isolated (per surgeon preference). Dissection was then extended caudally to identify and isolate the bilateral common iliac arteries. Given no gross evidence of infection, we elected to divide and retain a portion of the visceral stent graft. After systemic heparinization, the proximal aortic clamp was placed proximal to the accessory left renal artery, which was separately clamped. The clamp was placed on the self-expanding stent itself, between the two main renal arteries and the accessory left renal artery. The main renal vessels were exposed in case the pararenal clamp was insufficient for hemostasis or if issues arose with dividing between stent graft components, necessitating a suprarenal clamp to remove the device entirely. Upon opening the aneurysm sac, we observed the nearly 90-degree malposition of the visceral and bifurcated endograft components (Fig 3). The visceral component was sharply divided between the second and third stent rings, and the bifurcated and iliac limbs were removed without difficulty. The aorta was reconstructed using a 24 × 12 mm bifurcated Dacron graft pre-soaked in rifampin.⁴ Following graft implantation, heparin was reversed, and the aneurysm sac, retroperitoneum, and abdominal wall were closed in standard fashion. Estimated blood loss was 1200 mL.

Fig 3.

Intraoperative findings demonstrating dissociated and perpendicular endograft components after opening the aneurysm sac.

The patient’s postoperative course was relatively uncomplicated. Transverse abdominis plane nerve blocks were placed postoperatively, and cangrelor was resumed in the post-anesthesia care unit setting. Diet was advanced following return of bowel function. The patient was discharged home in good condition on postoperative day 7 and was noted to be recovering well without complications at clinic follow-up.

Discussion

Fenestrated and branched EVAR has significantly expanded treatment options for patients with juxtarenal and suprarenal abdominal aortic aneurysms who are poor candidates for open surgery. Recent studies have reported that fenestrated and branched EVAR is associated with reduced perioperative morbidity and mortality compared with open repair of complex aneurysms.⁵ Currently, the ZFEN endograft is the only commercially approved fenestrated device in the United States.⁶

Despite these advantages, patient selection is critical. The initial ZFEN repair was performed electively when the patient was in his 50s after he was offered both open and endovascular options, and the patient elected endovascular repair. Informed decision-making is essential, particularly in emphasizing the need for lifelong imaging surveillance after endovascular repair, as outlined in the Society for Vascular Surgery guidelines.⁷ Long-term outcomes remain an important part of informed consent discussions as well, as updated results from the EVAR-1 and DREAM trials suggest higher late mortality and reintervention rates with endovascular repair. In our practice, we generally favor open repair for younger, low-risk patients, even when endovascular repair is anatomically feasible.^1,2 Furthermore, as the field continues to shift toward increasingly sophisticated endovascular techniques, it is imperative that vascular surgeons maintain robust training in both open and endovascular aortic skillsets.⁸ Careful, individualized decision-making is essential to optimizing long-term outcomes.

There are several potential causes of type IIIa endoleaks. Early type IIIa endoleaks typically result from inadequate overlap between stent graft components, whereas late type IIIa endoleaks may develop due to growth or elongation of the aneurysm sac, endograft migration, or dilation of the proximal or distal attachment sites.⁹ In this scenario, the most likely etiology was a long-standing type II endoleak that led to progressive aneurysm sac expansion and eventual component dissociation. To support this hypothesis, the presence of a delayed type II endoleak was reported on the initial operative note. This case underscores the importance of continued, lifelong surveillance following EVAR, particularly given the passive fixation mechanisms of the ZFEN device, which lacks active anchoring. Although recent studies have suggested that even early type Ia and type III endoleaks after ZFEN implantation may resolve without intervention,¹⁰ this remains controversial. In this case, sac expansion went undetected for nearly a decade, and the resultant graft malposition could have been catastrophic. Open conversion for late complications after EVAR has been shown to be safe and associated with favorable mid-term durability.¹¹

Conclusions

Complete junctional dissociation of the visceral and bifurcated components of a fenestrated aortic endograft is a very rare complication. Although endovascular salvage may be the first-line treatment for delayed type IIIa endoleaks,³ open explantation is indicated for scenarios in which endovascular treatment is not feasible. This case report underscores the importance of lifelong surveillance after fenestrated aortic endografting.

Funding

None.

Disclosures

None.