The Initial Results of Physician-Modified Fenestrated-Branched Endovascular Repairs of the Aortic Arch and Lessons Learned From the First 21 Cases

Abstract

Introduction:

Physician-modified fenestrated-branched endovascular aortic repair (PM-FBEVAR) for the aortic arch provides a minimally invasive treatment option for patients who are too high-risk for open repair. Improvements in technique are gained with ongoing experience with these complex repairs. This study aims to describe outcomes of arch PM-FBEVAR and technical lessons.

Materials and Methods:

A retrospective review of consecutive patients who underwent PM-FBEVAR with zone 0 proximal sealing at a single institution between January 2019 and July 2023 was performed. Cases completed using initial techniques (early technique) were compared with cases using the current techniques (current technique). Modification technique changed to include a self-orienting spine trigger wire and anatomically specific fenestrations or inner branches in the current group. The primary outcome was in-hospital mortality. Secondary outcomes included technical success and 30 day stroke.

Results:

A total of 21 patients underwent arch PM-FBEVAR, with 7 in the early group and 14 in the current group. Severe comorbidities were present in both groups including chronic obstructive pulmonary disease (COPD) (43% vs 36%), prior open ascending aortic repair (57% vs 43%), and prior stroke (86% vs 21%), respectively. Technical success was the same (86% vs 86%, p=1.0). Fluoroscopy time (56 vs 24 min, p=0.012) and in-hospital death (43% vs 0%, p=0.026) were significantly lower in the current group. A 30 day stroke rate (29% vs 7%, p=0.247) was non-significantly decreased in the current group. All-cause mortality was 100% vs 7% during median follow-up of 8 and 6 months (p<0.001). Three deaths in the early group were related to their aortic arch repair including aortic rupture during endograft advancement and 2 postoperative strokes.

Conclusion:

There is a significant learning curve associated with aortic arch PM-FBEVAR. This study suggests that gained experience, use of the spine trigger wire technique, and precise creation of fenestrations or inner branches can lead to a shorter procedure time and lower complications.

Clinical Impact

Physician modified fenestrated branched endografting is feasible for the aortic arch. The high rate of stroke and perioperative mortality was reduced with incorporation of self-orienting spine trigger wire and anatomically specific inner branch creation.

Keywords

aortic arch aortic arch aneurysm zone 0 fenestrated-branched endovascular repair physician-modified endovascular graft

Introduction

Endovascular aortic repair of the descending thoracic aorta, visceral renal aorta, and infrarenal aorta are being widely performed with excellent reported outcomes.^1–3 For the aortic arch, open surgical repair remains the standard of care. For patients who are high risk for open arch repairs, total endovascular repair using custom-manufactured fenestrated-branched devices have been used successfully with acceptable short-term outcomes.^4–7 However, these devices are only available outside the United States and through select centers with investigational device exemption protocols. Other techniques have been used for endovascular arch repair including off-label use of off-the-shelf single-branched endografts, parallel grafting, in situ fenestration, and physician-modified endografts.^8–10 Among these techniques, physician-modified fenestrated-branched endovascular aortic repair (PM-FBEVAR) provides the operating surgeon the opportunity to create custom devices with desired features that resemble manufactured devices. Currently, the literature on PM-FBEVAR of the aortic arch incorporating all 3 aortic arch branches is limited to small series, mainly featuring fenestrations.^11–14 After extensive experience with PM-FBEVAR to treat thoracoabdominal aortic pathologies, our group began applying this approach to the aortic arch.

The aim of this study is to describe the initial experience with PM-FBEVAR for the aortic arch, comparing outcomes before and after changes in the technical approach based on lessons learned from the initial cases.

Materials and Methods

Study Design

An observational retrospective cohort analysis of consecutive patients undergoing PM-FBEVAR of the aortic arch at a single institution between January 2019 and July 2023 was completed. Included patients underwent fenestrated-branched endovascular repair of the aortic arch with proximal landing in zone 0, endograft coverage through zone 2 or greater, and all 3 aortic arch branch vessels incorporated into the repair or covered. Cases that used off-the-shelf fenestrated-branched devices and were completed without back table physician modification were excluded from the study. The study was approved by the Institutional Review Board.

