Off-the-Shelf Devices for Treatment of Thoracic Aortic Diseases: Midterm Follow-up of TEVAR With Chimneys or Physician-Made Fenestrations

Abstract

Purpose: To evaluate the midterm outcomes of thoracic endovascular aortic repair (TEVAR) using chimney grafts (ch-TEVAR) or thoracic stent-grafts with fenestrations made on the back table (f-TEVAR) to treat thoracic aortic dissection (TAD) and thoracic aortic aneurysm (TAA). Materials and Methods: A retrospective analysis was conducted of 474 consecutive patients (mean age 62.3±10.7 years; 346 men) treated with either f-TEVAR (n=110) or ch-TEVAR (n=364) for 352 TADs (81 f-TEVAR and 271 ch-TEVAR) or 122 TAAs (29 f-TEVAR and 93 ch-TEVAR) from 2008 to 2016. The primary endpoints at 30 days and during follow-up were overall mortality, aorta-related mortality, and major complications. The secondary endpoints were endoleak and reintervention. The patency of the target branches, cost of hospitalization, and the use of antiplatelet drugs were also analyzed. Results: Intraoperative type I endoleaks were treated in 69 (14.6%) cases (4 f-TEVAR and 65 ch-TEVAR, p<0.01) to achieve 100% technical success. Four (0.8%) patients died within 30 days [1 (0.9%) f-TEVAR and 3 (0.8%) ch-TEVAR]. Perioperative cerebral ischemia (1 fatal stroke) occurred in 9 (1.9%) patients: (1 f-TEVAR and 8 ch-TEVAR, p=0.39). During the mean follow-up of 50.6±20.0 months (49.5±18.3 months in f-TEVAR and 50.9±20.6 months in ch-TEVAR), 11 (2.3%) patients died of an aorta-related event. Type I endoleak was present in 40 (8.4%) patients (1 f-TEVAR and 39 ch-TEVAR, p<0.01). Eleven (2.3%) patients experienced stent-graft migration and 13 (2.7%) had a retrograde dissection. One hundred (16.9%) of the 593 branch stents occluded (4/75 in the f-TEVAR group and 96/518 in the ch-TEVAR group, p<0.01). The branch reintervention rate was 7.2% (34/474). The f-TEVAR group had a significantly higher probability of freedom from branch occlusion (92%) than the ch-TEVAR group (83%, p=0.007). Conclusion: Off-the-shelf techniques employing chimney grafts and homemade fenestrations are both suitable options for TAD and TAA involving the supra-aortic branches, with a low incidence of reintervention. Fenestrated TEVAR seems to have more favorable short- and midterm outcomes. Further study of these off-the-shelf techniques for aortic arch repair is warranted.

Keywords

aortic aneurysm aortic arch aortic dissection chimney graft covered stent endograft fenestration physician-modified stent-graft stent surgeon-modified stent-graft thoracic endovascular aortic repair thoracic stent-graft

Introduction

Thoracic aortic dissection (TAD) and thoracic aortic aneurysm (TAA) involving the aortic arch or its branches are not suitable for standard thoracic endovascular aortic repair (TEVAR), so the procedure is combined with debranching of the supra-aortic vessels¹ or fenestrated/branched stent-grafts.²

TEVAR has been widely used in China,^3–7 but no customized fenestrated stent-grafts are available. Thoracic aortic pathologies involving the arch have been treated mainly with chimney TEVAR (ch-TEVAR) or other parallel graft techniques^3,5,7 or on-site construction of fenestrations and branches in commercially available stent-grafts.^6,8,9 Our group has reported the satisfactory short-term outcomes of ch-TEVAR and thoracic stent-grafts with fenestrations made on the back table (f-TEVAR) in treating TAD and TAA involving the aortic arch.^10,11 The current retrospective analysis updates those early reports by examining a larger cohort treated with either f-TEVAR or ch-TEVAR for aortic dissections or aneurysms involving the supra-aortic branches or branches in need of coverage to achieve a sufficient landing zone.

Materials and Methods

Study Design and Patient Sample

A retrospective study was conducted of all consecutive patients who underwent endovascular treatment for TAD or TAA involving the aortic arch branches between January 2008 and June 2016. Indications for treatment were acute (<2 weeks after symptom onset), subacute (between 2 weeks and 3 months after symptoms), or chronic (>3 months) type B TAD and >6-cm-diameter TAA (or symptomatic presentation). The symptoms of TAD and TAA included chest pain, pulmonary infection, arrhythmia, dyspnea, dysphagia, hoarseness, or enlargement of the aneurysm or false lumen in TAD. Excluded were cases of penetrating aortic ulcers or intramural hematoma, TEVAR using custom-made branched stent-grafts, or hybrid procedures incorporating bypass grafts to reconstruct the supra-aortic branches in combination with TEVAR. All patients with TAD or TAA underwent computed tomography angiography (CTA) to confirm the diagnosis.

