Physician-modified inner-branched endovascular repair with re-intervention

Introduction

A physician-modified inner-branched endovascular repair (PMiBEVAR) is an effective treatment for high surgical risk patients. However, complications are expected as physician-modified stent-grafts have a complex structure. We experienced a case of re-intervention for PMiBEVAR failure.

Presentation of case

A 75-year-old man had undergone graft replacement of the ascending aorta using a 24 mm Gelweave (Vascutek Terumo Inc, Scotland, UK) 17 years prior, graft replacement of the aortic arch with an elephant trunk using a 26 mm J-graft Shield Neo (Japan Lifeline, Tokyo, Japan) 8 years prior, and thoracic endovascular aortic repair using 34 × 34 × 200 mm CTAG (W.L. Gore and Associates, Flagstaff, AZ, USA) and 34 × 34 × 200 mm CTAG (W.L. Gore and Associates) 4 months prior to the present complaint. The patient presented to our hospital with a thoracoabdominal aortic aneurysm (Crawford’s classification type IV) (Figure 1(a)). He had a history of multiple surgeries, chronic kidney disease, and pancytopenia. Due to the high surgical risk, we performed a PMiBEVAR. The physician-modified stent-graft was prepared sterilely using a 32 × 32 × 109 mm Zenith Alpha thoracic stent-graft (Cook Medical, Bloomington, IN, USA). The graft was partially unloaded using a delivery system, and the proximal fixation barbs were removed to facilitate re-sheathing. Based on the measurements from the preoperative computed tomographic angiography, 7 mm, 10 mm, and 7 mm fenestrations for the celiac artery (CA), superior mesenteric artery (SMA), and bilateral renal arteries (BRAs) were created in predetermined locations using electrocautery, respectively. The inner branch of the CA was not prepared for graft. The inner branch of the SMA was prepared using a 10 × 10 mm Viabahn (W.L. Gore and Associates, Flagstaff, AZ, USA). The inner branches of the BRA were prepared using a 7 × 10 mm Viabahn (W.L. Gore and Associates) and securely affixed to the fenestration using Prolene running 5-0 Prolene sutures placed within the stent-graft; coils were used as markers. The modified stent-graft was manually re-sheathed. The modification time was set to 100 min.OPEN IN VIEWER

The right femoral and axillary arteries were surgically exposed. After establishing through and through axillary-femoral access, the physician-modified stent-graft was advanced into the thoracoabdominal aorta via the right femoral artery. To assist in the cannulation from the right axillary artery to the inner branch, precannulation wires were inserted in the delivery system. This necessitated the use of an axillofemoral through-and-through access for wire placement. After unsheathing, we did not place the bridging stent-graft for the CA owing to its sufficient distance from the aortic aneurysm. The first inner-branch endograft of the SMA was expanded. A 9 × 58 mm LifeStream (Bard Peripheral Vascular, Tempe, AZ, USA) was deployed from the inner branch. A similar procedure was performed for the left renal artery using a 7 × 37 mm LifeStream (Bard Peripheral Vascular). However, only a microcatheter could be inserted into the right renal artery (RRA) owing to its ostial stenosis (Figure 1(b)). Moreover, the bridging stent could not be placed in the RRA due to the significant distance from the fenestration to the RRA as well as the severe angulation. Therefore, we placed coils in the RRA to prevent type II endoleak, and the inner branch was occluded by deploying a 32 × 32 × 45 mm Excluder AAA endoprosthesis (W.L. Gore and Associates, Flagstaff, AZ, USA) at the level of the fenestration in the physician-modified stent-graft to prevent it from graduating to type IIIc endoleak.

Postoperative computed tomography angiography revealed an endoleak from the inner branch of the RRA (Figure 2). We hypothesized that the aortic cuff deployed at the level of the inner branch was not expanded enough as we chose not to over-expand the Excluder AAA endoprosthesis. Re-intervention was performed with embolization of the endoleak using coils (Figure 3(a)). No endoleaks were detected on completion angiogram. Imaging after re-intervention showed successful endoleak obliteration (Figure 3(b) and (c)). Although serum creatine was increased because of RRA occlusion, the patient was discharged without any indication of dialysis.OPEN IN VIEWER

OPEN IN VIEWER

Figure 3.

Imaging after re-intervention. (a) Coils in the RRA (yellow) and embolization of the endoleak from the inner branch of the RRA with coils (light blue). (b), (c) Obliteration of the endoleak (white arrowhead). RRA: right renal artery.

Discussion

The reports of PMiBEVAR for thoracoabdominal aortic aneurysms are increasing.^1–3 Especially in cases of high surgical risk, it has proven to be an effective treatment. Zhang Y et al. described a successful total endovascular repair with physician-modified endograft with triple inner branches for aortic arch aneurysm in a complex repeated surgical setting involving high risks (advanced age, relevant comorbidities, and previous ascending aorta replacement). ⁴ Using the stent-graft with inner branches allowed us better handling and ease of target vessel cannulation compared with the use of outer branches that could have been difficult to open in narrow spaces. ⁵

Complications are expected as physician-modified stent-grafts have a complex structure, and these complications require management. However, there are no reports of additional treatment for PMiBEVAR failure. Silverberg et al. performed 14 re-interventions for endoleaks after fenestrated and branched endovascular repair. ⁶ One patient experienced detachment of the side branches within the RRA and underwent an unsuccessful attempt to bridge these components. The aortic cuff was occluded with an endovascular plug, resulting in kidney loss. Other procedures included coiling of the endoleak and bridging between separated components with additional stent-grafts.

In our case, the bridging stent could not be placed in the RRA because of the increased distance from the fenestration to RRA and the severe angulation. Therefore, we placed coils in the RRA to prevent type II endoleak, and the inner branch was occluded by the aortic cuff at the level of the fenestration in the physician-modified stent-graft to prevent type IIIc endoleak. However, the endoleak remained, thereby necessitating a re-intervention. Retrospectively we realized several strategies could have been employed to avoid re-intervention. We should have considered oversizing the aortic cuff to achieve a more robust and dependable sealing mechanism. Additionally, fenestration occlusion could have been performed by placing an endovascular plug or coils into the aneurysm sac. Moreover, we could have attempted pre-dilation of the right renal ostium especially, in a patient with pre-existing chronic renal disease. The patient underwent embolization of the endoleak from the inner branch of the RRA using coils. In spite of the decline in renal function, imaging after re-intervention showed successful obliteration of the endoleak.

We reported a successful case of re-intervention for PMiBEVAR failure.