An Early Technique group and a Current Technique group were created based on changes in technique that occurred part way through the series. The changes in technique included rerouting a trigger wire along the outer curve of the endograft as a self-orienting “spine” wire (Figure 1A) and constructing the inner branches and fenestrations at the precise location of each arch vessel without an intentional off-set for working room (Figure 1B). None of the cases in the Early Technique group used these techniques, whereas all the cases in the Current Technique group used these changes in technique. All patients had been deemed prohibitively high risk for open surgical arch repair by a team of cardiothoracic surgeons with extensive experience in aortic surgery. The primary outcome was in-hospital mortality. Secondary outcomes included technical success as defined by the Society for Vascular Surgery reporting standards,¹⁵ (1) 30 day stroke, endoleak, and reintervention. Stroke, endoleaks, and reinterventions were defined according to the SVS reporting standards. Early outcomes were defined as those that occurred within 30 postoperative days.

Device Design and Procedural Details

Anatomic suitability, including adequate landing zone consisting of at least 2 cm of healthy native aortic wall or previous open graft, burden of atheromatous disease or thrombus in the aortic arch and branch vessels, and vessel size, was evaluated using high-resolution computed tomography scans of the chest, abdomen, and pelvis with a preference for computed tomography angiography scans with 1 mm cuts or finer and centerline reconstruction (3mensio, Utrecht, The Netherlands). Anatomic suitability and device design for arch PM-FBEVAR were assessed by a vascular surgeon with 10 years of dedicated experience in complex endovascular aortic repairs and a physician-sponsored investigational device exception for PM-FBEVAR. The modified endografts used included the Zenith Alpha Thoracic Endovascular Graft, the Zenith Alpha Thoracic Endovascular Graft, and the Zenith TX2 Dissection Endovascular Graft (Cook Medical, Bloomington, Indiana). Device modifications for the arch branch vessels included fenestrations, inner branches, and outer branches depending on anatomical suitability and operative approach, based on previously described techniques for the visceral renal aorta.^16–18 Branches were either an antegrade or retrograde direction based on anatomic suitability and access site for bridging stent delivery.

Fenestrations were created using ophthalmic cautery, whereas inner branches were created using a VIABAHN (W. L. Gore & Associates, Flagstaff, Arizona) stent deployed through the appropriate fenestration. In the current technique group, one of the proximal trigger wires was rerouted longitudinally along the greater curvature of the endograft, corresponding to the 12 o’clock position of the hind view of the aortic arch, as the spine wire.

All cases were performed in a hybrid operating room with an Artis pheno robotic c-arm (Siemens Healthineers, Forchheim, Germany) and integrated 3D fusion overlay technology under general anesthesia and were completed by a team of cardiothoracic and vascular surgeons. The main body aortic endograft was delivered from the common femoral artery via percutaneous or open surgical access. Rapid ventricular pacing during deployment of the arch endograft was used consistently after the first 2 cases. Common carotid and brachial arteries were accessed for delivery of bridging stents, or to assist transfemoral delivery of the bridging stents, based on the device design and patient anatomy. 3D fusion overlay technology was used, whenever available, to guide deployment of the aortic endograft and cannulation of the target vessels.

Statistical Analysis

The patient demographics and characteristics, operative techniques and intraoperative metrics, early postoperative outcomes, and the midterm postoperative outcomes were compared between the 2 groups. Categorical variables were reported with frequency and percent and were compared using the Fisher exact test or χ². Continuous variables that were normally distributed were reported with mean and standard deviation, whereas those that were non-normally distributed were reported with median and interquartile range (IQR). Continuous variables were compared using the Student t-test or Wilcoxon rank sum test. Tests were 2-sided, and statistical significance was set at p<0.05. Data were analyzed using SPSS 28 (IBM SPSS, Chicago, Illinois).

Results

Patient Characteristics

There were 25 patients who underwent zone 0 endovascular aortic arch repair during the study period. A total of 21 patients were included in the study. Four patients were excluded as they underwent endovascular aortic arch repair with off-the-shelf branched thoracic aortic devices and without physician modification of the aortic endografts. Seven patients (33%) were in the early technique group and 14 patients (67%) were in the current technique group. Both groups had a similar mean age (75.9 ± 6.3 vs 74.9 ± 8.2 years). There were more females in the study overall and a greater proportion of females in the early technique than the current technique group (71% vs 57%, p=0.656). Most medical comorbidities were clinically similar between the groups including a history of hypertension (100% vs 100%), chronic obstructive pulmonary disease (43% vs 36%), and chronic kidney disease (29% vs 21%). History of prior stroke (86% vs 21%, p=0.016) and coronary artery disease (43% vs 21%, p=0.354) were more common in the early technique group. The most frequent aortic pathologies were previous type A aortic dissection (43% vs 29%) and degenerative aortic aneurysm (29% vs 36%). More than half the patients presented with symptomatic aortic disease in both groups (57% vs 57%). Nearly half of the patients had previously undergone at least 1 sternotomy with open ascending aortic replacement or hemiarch repair (57% vs 43%), and a minority of patients had undergone a prior zone 2 or 3 thoracic endovascular aortic repair (TEVAR) (14% vs 21%). Further demographics and characteristics can be seen in Table 1.