In the study period, 1528 TAD patients and 497 TAA patients were admitted for different types of treatment. After applying the exclusion criteria, 474 patients (mean age 62.3±10.7 years; 346 men) remained for analysis: 364 patients (mean age 61.9±10.9 years; 259 men) who underwent ch-TEVAR (271 TAD and 93 TAA) and 110 patients (mean age 63.6±11.0 years; 87 men) who received f-TEVAR (81 TAD and 29 TAA). Characteristics of the patients are listed in Table 1.

Table 1.

Patient Characteristics.^a

Variable	Total (n=474)	f-TEVAR (n=110)	ch-TEVAR (n=364)	p^b
Age, y	62.3±10.7	63.6±11.0	61.9±10.9	0.18
Men	346 (73.0)	87 (79.1)	259 (71.2)	0.10
Symptom
Chest pain	59 (12.4)	14 (12.7)	45 (12.4)	0.92
Pulmonary infection	22 (4.6)	5 (4.5)	17 (4.7)	0.96
Arrhythmia	30 (6.3)	8 (7.3)	22 (6.0)	0.64
Dyspnea	41 (8.6)	11 (10.0)	30 (8.2)	0.57
Dysphagia	18 (3.8)	4 (3.6)	14 (3.8)	0.92
Hoarseness	14 (3.0)	3 (2.7)	11 (3.0)	0.87
Asymptomatic	290 (61.2)	65 (59.1)	225 (61.8)	0.61
Comorbidities
Coronary heart disease	190 (40.1)	49 (45.8)	141 (38.7)	0.28
Ischemic stroke	40 (8.4)	11 (10.3)	29 (8.0)	0.50
Atrial fibrillation	125 (26.4)	31 (28.2)	94 (25.8)	0.63
Hypertension	358 (75.5)	84 (78.5)	274 (75.3)	0.82
Chronic obstructive pulmonary disease	73 (15.4)	19 (17.8)	54 (14.8)	0.53
Chronic renal dysfunction	104 (21.9)	26 (33.6)	78 (21.4)	0.62
Diabetes mellitus	96 (20.3)	25 (23.4)	71 (19.5)	0.46
Emergency	20 (4.2)	3 (2.8)	17 (4.7)	0.37
Target arch vessels
1	296 (62.4)	65 (59.1)	231 (63.5)	0.41
2	147 (31.0)	35 (31.8)	112 (30.8)	0.83
3	31 (6.5)	10 (9.1)	21 (5.8)	0.22

Abbreviations: ch-TEVAR, chimney thoracic endovascular aortic repair; f-TEVAR, fenestrated thoracic endovascular aortic repair; TAA, thoracic aortic aneurysm; TAD, thoracic aortic dissection.

Continuous data are presented as mean ± standard deviation; categorical data are given as the number (percentage).

f-TEVAR vs ch-TEVAR.

The study was performed according to the regulations of the institutional ethics committee. Because of the retrospective design, written informed consent for this study was not mandatory; all patients provided consent for the intervention.

Intervention

The selection of ch-TEVAR or f-TEVAR was at the discretion of the vascular surgeon. In case of emergent endovascular treatment, ch-TEVAR was typically used because of the time needed to perform calculations for f-TEVAR.

For all cases, 3-dimensional reconstructions of the CTA scans were generated on a workstation (Leonardo; Siemens, Erlangen, Germany). The relevant lengths, diameters, angles, lesions, and relationships with the supra-aortic branches were measured based on a center lumen line. The stent-grafts selected based on these measurements were oversized 0% to 5% for TADs and 15% to 20% for TAAs.¹²

The main stent-grafts and branch stents selected are shown in Table 2. Six different thoracic stent-grafts were used: Valiant (Medtronic Vascular, Santa Rosa, CA, USA); Ankura (Lifetech Scientific Corporation, Shenzhen, China); Zenith (Cook Medical, Bloomington, IN, USA); Hercules (MicroPort, Shanghai, China); TAG (Gore Medical, Flagstaff, AZ, USA); and Relay (Terumo Aortic/Bolton Medical, Sunrise, FL, USA). The bridging or chimney devices included covered stents [Fluency (BD/Bard Peripheral Vascular, Tempe, AZ, USA) or Viabahn (Gore Medical)] and bare stents [Sinus Superflex (Optimed, Ettlingen, Germany); S.M.A.R.T. (Cordis Corporation, a Cardinal Health company, Santa Clara, CA, USA); Complete SE (Medtronic Vascular); and Luminexx (BD/Bard Peripheral Vascular)].

Table 2.