Table 1.

Patient Demographics and Characteristics.

Characteristic	Early technique (N=7) (%)	Current technique (N=14) (%)	Total (N=21) (%)	p-value
Age (years)	75.9 ± 6.3	74.9 ± 8.2	75.2 ±7.5	0.795
Sex (male)	2 (28.6)	6 (42.9)	8 (38.2)	0.656
Body mass index (kg/m²)	25.6 ± 6.1	25.1 ± 6.8	25.3 ± 6.4	0.871
Coronary artery disease	3 (42.9)	3 (21.4)	6 (28.6)	0.354
Prior congestive heart failure	0 (0)	2 (14.3)	2 (9.5)	0.533
Diabetes	0 (0)	2 (14.3)	2 (9.5)	0.533
Hypertension	7 (100)	14 (100)	21 (100)	N/A
Hyperlipidemia	7 (100)	10 (71.4)	17 (81.0)	0.255
Smoking history	2 (28.6)	6 (42.9)	8 (38.1)	0.656
Chronic obstructive pulmonary disease	3 (42.9)	5 (35.7)	8 (38.1)	1.00
Chronic kidney disease	2 (28.6)	3 (21.4)	5 (23.8)	1.00
Peripheral artery disease	2 (28.6)	0 (0)	2 (9.5)	0.100
Stroke	6 (85.7)	3 (21.4)	9 (42.9)	0.016
Previous ascending/hemiarch open aortic repair	4 (57.1)	6 (42.9)	10 (47.6)	0.659
Previous zone 2 or 3 TEVAR	1 (14.3)	3 (21.4)	4 (19.0)	1.00
Previous type A dissection	3 (42.9)	4 (28.6)	7 (33.3)	0.638
Max aortic arch diameter (cm)	5.7 ± 0.9	5.2 ± 1.3	5.4 ± 1.2	0.456
Indication				0.798
Degenerative aneurysm	2 (28.6)	5 (35.7)	7 (33.3)
Post-dissection aneurysm	3 (42.9)	3 (21.4)	6 (28.6)
Acute dissection/penetrating aortic ulcer/intramural hematoma	1 (14.3)	3 (21.4)	4 (19.0)
Pseudoaneurysm	1 (14.3)	3 (21.4)	4 (19.0)
Symptomatic	4 (57.1)	8 (57.1)	12 (57.1)	1.00
ASA class				1.00
ASA III	4 (57.1)	7 (50)	11 (52.4)
ASA IV	3 (42.9)	7 (50)	10 (47.6)

Abbreviations: ASA, American Society of Anesthesiology; TEVAR, thoracic endovascular aortic repair.

Device Design and Procedural Metrics

The modified endograft was a proximal taper Zenith Alpha Thoracic Endovascular Graft for 19 cases, a non-tapered Zenith Alpha Thoracic Endovascular Graft in 1 case, and a Zenith TX2 Dissection Endovascular Graft in 1 case (Cook Medical). Device modification for the innominate artery was mostly antegrade inner branches with the early technique (71%) whereas the current technique group received inner branches (50%) or fenestrations (50%) (p=0.15). One patient in the early group and 7 patients in the current group had fenestrations without bridging stents. Modification for the left common carotid artery was either an antegrade inner branch (57%) or retrograde inner branch (29%) in the early technique, whereas the current technique group most often received a fenestration (43%) (p=0.337). Four patients in the current group had fenestrations without bridging stents. Three patients with bovine arch anatomy received a large unstented fenestration for the common trunk. Multiple modification strategies were used for the left subclavian branch in the early technique group including retrograde inner branch (29%), fenestration (29%), in situ fenestration (14%), parallel grafting (14%), and coverage with cervical debranching (14%). The retrograde inner branch was most commonly used for the left subclavian artery in the current technique group (64%). One patient in the current group had a fenestration without a bridging stent. Three patients in the current technique group had already undergone a zone 2 TEVAR with stenting of the left subclavian artery, which was incorporated and reinforced during the endovascular arch repair. One patient in each group underwent cervical debranching including a bilateral carotid to axillary bypass for residual dissection extending into the bilateral subclavian arteries and a left carotid to subclavian bypass in preparation for an open arch repair that instead underwent PM-FBEVAR. Bilateral common carotid artery access was obtained frequently in the early technique group (86%), whereas in the current technique group, 36% underwent bilateral common carotid artery access and 36% had neither carotid artery accessed as transfemoral delivery was used for the carotid stents (p=0.313). The median total procedure time was 339 (IQR=125) vs 231 (IQR=127) minutes (p=0.046), the median fluoroscopy time was 56.0 (IQR=17.9) vs 24.2 (IQR=19.4) minutes (p=0.012), the median iodinated contrast use was 80 (IQR=49) vs 85 (IQR=45) cc (p=0.547), and the median estimated blood loss was 700 (IQR=2800) vs 100 (IQR=163) cc (p=0.003) in the early and the current technique groups, respectively (Table 2).

Table 2.

Operative Techniques and Metrics.

Technique/metric	Early technique (N=7) (%)	Current technique (N=14) (%)	Total (N=21) (%)	p-value
Spine trigger wire	0 (0)	14 (100)	14 (66.7)	N/A
Innominate modification				0.150
Antegrade inner branch	5 (71.4)	7 (50.0)	12 (57.1)
Retrograde inner branch	1 (14.3)	0 (0)	1 (4.8)
Fenestration	1 (14.3)	7 (50.0)	8 (38.1)
Left common carotid modification				0.337
Antegrade inner branch	4 (57.1)	3 (21.4)	7 (33.3)
Retrograde inner branch	2 (28.6)	3 (21.4)	5 (23.8)
Fenestration	0 (0)	6 (42.9)	6 (28.6)
No modification due to bovine arch	1 (14.3)	2 (14.3)	3 (14.3)
Left subclavian modification				0.206
Retrograde inner branch	2 (28.6)	9 (64.3)	11 (52.4)
Fenestration	2 (28.6)	2 (14.3)	4 (19.0)
In situ fenestration	1 (14.3)	0 (0)	1 (4.8)
Parallel graft	1 (14.3)	0 (0)	1 (4.8)
Covered	1 (14.3)	1 (7.1)	2 (9.5)
Prior fenestrated TEVAR	0 (0)	2 (14.3)	2 (9.5)
Cervical debranching	1 (14.3)	1 (7.1)	2 (9.5)	1.00
Cerebrospinal fluid drain	0 (0)	1 (7.1)	1 (4.8)	1.00
Rapid ventricular pacing	5 (71.4)	14 (100)	19 (90.5)	0.100
Fusion imaging	4 (57.1)	9 (64.3)	13 (61.9)	1.00
Carotid access				0.313
None	1 (14.3)	5 (35.7)	6 (28.6)
Bilateral	6 (85.7)	5 (35.7)	11 (52.4)
Right only	0 (0)	2 (14.3)	2 (9.5)
Left only	0 (0)	2 (14.3)	2 (9.5)
Total procedure time (median)	339 [125]	231 [127]	254 [173]	0.046
Fluoroscopy time (median)	56.0 [17.9]	24.2 [19.4]	34.0 [36.9]	0.012
Iodinated contrast (median)	80 [49]	85 [45]	80 [35]	0.547
Radiation exposure mGy (median)	800 [104]	776 [350]	778 [283]	0.859
Estimated blood loss (median)	700 [2800]	100 [163]	100 [400]	0.003
Technical success	6 (85.7)	12 (85.7)	18 (85.7)	1.00

Abbreviation: TEVAR, thoracic endovascular aortic repair.

Technical success was achieved in 86% of the early technique group and 86% of the current technique group (p=1.0). The technical failure in the early technique group was due to deployment of a bridging stent external to the aortic endograft rather than through the intended inner branch, which was rescued with parallel grafting (Figure 2A). Two patients in the current technique group had technical failures, including an innominate bridging stent that was placed into the false lumen of a dissected innominate artery (Figure 2B and C) and a misaligned innominate artery fenestration causing persistent perfusion of an ascending aortic pseudoaneurysm (Figure 2D).