Characteristics of the Aortic Stent-Grafts and Chimney/Bridging Devices.^a

Variable	f-TEVAR	ch-TEVAR	p
Oversizing, %			0.918
TAD	3.1±1.3	3.1±1.3
TAA	17.8±1.2	17.6±1.3
Aortic stent-graft model			0.655
Valiant	93/110 (84.5)	112/364 (30.8)	<0.001
Ankura	12/110 (10.9)	21/364 (5.8)	0.063
Zenith	5/110 (4.5)	98/364 (26.9)	<0.001
Hercules	0	29/364 (8.0)	0.002
TAG	0	96/364 (26.4)	<0.001
Relay	0	8/364 (2.2)	0.117
Bridging or chimney devices
Covered stent
Fluency	19/75 (25.3)	258/518 (49.8)	<0.001
Viabahn	29/75 (38.7)	260/518 (50.2)	0.062
Bare stent
Superflex	10/75 (13.3)	0	<0.001
Cordis	6/75 (8.0)	0	<0.001
Complete SE	4/75 (5.3)	0	<0.001
Luminexx	7/75 (9.3)	0	<0.001

Abbreviations: ch-TEVAR, chimney thoracic endovascular aortic repair; f-TEVAR, fenestrated thoracic endovascular aortic repair; TAA, thoracic aortic aneurysm; TAD, thoracic aortic dissection.

Continuous data are presented as mean ± standard deviation; categorical data are given as the number/sample (percentage).

Back Table Fenestrations

The branch vessel diameters, lengths, angles to the arch, clock positions, and relative relationships were measured on the CTA reconstructions (Figure 1A). The diameter of the aorta and its relationship with the midline of the outer curve at the supra-aortic branches, the aortic arch angle, the space between the branches, and the length of the landing zone at the outer curve were clearly defined (Figure 1B), along with the spatial relationship between the branches and the main stent-graft (Valiant, Ankura, or Zenith) in order to develop the preoperative plan for the location of the fenestrations (Figure 1C).⁹

Two types of fenestrations were made: small for a single branch (distance between the branch and lesion <15 mm; Figure 2) or large for 2 or 3 branches (distance between the branch and lesion >15 mm; Figure 3). The diameters of both types of fenestrations were matched to the branch diameters.

After general anesthesia, the common femoral artery was exposed via a cutdown or accessed percutaneously. Angiography was performed to identify the branches. After restudying both the angiogram and the CTA images, the selected thoracic stent-graft was partially released on a back table in the operating room (Video 1). According to the predesigned plan, the fenestration site was marked at the specified position on the stent-graft, with the flush port corresponding to the posterior midline pointing up. The fabric was cut away with a pair of scissors; the stent-graft was carefully handled so as not to damage the struts or the post-release device (Video 2). After fenestration was complete (Video 3), the stent-graft was resheathed in the delivery system (Video 4), taking care to compress only the fabric and keep the axis of the flush port centered to guarantee correct alignment of the stent-graft. When the fenestrated stent-graft was delivered, the flush port was turned 180° to align the fenestration with the outer curve of the arch. The position of the fenestration relative to the branch artery was verified based on the stent-graft’s metallic marker. After placement of the stent-graft in the designated position, the device was deployed; its position and patency of the branches were confirmed by angiography.

For small fenestrations, the selection of the type of stents deployed in the branches depended mainly on the results of intraoperative angiography. If endoleak around the fenestration was confirmed by angiography, covered stents were used, while bare stents were employed when no endoleak was observed.

After surgery, patients with a single bare stent in the branch were prescribed aspirin, while patients with a covered stent or 2 bare stents were given aspirin and clopidogrel. Patients who did not receive a branch stent had no additional antiplatelet therapy to accompany the existing cardiovascular risk prevention medication.

Chimney TEVAR

Under general anesthesia, open access was obtained to the common femoral artery as well as to the brachial or carotid arteries for delivery of the chimney grafts. The main thoracic stent-graft was sized according to the formula of Chou et al,¹³ which was based on the principle of isoperimetric inequality and the known diameters of the aorta and chimney stents. Chimney TEVAR was performed as described previously using covered stents exclusively for the chimneys.¹⁰ The proximal end of a chimney covered stent was sealed on relatively normal vessel wall at least 1.5 cm from the aneurysm or the proximal end of the tear; its length exceeded the proximal end of the main stent-graft by about 0.5 to 1 cm (free flare zone). The sealing zone between the chimney and the main stent-graft was 0.5 to 1 cm. In case of multiple chimneys, the stents were placed on the same side.

To reduce the gutter and seal any proximal type I endoleak, the main and chimney grafts were expanded simultaneously using kissing balloons. If type I endoleak or filling of a false lumen in TAD was still visible on angiography, embolization (coil or glue) or a proximal cuff was used.^4,10 Dual antiplatelet therapy was administered after the surgical procedure.