Early Outcomes

The median length of intensive care unit stay was 4.0 (IQR=8.0) vs 3.0 (IQR=4.3) days and total hospital length of stay 9.0 (IQR=6.0) vs 5.5 (IQR=12.3) days for the early technique and current technique groups, respectively. Stroke within 30 days occurred in 2 patients (29%) in the early technique and 1 patient (7%) in the current technique group (p=0.247). Spinal cord ischemia occurred in 1 patient in both groups (14% vs 7%, p=1.0). The first patient experienced immediate postoperative bilateral lower extremity paralysis following zone 0 to 3 PM-FBEVAR that did not improve on follow-up. The second patient experienced delayed bilateral lower extremity paralysis with bowel and bladder incontinence following an episode of hypotension 10 days after zone 0 to 9 PM-FBEVAR, with improvement to ambulating with assistance on follow-up. The discharge location was most commonly home for the current group (57%), whereas the early group was as likely to discharge home as to a short-term nursing facility or acute rehabilitation unit (29 and 29%). In-hospital death was significantly higher in the early technique group at 43% compared with no deaths in the current group (p=0.026). The in-hospital deaths occurred due to intraoperative rupture of the aorta during aortic endograft advancement in 1 patient and intraoperative strokes in 2 patients. All 3 in-hospital mortalities occurred in female patients (Table 3).

Table 3.

Early Outcomes Following Physician-Modified Fenestrated-Branched Endovascular Aortic Repair.

Outcome	Early technique (N=7) (%)	Current technique (N=14) (%)	Total (N=21) (%)	p-value
ICU length of stay days (median)	4.0 [8.0]	3 [4.3]	3 [4.5]	0.743
Hospital length of stay days (median)	9.0 [6.0]	5.5 [12.3]	7 [8.5]	0.287
Major adverse event	5 (71.4)	5 (35.7)	10 (47.6)	0.183
Myocardial infarction	1 (14.3)	0 (0)	1 (4.8)	0.333
Respiratory failure	1 (14.3)	1 (7.1)	2 (9.5)	1.00
Stroke	2 (28.6)	1 (7.1)	3 (14.3)	0.247
Spinal cord ischemia	1 (14.3)	1 (7.1)	2 (9.5)	1.00
Acute kidney injury	3 (42.9)	3 (21.4)	6 (28.6)	0.354
New onset dialysis	0 (0)	0 (0)	0 (0)	N/A
Access complication	3 (42.9)	2 (14.3)	5 (23.8)	0.280
Discharged	4 (57.1)	14 (100)	18 (85.7)	1.00
Home	2 (28.6)	8 (57.1)	10 (47.6)
Short-term nursing facility	1 (14.3)	3 (21.4)	4 (19.0)
Acute rehabilitation unit	1 (14.3)	2 (14.3)	3 (14.3)
Other hospital	0 (0)	1 (7.1)	1 (4.8)
In-hospital death	3 (42.9)	0 (0)	3 (14.3)	.026

Abbreviation: ICU, intensive care unit.

Midterm Outcomes

The median follow-up was 7.6 (IQR=18.4) months for the early technique and 5.6 (IQR=11.4) months for the current technique group (p=0.856). Endoleaks occurred at similar rates between the 2 groups (29% vs 36%). In the early group, there was 1 type IC endoleak at the left subclavian artery bridging stent and 1 type IIIB endoleak suspected to be from diameter reducing suture holes that resolved on follow-up. There was 1 type IA noted intra-operatively that resolved on follow-up, a type II from an innominate artery following reintervention with coverage of a misaligned innominate fenestration and carotid-carotid bypass, a type IIIB suspected to be from spine trigger wire holes, and 2 type IIIC endoleaks at the left subclavian stent in 1 patient and from an unstented innominate artery fenestration in another patient in the current technique group (Table 4). One reintervention was required in the early technique group with stenting of an innominate artery for bridging stent stenosis. Four reinterventions were required for the current technique group including innominate artery stenting for a bridging stent that was deployed in the false lumen, open surgical arch repair for a new pseudoaneurysm and for persistent endoleak and filling of a pseudoaneurysm from misaligned unstented innominate fenestration, and right subclavian steal syndrome due to dissection limiting flow to the right subclavian artery.

Table 4.

Midterm Outcomes.