The follow-up protocol for both procedures featured clinical examination and CTA performed at 3, 6, and 12 months and annually thereafter. If the false lumen in TAD patients increased during follow-up, the gutter between the main and chimney grafts was thrombosed using either coils or glue if the endoleak was limited (<5 mm). If the endoleak was >5 mm, coils were delivered first followed by glue. If the chimney stent was misplaced or needed to be extended, additional stents were used.

Definitions and Endpoints

Technical success was defined as successful deployment of stent-graft to the planned location and complete exclusion of the proximal entry tear in TAD or aneurysm sac in the case of TAA, maintained patency of the target branch vessel, and no type I endoleak.

The primary endpoints were overall mortality and aorta-related mortality at 30 days and during follow-up and major complications (stroke, paraplegia, or new dissection or aneurysm) at 30 days and during follow-up. Secondary endpoints were endoleaks and reintervention. Target branch patency was assessed during follow-up. The cost of hospitalization and the use of antiplatelet drugs were analyzed.

Statistical Analysis

Continuous data are presented as the mean ± standard deviation; groups were compared using the t test for normally distributed data or the Mann-Whitney U test for nonnormally distributed variables. Categorical data are given as the number (percentage) and were compared in contingency tables using the Fisher exact test. The endpoints were analyzed using the Kaplan-Meier method; the curves were compared using the log-rank test. The estimates are presented with the 95% confidence interval (CI). The threshold of statistical significance was a 2-tailed p<0.05. Statistical analyses were performed using SPSS (version 22; IBM Corporation, Armonk, NY, USA) and Empower Stats (http://www.empowerstats.net/en/index.html) software.

Results

Perioperative Outcomes

The intended procedure was performed successfully in all patients. The operative time was 81.0±18.9 minutes in total, 10.4 minutes longer in the f-TEVAR group (89.0±14.7 minutes) than the ch-TEVAR patients (78.6±19.3 minutes, p<0.01). Intraoperative type I endoleak was treated in 37 TAD and 32 TAA cases in 65 (17.9%) of the 364 ch-TEVAR patients and 4 (3.6%) of the 110 f-TEVAR patients (p<0.01; Table 3). Ancillary procedures included additional branch stents, embolotherapy, or cuffs in 63 patients (4 f-TEVAR and 59 ch-TEVAR, p<0.01). The proportion of additional endovascular techniques used in the ch-TEVAR group (16.2%) was higher than in f-TEVAR (3.6%, p<0.01), mainly related to greater use of cuffs (n=15) and coil embolization (n=31). Postoperatively, 43.6% patients in the f-TEVAR group used 2 antiplatelet drugs compared with 100% in the ch-TEVAR group (p<0.01). The costs in the ch-TEVAR group were higher than in the f-TEVAR group (137.2±15.5 vs 117.6±13.9 thousand RMB, p<0.01).

Table 3.

Intraoperative, Perioperative, and Midterm Follow-up of Off-the-Shelf TEVAR for the Entire Cohort and by Subgroup.^a