Outcome	Early technique (N=7) (%)	Current technique (N=14) (%)	Total (N=21) (%)	p-value
Follow-up months (median)	7.6 [18.4]	5.6 [11.4]	6.3 [13.0]	0.856
Endoleak	2 (28.6)	5 (35.7)	7 (33.3)
Type IA	0 (0)	1 (7.1)	1 (4.8)	1.00
Type IB	0 (0)	0 (0)	0 (0)	N/A
Type IC	1 (14.3)	0 (0)	1 (4.8)	0.333
Type II	0 (0)	2 (14.3)	2 (9.5)	0.533
Type IIIA	0 (0)	0 (0)	0 (0)	N/A
Type IIIB	1 (14.3)	1 (7.1)	2 (9.5)	1.00
Type IIIC	0 (0)	2 (14.3)	2 (9.5)	1.00
Total target vessels	19	38	57
Target vessel patency	100%	100%	100%
Target vessel instability	5.3%	0%	1.8%
Aortic rupture	1 (14.3)	0 (0)	1 (4.8)	0.333
Reintervention	1 (14.3)	4 (28.6)	5 (23.8)	0.624
Target vessel stenosis	1 (14.3)	0 (0)	1 (4.8)
False lumen deployment of bridging stent	0 (0)	1 (7.1)	1 (4.8)
Additional arch repair for pseudoaneurysm	0 (0)	2 (14.3)	2 (9.5)
Revascularization for subclavian steal syndrome	0 (0)	1 (7.1)	1 (4.8)
All-cause mortality	7 (100)	1 (7.1)	8 (38.1)	<0.001

All-cause mortality during follow-up was significantly different between the groups at 100% for the early technique group and 7% for the current technique group (p<0.001).

Discussion

This study demonstrates that PM-FBEVAR for the aortic arch is feasible, albeit with a steep and ongoing learning curve, marked with severe complications. The difference of in-hospital mortality of 43% compared with 0% between the groups reflects important changes in technical approach, concentrated experience at the institution, and patient selection by the dedicated multidisciplinary aortic team. The difference in all-cause mortality of 100% vs 7% is likely due to the combination of higher in-hospital mortality in the early group and improved patient selection in the current group. The technical success of 86% in both groups supports the feasibility of endovascular arch repair but also highlights the technical challenge of repairs within the arch compared with the visceral renal aorta, especially in patients with difficult anatomy. Similar to fenestrated-branched repairs of the visceral renal aorta, endoleaks and reinterventions remain common. Due to the limited number of events in this study, no particular pattern of endoleak or reintervention was elucidated, and additional investigation will be needed to determine relevant risk factors as further experience is gained.

Open aortic arch repair remains the standard of care for aortic arch disease and can be performed with excellent outcomes in several centers.¹⁹ As such, all patients in this series were deemed prohibitively high risk for open arch repair by a team of cardiothoracic surgeons with extensive experience in open ascending and aortic arch repairs. During this study, our center was participating in 2 zone 0 arch-branched device trials, and all patients were initially screened for eligibility in these trials before proceeding with PM-FBEVAR. Enrollment in these trials was challenging due to strict inclusion criteria and typically not feasible for symptomatic patients who need urgent repair, which was common in this series. Physician-modified fenestrated-branched endovascular aortic repair is a solution to these constraints by providing an additional option for total endovascular arch repair.

As endovascular aortic arch repair is still in the early stages of experience and this series begins with the first total endovascular arch repair at our center, improvements in technical approach were observed during the series. Our initial approach was to replicate the Cook Medical custom-manufactured devices as closely as possible due to the encouraging early results from European centers using Cook Arch Branch and Arch Fenestrated stent grafts (Cook Medical).^20,21 Therefore, early arch PM-FBEVAR devices were built with antegrade inner branches for the innominate and left carotid, and retrograde outer/inner branches for the left subclavian with an off-set 1 to 2 cm away from the vessel origin to allow working space for branch cannulation. Permanent constraining sutures were added along the working space (Figure 3A). Other early designs used large unstented fenestrations for the innominate and left carotid, and stented fenestration or branches for the left subclavian to replicate the Cook Medical or Najuta Fenestrated TEVAR devices.⁷ A permanent wire was added along the outer curve (Figure 3B). These early designs were focused on tolerance to radial misalignment that were anticipated during deployment. Two notable changes were incorporated in the design in the current technique group. First, one of the trigger wires was rerouted along the outer curve of the stent graft as the self-orienting “spine” wire as demonstrated by Lee et al due to their success with custom-manufactured and physician-modified aortic arch endografts that used a similar design.¹¹ Combined with the pre-curved nose cone of the Zenith Alpha thoracic endograft, incorporation of the spine wire enhanced predictability of rotational alignment during deployment. This in turn led to changing branch modifications to the precise location of the arch vessels without intentional off-set (Figure 3C). In our experience, these changes improved our ability to quickly and precisely align the branch vessels with the device, which made cannulation of the branch vessels easier with decreased wire and catheter manipulation. This is reflected by the decrease in total procedure time and fluoroscopy time of the current technique group and may have aided in decreasing the stroke rate as increased manipulation of the aortic arch branches with wires and catheters has been previously associated with increased risk of stroke.²²