Variable	Total	f-TEVAR	ch-TEVAR	p^b
Intraoperative
Landing zone, mm	24.9±10.9	24.5±10.6	25.0±10.9	0.64
Type I endoleak	69/474 (14.6)	4/110 (3.6)	65/364 (17.9)	<0.01
TAD	37/352 (10.5)	1/81 (1.2)	36/271 (13.3)	<0.01
TAA	32/122 (26.2)	3/29 (10.3)^c	29/93 (31.2)^d	0.03
Additional procedures	63/474 (13.3)	4/110 (3.6)	59/364 (16.2)	0.00
Stents (arch only, not visceral)	11/474 (2.3)	4/110 (3.6)^e	7/364 (1.9)	0.30
Coil embolization	31/474 (6.5)	0/110	31/364 (8.5)	<0.01
Onyx embolization	6/474 (1.3)	0/110	6/364 (1.6)	0.18
Cuff	15/474 (3.2)	0/110	15/364 (4.1)	0.03
Procedure duration, min	81.0±18.9	89.0±14.7	78.6±19.3	<0.01
Cost, ×10³ RMB yuan	132.6±17.2	117.6±13.9	137.2±15.5	<0.01
Perioperative
In-hospital cerebral ischemia	9/474 (1.9)	1/110 (0.9)	8/364 (2.2)	0.39
Stroke	1/474 (0.2)	0/110	1/364 (0.3)	0.58
TIA	8/474 (1.7)	1/110 (0.9)	7/364 (1.9)^f	0.47
Antiplatelet drug
Aspirin or clopidogrel	27/474 (5.7)	27/110 (24.5)	0/364	<0.01
Aspirin and clopidogrel	412/474 (86.9)	48/110 (43.6)	364/364 (100)	<0.01
In-hospital mortality	4/474 (0.8)	1/110 (0.9)	3/364 (0.8)	0.93
Follow-up, mo	50.6±20.0	49.5±18.3	50.9±20.6	0.53
TAD or TAA-related mortality	11/474 (2.3)	3/110 (2.7)	8/364 (2.2)	0.75
Mortality overall	51/474 (10.8)	12/110 (10.9)	39/364 (10.7)	0.95
Type I endoleak	40/474 (8.4)	1/110 (0.9)	39/364 (10.7)	<0.01
TAD	18/352 (5.1)	0/81(0)	18/271 (6.6)	0.02
TAA	22/122 (18.0)	1/29 (3.4)	21/93 (22.6)^g	0.02
Type I endoleak reintervention	36/474 (7.6)	0/110	36/364 (9.9)	<0.01
Coil embolization	30/474 (6.3)	0/110	30/364 (8.2)	<0.01
Onyx embolization	6/474 (1.3)	0/110	6/364 (1.6)	0.03
Diminished by reintervention	31/474 (6.5)	0/110	31/364 (8.5)	<0.01
Cerebral ischemia	22/474 (4.6)	4/110 (3.6)	18/364 (4.9)	0.57
Stroke	4/474 (0.8)	0/110	4/364 (1.1)	0.27
TIA	18/474 (3.8)	4/110 (3.6)^h	14/364 (3.8)ⁱ	0.92
Stent migration	11/474 (2.3)	3/110 (2.7)	8/364 (2.2)	0.75
Retrograde dissection	13/474 (2.7)	2/110 (1.8)	11/364 (3.0)	0.50
Branch occlusion
Per stent	100/593 (16.9)	4/75 (5.3)	96/518 (18.5)	0.01
Per patient	62/474 (13.1)	4/110 (3.6)	58/364 (15.9)	<0.01
Reintervention for branch occlusion	34/474 (7.2)	3/110 (2.7)	31/364 (8.5)	0.04
Paraplegia	1/474 (0.2)	0/110	1/364 (0.3)	0.58

Abbreviations: ch-TEVAR, chimney thoracic endovascular aortic repair; f-TEVAR, fenestrated thoracic endovascular aortic repair; RMB, Renminbi (the people’s currency); TAA, thoracic aortic aneurysm; TAD, thoracic aortic dissection; TIA, transient ischemic attack.

Continuous data are presented as the mean ± standard deviation; categorical data are given as the number (percentage).

Between f-TEVAR and ch-TEVAR unless otherwise noted.

p=0.025 compared with the TAD group.

p<0.001 compared with the TAD group.

The branch was partly blocked by the stent and blood flow was affected in 4 patients; the treatment was replaced by ch-TEVAR.

p=0.033 compared with the stroke group.

p<0.001 compared with the TAD group.

p=0.043 compared with the stroke group.

p=0.017 compared with the stroke group.

Mortality and Complications

Four (0.8%) patients died within 30 days. In the 110-patient f-TEVAR group, 1 (0.9%) patient with TAD suffered a retrograde dissection, which resulted in death on the third day. Of the 3 (0.8%) deaths among the 364 ch-TEVAR patients, one was due to cerebral infarction 3 days after the operation, another to retrograde dissection at 2 days, and the third to myocardial infarction 4 days after TEVAR.

Perioperative cerebral ischemia occurred in 9 (1.9%) patients: 1 f-TEVAR patient had a transient ischemic attack (TIA) with full recovery in follow-up vs 7 TIAs in the ch-TEVAR group (all treated and discharged), in addition to the fatal stroke noted above.

Midterm Outcomes

The mean follow-up was 50.6±20.0 months in the entire cohort: 49.5±18.3 months for the f-TEVAR group and 50.9±20.6 months in the ch-TEVAR group. Overall survival was 88% (95% CI 85.1% to 90.9%) in both groups (p=0.9; Figure 4A). In the entire cohort, aorta-related mortality was 2.3% (11/474). Two f-TEVAR patients died of retrograde dissection at 8 and 16 months in addition to the one in hospital. In the ch-TEVAR group, 7 patients died in follow-up in addition to the fatal retrograde dissection in hospital. Of these 7 late deaths, 2 were sudden deaths after complaining about chest tightness and uncontrollable hypertension that was believed to be related to TAD. One patient died of acute cerebral infarction, 3 patients died of retrograde dissection after open surgery, and 1 patient died of pulmonary infection subsequent to paraplegia.