Despite these overall improvements, there are several important lessons we learned through this initial experience. First, placement of innominate and left carotid inner branches close together in tapered devices can lead to difficult re-sheathing and deployment. This can result in inaccurate and slow deployment, which can induce cardiac failure from prolonged aortic insufficiency in the setting of induced ventricular tachycardia. Second, the spine trigger wire should not be reinserted back into the nose cone if its original position is not exactly at the outer curve. This can cause misalignment during retrieval, leading to branch shuttering and type IIIC endoleak. Third, cutting the pre-curved nose cone can lead to increased risk of stroke and device misalignment. In 1 patient, the nose cone was cut to prevent the device from crossing the aortic valve. This patient experienced diffuse ischemic stroke leading to an in-hospital mortality, which we believe may have been caused by the cut proximal end of the device creating emboli as it moved along aortic mural thrombus. Therefore, we routinely cross the aortic valve. Finally, there is the close collaboration between vascular and cardiac surgery. At our center, all TEVARs are performed with vascular and cardiac surgeons as cosurgeons. Vascular surgery provides expertise in device sizing, access vessel management, and target vessel management, whereas cardiac surgery provides expertise in hemodynamic support maneuvers, crossing the aortic valve, management of arrhythmias, and hemodynamic instability. In our experience, arch PM-FBEVAR requires merging of these skill sets.

Although arch PM-FBEVAR has become an important treatment option for patients who are high surgical risk and do not have access to dedicated manufactured devices, our experience suggests a significant learning curve marked with occurrence of devastating complications including stroke and death. This was also demonstrated by the results of early feasibility trials and in “real-world” Vascular Quality Initiative data.^4,8,23 Much work remains to determine the optimal configuration of the arch fenestrated-branched endografts and in stroke mitigation strategies.

This study is limited by the small sample size which decreases statistical power. The retrospective nature of the study does not allow for standardization within the 2 groups and introduces bias. Results of the comparison between the 2 groups must also factor in the learning curve that was encountered during the series, which may be increasing the differences seen between the 2 groups.

Conclusion

There is a significant learning curve associated with PM-FBEVAR for the aortic arch. This study suggests that gained experience, use of a spine trigger wire, and an anatomically specific device design can facilitate branch alignment, leading to a shorter procedure and fewer complications.

Footnotes

Acknowledgements

The senior author thanks Drs Carlos Timaran and Mirza Baig for sharing their technique of incorporating a spine trigger wire during arch PM-FBEVAR.

Presentation Information

This study was presented at the 38th Western Vascular Society Annual Meeting, Kauai, Hawaii, September 9 to 12, 2023.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: S.M.H. is a consultant for consultant for W. L. Gore & Associates, Cook Medical, Terumo Aortic, and Vestek and is on the scientific advisory board for W. L. Gore & Associates and Vestek. F.F. is a consultant for W. L. Gore & Associates, Cook Medical, Terumo, and Artivion.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Alexander D. DiBartolomeo

Fernando Fleischman

Sukgu M. Han

References

Harky

Kai Chan

Ming Wong

, et al Open versus endovascular repair of descending thoracic aortic aneurysm disease: a systematic review and meta-analysis. Ann Vasc Surg. 2019;54:304–315.

Kärkkäinen

Pather

Tenorio

, et al Should endovascular approach be considered as the first option for thoraco-abdominal aortic aneurysms. J Cardiovasc Surg (Torino). 2019;60(3):298–312.

DiBartolomeo

Pyun

Ding

, et al Comparative outcomes of physician-modified fenestrated-branched endovascular repair of post-dissection and degenerative complex abdominal or thoracoabdominal aortic aneurysms. J Vasc Surg. 2023;78:565–574e2.