During the entire follow-up period, 22 (4.6%) patients suffered cerebral ischemia (4 in the f-TEVAR group and 18 in the ch-TEVAR group, p=0.57). Stent migration and retrograde dissection occurred in 11 (2.3%) and 13 (2.7%) of all patients, respectively, without significant differences between treatment approaches. Type I endoleak was identified in 40 (8.4%) patients (only 1 f-TEVAR case): 18 TAD cases and 22 TAA cases of the total 474 patients. Thirty-six (7.6%) patients underwent reintervention for proximal type I endoleak, with diminished flow in 31.

One hundred (16.9%) of the 593 branch stents occluded (4/75 in the f-TEVAR group and 96/518 in the ch-TEVAR group, p<0.01). Of the 62 (13.1%) patients with branch occlusion, 28 were asymptomatic, and no reintervention was performed (7.2% branch reintervention rate). More covered stents in the f-TEVAR group were patent (45/48, 93.8%) than in the ch-TEVAR group (422/518, 81.5%). As for the bare stents used in f-TEVAR, the majority (26/27, 96.3%) were patent. The estimated patency of all branch stents was 83% (95% CI 79.8% to 86.3%) in the ch-TEVAR group and 92% (95% CI 85.7% to 98.3%) in the f-TEVAR group (p=0.007; Figure 4B). Thus, the probability of reintervention for branch occlusion was significantly higher in the ch-TEVAR group [10% (95% CI 6.8% to 13%)] than in f-TEVAR group [6% (95% CI 1.7% to 11%), p=0.04; Figure 4C].

Discussion

Endovascular repair of type A dissection involving the aortic arch has been performed in recent years in China,^9,10,14 but open repair is still favored in cases involving the ascending aorta, coronary arteries, or aortic valve. The treatment of TAD and TAA involving the aortic arch are high-risk procedures, but with the evolution in endovascular techniques in recent years, a less invasive option with relatively satisfactory clinical outcomes is available for these patients.^{2–7,11,14,15}

Ten years ago, our team started to use endovascular procedures, such as physician-customized fenestrations and chimney techniques, in the aortic arch with satisfactory clinical efficacy.^10,11 ch-TEVAR and f-TEVAR were used for patients with aortic dissections or aneurysms involving the origin of the supra-aortic branches or if a TEVAR landing zone required coverage of a supra-aortic branch. In emergency situations, ch-TEVAR was used in the majority of cases because of the time needed to make calculations for f-TEVAR.

Type I endoleak is a known and common complication of ch-TEVAR, with a reported incidence of 15% to 30%.^15–17 Our 15% and 8% rates of intraoperative and late endoleaks, respectively, were typical, and endoleaks were more often seen after TAA than TAD. The main source of these leaks is the gutter that arises from inadequate sealing between the parallel grafts.^18–20 Such gutters are difficult to seal completely with the current techniques²⁰; however, longer overlapping of chimney grafts with the main graft may reduce the gutter size.³

The fenestration technique, on the other hand, can result in complete sealing of the main graft and branches to the aortic wall and arteries, respectively.^11,21,22 Type Ia endoleak after f-TEVAR can arise from the gap between the proximal endograft and the aortic wall as no perfect apposition can sometimes be reached because of the rigidity of the endograft.²³ We recommend balloon molding for the aortic stent-graft and deploying a covered stent in small fenestrations to avoid endoleaks and to provide a better seal. In our experience, type I endoleak was infrequent after f-TEVAR, occurring primarily in TAA cases.

In this study, cerebral ischemic events were more frequent in ch-TEVAR than f-TEVAR. If we speculate about the cause, we think that long-term compression of the parallel stents by the main graft could cause the cerebral events. A high rate of branch occlusion could mean a higher possibility of cerebral ischemia.²⁴ However, in our study, there was no significant difference in the incidence of cerebral infarction or TIA between the 2 groups during hospitalization or follow-up. The cerebral ischemia events in the ch-TEVAR group were mainly TIAs, and stroke was rare. Since branch occlusion was mostly chronic, the body had enough time to establish collaterals to provide the distal cerebral blood supply. Thus, about half of these occlusions did not cause serious or acute cerebral ischemia. Therefore, even if the branch occlusion rate was high in the ch-TEVAR group, the incidence of cerebral ischemia was not significantly higher than that in the f-TEVAR group. However, patency might be improved in the f-TEVAR group if aspirin and clopidogrel were routinely prescribed (fewer than half of the f-TEVAR patients in this study received dual therapy). In ch-TEVAR patients, long-term compression should be avoided by using fewer chimneys and avoiding long stents (use short stent-grafts as much as possible).

In our study, the proportion of covered stents deployed in the branches in the f-TEVAR group was about half that of ch-TEVAR. The proportion of additional endovascular devices used in the procedure or during follow-up was also significantly lower in the f-TEVAR group, as was the hospitalization expense. The lower rates of arch stent-graft use and reintervention favored the f-TEVAR technique if compared with the ch-TEVAR.

There are no commercially available customized fenestrated stent-grafts in China, which is why we use physician-made fenestrations in our center. Other options in Western countries are custom-made² or off-the-shelf branch²⁵ devices, but mid- to long-term results are scarce, and off-the-shelf devices have limited anatomical suitability.²⁵ Although results are acceptable, high reintervention rates exist.^26,27 Direct comparison between custom-made and physician-made fenestrations should be studied in a multicenter randomized trial in the future.

In our center, in situ fenestration has not been used to treat TAD or TAA involving the aortic arch, though the technique is popular and has been used in several centers.^28–30 It has been suggested as being more accurate compared to on table fenestrations; however, in situ fenestration requires the use of needles²⁸ and even lasers,^8,30 with a potential risk of aortic injury. Furthermore, the site of in situ fenestration is randomly chosen and the operator cannot necessarily avoid damaging the metal wire of the main stent-graft. In addition, balloon-expandable stents must be used in all the cases to dilate the fenestration hole. Finally, there might be more risk of cerebral ischemia with multiple in situ fenestrations.³¹ However, a study would be needed to compare the advantages and disadvantages of the two fenestration techniques.

Although f-TEVAR has some advantages compared with ch-TEVAR based on our data, there are still some shortcomings. Technical difficulties could occur in which the branch cannot be targeted, and the f-TEVAR technique would have to be changed into a chimney technique. This occurred in 4 f-TEVAR patients in our study who were converted to a chimney procedure owing to an inaccurately positioned fenestration.

Furthermore, the duration of a f-TEVAR procedure is significantly longer than for ch-TEVAR. Additionally, f-TEVAR demands a higher quality of preoperative assessment and intraoperative planning from the operator.^22,32 Therefore, experience in TEVAR procedures is necessary to perform f-TEVAR.

Limitations

We have to emphasize that our study is a retrospective comparative study, which cannot rule out selection bias on the part of the vascular surgeon. A true comparison can be performed only in a prospective randomized study. Also, the results of this study are particular to an Asian population and so might not be generalizable to other races.

Conclusion

For TAD and TAA involving aortic arch, off-the-shelf techniques such as back table fenestrations and chimney procedures provide acceptable midterm outcomes. The f-TEVAR group had a lower reintervention rate with higher branch patency than the ch-TEVAR group; however, f-TEVAR required a longer operation. Our study warrants further research in these off-the-shelf techniques for supra-aortic vessel revascularization.

Footnotes

Authors’ Note

This study was presented at the 2017 Veith Symposium in collaboration with the Society for Vascular Surgery (November 14–18, 2017; New York, NY, USA).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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

ORCID iDs

Meng-tao Wu

Chao Song

Qing-sheng Lu

Supplemental Material

The videos are available at

References

Tsilimparis

Debus

Von Kodolitsch

, et al. Branched versus fenestrated endografts for endovascular repair of aortic arch lesions. J Vasc Surg. 2016;64:592–599.

Konstantinou

Debus

Vermeulen

CFW

, et al. Cervical debranching in the endovascular era: a single centre experience. Eur J Vasc Endovasc Surg. 2019;58:34–40.

Wang

Shu

, et al. Thoracic endovascular aortic repair with single/double chimney technique for aortic arch pathologies. J Endovasc Ther. 2017;24:383–393.

Liu

Chang

, et al. Mid-term outcomes from a multicenter study: is TEVAR safe for ascending aortic dissection? Int J Cardiol. 2018;265:218–222.

Zhao

Feng

Yan

, et al. Outcomes of the chimney technique for endovascular repair of aortic dissection involving the arch branches. Ann Vasc Surg. 2019;58:238–247.e3

Feng

Zhou

, et al. Endovascular repair by customized branched stent-graft: a promising treatment for chronic aortic dissection involving the arch branches. J Thorac Cardiovasc Surg. 2015;150:1631–1638.e5.

Zhu

Guo

Liu

, et al. The single-centre experience of the supra-arch chimney technique in endovascular repair of type B aortic dissections. Eur J Vasc Endovasc Surg. 2013;45:633–638.

Qin

Zhao

Liu

, et al. In situ diode laser fenestration of aortic arch stent grafts during thoracic endovascular aortic repair of Stanford type A aortic dissection. EuroIntervention. 2019;14:e1854–e1860.

Chen

, et al. Total arch repair for acute type A aortic dissection with 2 modified techniques: open single-branched stent graft placement and reinforcement of the dissected arch vessel stump with stent graft. Circulation. 2011;123:2536–2541.

10.

Feng

Zhao

Bao

, et al. Double-chimney technology for treating secondary type I endoleak after endovascular repair for complicated thoracic aortic dissection. J Vasc Surg. 2011;54:212–215.