Tazaki

Inoue

Higami

, et al Thoracic endovascular aortic repair with branched Inoue Stent Graft for arch aortic aneurysms. J Vasc Surg. 2017;66(5):1340–1348.

Haulon

Greenberg

Spear

, et al Global experience with an inner branched arch endograft. J Thorac Cardiovasc Surg. 2014;148(4):1709–1716.

Czerny

Rylski

Morlock

, et al Orthotopic branched endovascular aortic arch repair in patients who cannot undergo classical surgery. Eur J Cardiothorac Surg. 2018;53:1007–1012.

Iwakoshi

Ichihashi

Itoh

, et al Clinical outcomes of thoracic endovascular aneurysm repair using commercially available fenestrated stent graft (Najuta endograft). J Vasc Surg. 2015;62(6):1473–1478.

Liang

Dake

Fischbein

, et al Midterm outcomes of endovascular repair of aortic arch aneurysms with the gore thoracic branch endoprosthesis. Eur J Vasc Endovasc Surg. 2022;64(6):639–645.

Hogendoorn

Schlösser

Moll

, et al Thoracic endovascular aortic repair with the chimney graft technique. J Vasc Surg. 2013;58:502–511.

10.

Chan

Jia

, et al Methods and clinical outcomes of in situ fenestration for aortic arch revascularization during thoracic endovascular aortic repair. Vascular. 2020;28(4):333–341.

11.

Lee

Porras-Colon

Scott

, et al Early results and feasibility of total endovascular aortic arch repair using 3-vessel company-manufactured and physician-modified stent-grafts [published online ahead of print March 28, 2023] J Endovasc Ther. doi:10.1177/15266028231163069.

12.

Zhu

Dai

Noiniyom

, et al Fenestrated thoracic endovascular aortic repair using physician-modified stent grafts (PMSGs) in zone 0 and zone 1 for aortic arch diseases. Cardiovasc Intervent Radiol. 2019;42(1):19–27.

13.

Kuo

Huang

Chen

JS.

Handmade fenestrated stent grafts to preserve all supra-aortic branches in thoracic endovascular aortic repair. J Thorac Cardiovasc Surg. 2020;160(3):629–639.

14.

Zhang

Song

, et al Outcomes of total endovascular aortic arch repair with surgeon-modified fenestrated stent-grafts on zone 0 landing for aortic arch pathologies. J Endovasc Ther. 2022;29(1):109–116.

15.

Oderich

Forbes

Chaer

, et al Reporting standards for endovascular aortic repair of aneurysms involving the renal-mesenteric arteries. J Vasc Surg. 2021;73:4s–52s.

16.

Kuo

Han

SM.

Treatment of complex thoracoabdominal aortic disease. Cardiol Clin. 2017;35:411–429.

17.

Pyun

Zhang

Magee

, et al Use of inner branches during physician-modified endografting for complex abdominal and thoracoabdominal aortic aneurysms. Ann Vasc Surg. 2021;76:244–253.

18.

Han

Tenorio

Mirza

, et al Low-profile Zenith Alpha™ thoracic stent graft modification using preloaded wires for urgent repair of thoracoabdominal and pararenal abdominal aortic aneurysms. Ann Vasc Surg. 2020;67:14–25.

19.

Shrestha

Bachet

Bavaria

, et al Current status and recommendations for use of the frozen elephant trunk technique: a position paper by the Vascular Domain of EACTS. Eur J Cardiothorac Surg. 2015;47(5):759–769.

20.

Tenorio

Oderich

Kölbel

, et al Multicenter global early feasibility study to evaluate total endovascular arch repair using three-vessel inner branch stent-grafts for aneurysms and dissections. J Vasc Surg. 2021;74(4):1055–1065.

21.

Tsilimparis

Law

Rohlffs

, et al Fenestrated endovascular repair for diseases involving the aortic arch. J Vasc Surg. 2020;71(5):1464–1471.

22.

Buth

Harris

Hobo

, et al Neurologic complications associated with endovascular repair of thoracic aortic pathology: incidence and risk factors. a study from the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) registry. J Vasc Surg. 2007;46(6):1103–1110.

23.

DiBartolomeo

Ding

Weaver

, et al Risk of stroke with thoracic endovascular aortic repair of the aortic arch. Ann Vasc Surg. 2023;97:37–48.