11.

Yuan

Feng

Jing

Endovascular repair of a thoracic arch aneurysm with a fenestrated stent-graft. J Endovasc Ther. 2008;15:539–543.

12.

Liu

Zhang

, et al. Impact of oversizing on the risk of retrograde dissection after TEVAR for acute and chronic type B dissection. J Endovasc Ther. 2016;23:620–625.

13.

Chou

Chan

Wang

, et al. How to size the main aortic endograft in a chimney procedure. J Thorac Cardiovasc Surg. 2014;147:1099–1101.

14.

Jing

Zhao

, et al. Endovascular stent graft repair of aortic dissection type B extending to the aortic arch. Eur J Vasc Endovasc Surg. 2011;42:456–463.

15.

Lindblad

Bin Jabr

Holst

, et al. Chimney grafts in aortic stent grafting: hazardous or useful technique? Systematic review of current data. Eur J Vasc Endovasc Surg. 2015;50:722–731.

16.

Parmer

Carpenter

Stavropoulos

, et al. Endoleaks after endovascular repair of thoracic aortic aneurysms. J Vasc Surg. 2006;44:447–452.

17.

Preventza

Wheatley

3rd Ramaiah

, et al. Management of endoleaks associated with endovascular treatment of descending thoracic aortic diseases. J Vasc Surg. 2008;48:69–73.

18.

Xue

Sun

Zheng

, et al. The chimney technique for preserving the left subclavian artery in thoracic endovascular aortic repair. Eur J Cardiothorac Surg. 2015;47:623–629.

19.

Ullery

Tran

Itoga

, et al. Natural history of gutter-related type Ia endoleaks after snorkel/chimney endovascular aneurysm repair. J Vasc Surg. 2017;65:981–990.

20.

de Beaufort

HWL

Cellitti

de Ruiter

QMB

, et al. Midterm outcomes and evolution of gutter area after endovascular aneurysm repair with the chimney graft procedure. J Vasc Surg. 2018;67:104–112.

21.

Azuma

Yokoi

Yamazaki

The next generation of fenestrated endografts: results of a clinical trial to support an expanded indication for aortic arch aneurysm treatment. Eur J Cardiothorac Surg. 2013;44:e156–e163.

22.

Canaud

Ozdemir

Chassin-Trubert

, et al. Homemade fenestrated stent-grafts for complete endovascular repair of aortic arch dissections. J Endovasc Ther. 2019;26:645–651.

23.

Marrocco-Trischitta

Spampinato

Mazzeo

, et al. Impact of the bird-beak configuration on postoperative outcome after thoracic endovascular aortic repair: a meta-analysis. J Endovasc Ther. 2019;26:771–778.

24.

Bosiers

Tran

Lee

, et al. Incidence and prognostic factors related to major adverse cerebrovascular events in patients with complex aortic diseases treated by the chimney technique. J Vasc Surg. 2018;67:1372–1379.

25.

Magee

Veranyan

Kuo

, et al. Anatomic suitability for “off-the-shelf” thoracic single side-branched endograft in patients with type B aortic dissection [published online June 24, 2019]. J Vasc Surg. doi:10.1016/j.jvs.2019.04.461.

26.

Andrási Terézia

Grossmann

Zenker

, et al. Supra-aortic interventions for endovascular exclusion of the entire aortic arch. J Vasc Surg. 2017;66:281–297.e2.

27.

Verscheure

Haulon

Tsilimparis

, et al. Endovascular treatment of post type A chronic aortic arch dissection with a branched endograft: early results from a retrospective international multicenter study [published online April 8, 2019]. Ann Surg. doi:10.1097/SLA.0000000000003310.

28.

Wang

Zhou

Guo

, et al. A new adjustable puncture device for in situ fenestration during thoracic endovascular aortic repair. J Endovasc Ther. 2018;25:474–479.

29.

Tse

Lindsay

Roche-Nagle

, et al. Radiofrequency in situ fenestration for aortic arch vessels during thoracic endovascular repair. J Endovasc Ther. 2015;22:116–121.

30.

Talaie

Werter

Drucker

, et al. Laser fenestration for treatment of a complicated chronic type B aortic dissection. Vasc Endovascular Surg. 2018;52:212–217.

31.

Wang

Zhang

Liu

, et al. A novel internal and external bypass method to safeguard cerebral blood flow during aortic arch endovascular repair requiring triple in situ fenestrations. J Endovasc Ther. 2019;26:652–657.

32.

Lucatelli

Cini

Benvenuti

, et al. Custom-made endograft for endovascular repair of thoraco-abdominal aneurysm and type B dissection: single-centre experience. Cardiovasc Intervent Radiol. 2018;41:1174–1183.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB