Outcomes of directional branches using self-expandable or balloon-expandable stent grafts during endovascular repair of thoracoabdominal aortic aneurysms

From the Society for Clinical Vascular Surgery

Outcomes of directional branches using self-expandable or balloon-expandable stent grafts during endovascular repair of thoracoabdominal aortic aneurysms Emanuel R. Tenorio, MD, PhD, Jussi M. Kärkkäinen, MD, PhD, Bernardo C. Mendes, MD, Randall R. DeMartino, MD, Thanila A. Macedo, MD, Alisa Diderrich, RN, Jan Hofer, RN, and Gustavo S. Oderich, MD, Rochester, Minn

ABSTRACT Objective: The objective of this study was to evaluate outcomes of directional branches using self-expandable stent grafts (SESGs) or balloon-expandable stent grafts (BESGs) during fenestrated-branched endovascular aneurysm repair of thoracoabdominal aortic aneurysms. Methods: Patients treated by fenestrated-branched endovascular aneurysm repair were enrolled in a prospective study from 2014 to 2018. We included in the analysis patients who had target vessels incorporated by directional branches using either SESG (Fluency [Bard, Covington Ga] or Gore Viabahn [W. L. Gore & Associates, Flagstaff, Ariz]) or BESG (Gore VBX). Target artery instability (TAI) was defined by a composite of any stent stenosis, separation, or type IC or type IIIC endoleak requiring reintervention and stent occlusion, aneurysm rupture, or death due to target artery complication. End points included technical success, target artery patency, freedom from TAI, freedom from type IC or type IIIC endoleak, and freedom from target artery reintervention. Results: There were 126 patients (61% male; mean age, 73 6 8 years) included in the study. A total of 335 renal-mesenteric arteries were targeted by directional branches using SESGs in 62 patients and 176 arteries or BESGs in 54 patients and 159 arteries. Patients in both groups had similar thoracoabdominal aortic aneurysm classification and aneurysm and target artery diameter, but SESG patients had significantly (P < .05) shorter stent length (
7 mm) and larger stent diameter (þ1 mm) and more often had adjunctive bare-metal stents (72% vs 15%). Technical success was achieved in 99% of patients, with one 30-day death (0.7%). Mean follow-up was significantly longer among patients treated by SESGs compared with BESGs (23 6 12 months vs 868 months; P < .0001). TAI occurred in 27 directional branches (8%), including 11 type IC endoleaks (2 SESGs, 9 BESGs), 10 stenoses (3 SESGs, 7 BESGs), 4 occlusions (3 SESGs, 1 BESGs), 4 type IIIC endoleaks (2 SESGs, 2 BESGs), and 1 stent separation (SESG), resulting in 20 target artery reinterventions in 16 patients (5 SESGs and 11 BESGs). At 1 year, SESGs had higher primary patency (97% 6 2% vs 96% 6 2%; P ¼ .004), freedom from TAI (96% 6 2% vs 88% 6 3%; P < .0001), freedom from type IC or type IIIC endoleaks (98% 6 1% vs 92% 6 3%; P ¼ .0004), and freedom from target artery reinterventions (98% 6 1% vs 88% 6 4%; P < .0001) compared with BESGs. There was no difference in secondary patency for SESGs and BESGs (98% 6 1% vs 99% 6 1%; P ¼ .75). Factors associated with TAI were large stent diameter (odds ratio, 0.6; P < .0001) and use of VBX stent graft (odds ratio, 6.5; P < .0001). Conclusions: Directional branches were associated with high technical success and low rates of stent occlusion, independent of stent type. However, primary patency, freedom from TAI, and freedom from type IC or type IIIC endoleaks was lower for BESGs compared with SESGs. (J Vasc Surg 2019;-:1-14.) Keywords: Bridging stent; Self-expandable stent graft; Balloon-expandable stent graft; Fenestrated-branched endovascular aneurysm repair

Fenestrated-branched endovascular aneurysm repair (F-BEVAR) has allowed successful treatment of patients with pararenal aneurysms (PRAs) and thoracoabdominal aortic aneurysms (TAAAs). Different reports have shown the safety, feasibility, and effectiveness of these procedures, but the materials employed are not standardized,

and most decisions are left to the discretion of the treating physician.1-8 Controversy remains with respect to selection of fenestration or directional branches and which type of bridging stent should be used. Furthermore, few or none of the bridging stents have been tested or are approved for use with F-BEVAR

From the Advanced Endovascular Aortic Research Program, Division of

Correspondence: Gustavo S. Oderich, MD, Gonda Vascular Center, Mayo Clinic,

Vascular and Endovascular Surgery, Mayo Clinic. ClinicalTrials.gov identifier: NCT1937949 and NCT2089607.

200 First Street SW, Rochester, MN 55905 (e-mail: oderich.gustavo@mayo. edu).

Author conflict of interest: G.S.O. has received consulting fees and grants from

The editors and reviewers of this article have no relevant financial relationships to

Cook Medical, W. L. Gore, and GE Healthcare (all paid to Mayo Clinic with no

disclose per the JVS policy that requires reviewers to decline review of any

personal income). Presented in the Plenary Session of the Forty-seventh Annual Symposium of the Society for Clinical Vascular Surgery, Boca Raton, Fla, March 16-20, 2019. Additional material for this article may be found online at www.jvascsurg.org.

manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2019 by the Society for Vascular Surgery. Published by Elsevier Inc. https://doi.org/10.1016/j.jvs.2019.07.079

1

2

Journal of Vascular Surgery

Tenorio et al

---

procedures.9-12 Therefore, bridging stents are used off label, and different stent combinations are often needed to overcome anatomic challenges and to achieve desired stent characteristics, such as flexibility, radial force, optimal length, diameter, and profile.13 Balloon-expandable covered stents or stent grafts (BESGs) have been used traditionally for alignment of fenestrations, whereas self-expandable stent grafts (SESGs) have been preferred for directional branches. BESGs are available in shorter length, have higher radial force, and have been preferentially used for transversely oriented vessels aligned by fenestrations, whereas SESGs can better accommodate tortuous and longer lengths that are needed with directional branches. Novel BESGs have challenged this concept by adding features of increased flexibility, longer lengths, and smaller profile to their known advantages of radial force, precise deployment, and ability to flare. The balloon-expandable Viabahn VBX stent graft (W. L. Gore & Associates, Flagstaff, Ariz) was introduced in 2016 and used in conjunction with the Gore Thoracoabdominal Multibranched Endoprosthesis (TAMBE) in patients with PRAs and extent IV TAAAs.13,14 Since its introduction, the stent gained in popularity, but clinical data are lacking on its performance for more extensive TAAAs. The purpose of this study was to evaluate the outcomes between SESGs and BESGs for directional branches during F-BEVAR.

METHODS The study was approved by the Institutional Review Board of the Mayo Clinic. All patients consented for participation in the research study and were enrolled in a prospective physician-sponsored investigational device exemption protocol to evaluate F-BEVAR using a manufactured patient-specific or off-the-shelf Cook Zenith device (Cook Medical, Bloomington, Ind) between April 2014 and October 2018 (G130030 and G130266). Patients who had devices with at least one directional branch were included in the analysis. Patients with devices based on fenestrations without directional branches were excluded from the analysis. Demographics, clinical characteristics, cardiovascular risk factors, and operative and postoperative variables were collected prospectively and stored on case report forms and the Medidata Rave (New York, NY) database. Device design. Selection of stent design has been summarized elsewhere.1 Aneurysm morphology was determined by high-resolution computed tomography angiography (CTA), and devices were designed with a minimum proximal landing zone of 25 mm. Selection of directional branches was preferred in segments with larger inner aortic diameter (>40 mm) coupled with vessels with downgoing orientation and without excessive tortuosity. All devices were designed and approved by the primary investigator (G.S.O).

2019

ARTICLE HIGHLIGHTS d

d

d

Type of Research: Prospective, single-center cohort study Key Findings: In 126 patients, 335 renal-mesenteric arteries were targeted during fenestrated-branched endovascular aneurysm repair by directional branches using self-expandable stent grafts in 176 arteries and balloon-expandable stent grafts (BESGs) in 159 arteries. Directional branches were associated with high technical success (99%) and low rates of stent occlusion (1%), independent of stent type. However, at 1 year, primary patency, freedom from target artery instability, and freedom from type IC and type IIIC endoleaks were lower for BESGs compared with self-expandable stent grafts. Take Home Message: This study represents an analysis of specific changes in the choice of bridging stent for directional branches and shows some of the shortcomings of BESG on early follow-up.

Selection of alignment or bridging stents. Before February 2017, directional branches were targeted using SESGs, which included Fluency stent graft (Bard, Covington, Ga) for the celiac axis and superior mesenteric arteries (SMAs) or Viabahn stent grafts (W. L. Gore) for the renal arteries (Supplementary Fig 1, online only). In patients with small renal arteries (4-5 mm) treated by 5to 6-mm Viabahn SESGs, a larger 7-mm Viabahn SESG was used for sealing at the 6-mm directional branch. The cuff segment was often reinforced with a short balloonexpandable stent to prevent separation. An adjunctive bare-metal self-expandable stent was often used to prevent kink. Since the introduction of the Viabahn VBX BESG in February 2017, this became the bridging stent graft of choice for all directional branches. Surveillance protocol. Follow-up consisted of clinical examination, laboratory studies, and imaging before discharge and at 2, 6, and 12 months and annually thereafter for the first 5 years. Imaging evaluation included CTA or computed tomography without contrast enhancement (in patients with stage 3B or stage 4 chronic kidney disease) and duplex ultrasound of the renal-mesenteric arteries. All imaging studies were independently evaluated by dedicated vascular radiologists and by the primary investigator. Clinical events and imaging findings were adjudicated by a Clinical Events Committee and Data and Safety Monitoring Board. Statistical analysis, outcomes measures, and definitions. Data were managed and stored in the Medidata Rave database and analyzed using IBM SPSS statistics 25 (IBM, Armonk, NY). Outcomes were compared in patients with directional branches targeted by SESG or

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

3

-

BESG using analysis of patients or branches at risk, when pertinent. For analysis of target artery outcomes, we considered only the VBX stent graft as BESG and excluded five patients (eight vessels) treated by iCAST covered stents (Atrium Maquet, Hudson, NH). For analysis of renal directional branches, only Viabahn and VBX were considered SESG or BESG, respectively. The reporting standards of the Society for Vascular Surgery were used to define stent graft-related outcomes and major adverse events (MAEs).15,16 Type IC and type IIIC endoleaks were defined as endoleak originating from the distal target vessel sealing zone or from bridging stent disruption, separation, or insufficient overlap, respectively. End points included technical success, target artery patency, freedom from target artery instability (TAI), freedom from type IC or type IIIC endoleak, and freedom from reintervention. TAI was defined as a composite end point of any branch-related complication leading to aneurysm rupture, death, or occlusion or a stent separation, stenosis, kink, or endoleak requiring reintervention.7 Primary patency and secondary patency were defined by uninterrupted patency from the index procedure until occlusion or branch stent reintervention for stenosis and by an occlusion treated by surgical bypass or not suitable to endovascular salvage, respectively. Time-dependent outcomes were reported using KaplanMeier estimates, and differences were determined by the log-rank test. Results were reported as percentage for categorical variables and mean 6 standard deviation for continuous variables. The Pearson c2 or Fisher exact test was used for analysis of categorical variables. Differences between means were tested with two-sided Student ttest, Wilcoxon rank sum test, or Mann-Whitney test. Multivariate Cox regression model was used to evaluate independent predictors for TAI after selection of important independent variables or those with positive association by univariate analysis. A P value of <.05 was used to determine statistical significance.

RESULTS Study patients A total of 300 patients treated by F-BEVAR were enrolled in the physician-sponsored investigational device exemption study. From this group, 126 patients (42%), 77 male (61%) and 49 female, with a mean age of 73 6 8 years were included in the analysis (Table I). Directional branches were stented using SESGs in 62 patients (49%) and 176 vessels or BESGs in 54 patients (43%) and 159 vessels. Ten patients (8%) had vessels targeted by combinations of both types of stents. Patients treated by SESGs had significantly higher rates (P < .05) of cigarette smoking and pulmonary, cardiac, and total Society for Vascular Surgery comorbidity scores. Patients in both groups had similar TAAA classification, aneurysm

diameter, target artery diameter, inner aortic diameter for each vessel, and renal angulation. TAAA classification was extent I to III in 112 patients (89%) in both groups. Bridging stent type SESGs were predominantly used in the first three-fifths of the study period (Fig 1). Use of SESGs declined from 100% (n ¼ 152/152) between 2014 and 2016 to 13% (n ¼ 24/183) between 2017 and 2018 (P < .0001). For mesenteric vessels, choice of stent changed from Fluency stent graft in 71 of 101 vessels (70%) between 2014 and 2016 to Viabahn VBX BESG in 102 of 112 vessels (91%) between 2017 and 2018. For renal artery incorporation, choice of stent changed from Viabahn stent graft in 51 of 51 vessels (100%) to Viabahn VBX BESG in 57 of 71 vessels (80%), respectively. Eight vessels (six mesenteric arteries and two renal arteries) in five patients were targeted by iCAST covered stents, all before 2017. Stent graft design A patient-specific device was designed in 93 patients (74%), and an off-the-shelf multibranched stent graft (t-Branch; Cook Medical) was selected in 33 patients (26%). Among patients treated by patient-specific devices, there were 13 different types of stent configuration. The most common design, which was selected in 34 of the 93 patients (37%), consisted of two directional branches for the celiac axis and SMA and two renal fenestrations (Supplementary Table I, online only). There was no difference in rates of TAI between patients with offthe-shelf t-Branch and patient-specific device (8% vs 8%; P ¼ .78). Furthermore, none of 13 different types of stent configuration was associated with increase of the risk of TAI. A total of 335 renal-mesenteric arteries (105 celiac arteries, 108 SMAs, and 122 renal arteries) were incorporated by directional branches using BESGs in 159 vessels (47%) or SESGs in 176 vessels (53%). The average number of directional branches per patient was 2.5 6 1.1, with no difference between groups (P > .05). Procedure details All procedures were performed in dedicated hybrid operating rooms equipped with a floor-mounted and laser-guided mobile gantry imaging system using general endotracheal anesthesia (Table II). Contrast material volume and total endovascular time were higher (P < .05) in patients treated by SESGs. Patients treated by BESGs had significantly (P < .05) longer stent length (þ7; 70 6 11 mm vs 63 6 13 mm), smaller stent diameter (
0.4; 7.4 6 1.2 mm vs 7.8 6 1.4 mm), and less oversizing (
0.4; 0.6 6 1.1 mm vs 1.0 6 0.9 mm) and less often had adjunctive bare-metal stents (15% vs 72%). Technical success, defined by deployment of the aortic stent graft and all intended side branch components, was achieved in 97% (122/126) of patients and 99% (335/ 339) of the vessels. Among patients with technical

4

Journal of Vascular Surgery

Tenorio et al

---

2019

Table I. Demographics and clinical and anatomic characteristics of 126 patients treated by fenestrated and branched endovascular aneurysm repair (F-BEVAR) using balloon-expandable stent graft (BESG) or self-expandable stent graft (SESG) for directional branches Overall (N ¼ 126)

BESGa (n ¼ 54)

SESGa (n ¼ 62)

SESG and BESG (n ¼ 10)

Age, years

73 6 8

74 6 8

74 6 6

68 6 13

.75

Age >80 years

29 (23)

15 (28)

11 (18)

3 (30)

.20

Male sex

77 (61)

37 (69)

37 (60)

3 (30)

.32

104 (83)

39 (72)

59 (95)

6 (60)

.001

Hypertension

118 (94)

50 (93)

59 (95)

9 (90)

.56

Hypercholesterolemia

102 (81)

43 (80)

52 (84)

7 (70)

.55

Coronary artery disease

63 (50)

23 (43)

36 (58)

4 (40)

.10

COPD

47 (37)

18 (33)

28 (45)

1 (10)

.19

Peripheral artery disease

24 (19)

6 (11)

18 (29)

0

.02

Atrial fibrillation

30 (24)

13 (24)

15 (24)

2 (20)

.98

1

7 (6)

1 (2)

4 (7)

2 (20)

2

59 (47)

23 (43)

33 (53)

3 (30)

3

55 (44)

29 (54)

22 (35)

4 (40)

4

3 (2)

1 (2)

2 (3)

0

5

2 (2)

0

1 (2)

1 (10)

CKD stage 3-5

60 (48)

30 (56)

25 (40)

5 (50)

.10

Diabetes mellitus

P valuea

Demographics

Cardiovascular risk factors Cigarette smoking

CKD stage

.25

20 (16)

9 (17)

11 (18)

0

.88

Congestive heart failure

13 (10)

4 (7)

7 (11)

2 (20)

.48

Stroke/TIA

18 (14)

8 (15)

9 (15)

1 (10)

.96

Cancer

29 (15)

6 (11)

13 (21)

0

.15

Prior aortic repair

72 (57)

27 (50)

38 (61)

7 (70)

.22 .75

Preoperative evaluation Positive cardiac stress test

18 (14)

9 (17)

9 (15)

0

Ejection fraction, %

59 6 10

59 6 10

59 6 10

63 6 4

.75

Serum creatinine, mg/dL

1.3 6 0.8

1.2 6 0.4

1.3 6 0.9

1.5 6 1.4

.21

eGFR, mL/min/1.73 m2

59 6 21

57 6 15

60 6 19

65 6 45

.32

Body mass index, kg/m2

28 6 6

28 6 5

27 6 5

34 6 12

.19

10 6 5

865

12 6 4

663

<.0001

Cardiac score

3.4 6 3.2

2.2 6 3.0

4.9 6 2.7

0.6 6 1.1

<.0001

Pulmonary score

1.8 6 1.9

1.3 6 1.8

2.5 6 1.7

0.6 6 1.0

<.0001

Renal score

0.7 6 1.3

0.7 6 1.2

0.7 6 1.4

1.1 6 2.0

.50

Hypertension score

2.0 6 0.9

1.9 6 0.9

2.0 6 0.9

2.4 6 0.8

.89

Age score

1.9 6 0.8

1.9 6 0.9

1.9 6 0.6

1.3 6 1.1

.87

1

6 (5)

1 (2)

5 (8)

0

2

73 (58)

33 (61)

37 (60)

3 (30)

3

44 (35)

19 (35)

18 (29)

7 (70)

4

3 (2)

1 (2)

2 (3)

0

5

0

0

0

0

Crawford extent I

11 (9)

5 (9)

5 (8)

1 (10)

Crawford extent II

70 (56)

29 (54)

33 (53)

8 (80)

Risk assessment SVS total score (0-30)

ASA class

.43

Anatomic characteristics Type of aneurysm

.62

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

5

-

Table I. Continued. Overall (N ¼ 126)

BESGa (n ¼ 54)

SESGa (n ¼ 62)

SESG and BESG (n ¼ 10)

P valuea

Crawford extent III

31 (25)

15 (28)

15 (24)

1 (10)

Crawford extent IV

13 (10)

4 (8)

9 (15)

0

1 (1)

1 (2)

0

0

Postdissection thoracoabdominal aneurysm

20 (16)

13 (24)

6 (10)

1 (10)

.04

Maximum aortic diameter, mm

70 6 12

67 6 10

71 6 13

72 6 14

.11

Celiac axis

8.2 6 1.7

8.2 6 1.6

8.2 6 1.7

e

.79

Superior mesenteric artery

7.9 6 1.3

8.1 6 1.4

7.8 6 1.2

e

.08

Pararenal

Vessel diameter,b mm

Right renal artery

5.5 6 1.0

5.7 6 1.1

5.4 6 0.8

e

.44

Left renal artery

5.8 6 1.0

5.8 6 1.1

5.6 6 0.9

e

.59

Inner aortic diameter for each vessel,b mm Celiac axis

35 6 10

38 6 13

34 6 7

e

.43

Superior mesenteric artery

34 6 8

35 6 8

32 6 8

e

.14

Right renal artery

34 6 10

35 6 10

33 6 9

e

.95

Left renal artery Renal angulation, degrees

34 6 9

35 6 10

34 6 9

e

.73


24 6 20


24 6 17


24 6 21

e

.83

ASA, American Society of Anesthesiologists; COPD, chronic obstructive pulmonary disease; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; SVS, Society for Vascular Surgery; TIA, transient ischemic attack. Categorical variables are presented as number of patients (%). Continuous variables are presented as mean 6 standard deviation. a Difference statistically significant between BESG and SESG. b Analysis per vessel.

imaging during the study. Intraoperative target vessel revisions based on cone beam computed tomography findings were needed in 6% (9/159) of BESG and 1% (2/ 176) of SESG target vessels (P ¼ .03). None of the technical failures were attributed to stent choice.

Fig 1. Evolution of the use of a self-expandable stent graft (SESG) over time in relation to the use of a balloonexpandable stent graft (BESG).

failures, three had inability to catheterize renal arteries (three left renal, one right renal) and one had renal artery disruption requiring intentional embolization leading to loss of the kidney. The three patients with technical difficulties during renal stenting were successfully treated by retrograde stenting through laparotomy in two patients or ileorenal bypass in one patient. The use of cone beam computed tomography was more frequent in patients treated by BESGs (100%) compared with SEGS (80%; P ¼ .001), which reflects a change in advanced

Clinical outcomes One patient (0.7%) treated for extent II TAAA using BESG died of presumed stroke on postoperative day 23 after an uncomplicated hospital course (Supplementary Table II, online only). The patient was discharged home on postoperative day 10 with predismissal CTA showing widely patent branches and no complications. Forty-one patients (32%) had MAEs, with similar rates in both groups. The most common MAEs were blood loss >1 L in 15 patients (12%), acute kidney injury in 15 patients (12%), paraplegia in 7 patients (6%), respiratory failure requiring reintubation in 4 patients (3%), myocardial infarction in 3 patients (2%,) and ischemic colitis in 1 patient (1%). Four patients (3%) required early secondary interventions (three BESGs, one SESG) to treat three type IC endoleaks (two BESGs, one SESG) or stent kink (one BESG). Mean follow-up averaged 8 6 8 months for patients treated by BESGs and 23 6 12 months for those who had SESGs (P < .0001). Follow-up imaging was obtained in all patients with a mean of 2.4 studies per patient. There was no aorta- or branch-related death or aneurysm rupture during the study.

6

Journal of Vascular Surgery

Tenorio et al

---

2019

Table II. Procedural details and device design of 126 patients treated by fenestrated and branched endovascular aneurysm repair (F-BEVAR) using balloon-expandable stent graft (BESG) or self-expandable stent graft (SESG) for directional branches Overall (N ¼ 126)

BESGa (n ¼ 54)

SESGa (n ¼ 62)

SESG and BESG (n ¼ 10)

P valuea

General anesthesia

126 (100)

54 (100)

62 (100)

10 (100)

Cerebrospinal fluid drainage

116 (92)

50 (93)

59 (95)

7 (70)

.56

NA

Fusion/cone beam computed tomography

113 (90)

54 (100)

50 (81)

9 (90)

.001

Somatosensory evoked potential/motor evoked potentials

123 (98)

53 (98)

60 (97)

10 (100)

.64

Brachial access

10 (100)

.28

125 (99)

53 (98)

62 (100)

Left side

87 (69)

30 (56)

50 (81)

7 (70)

Right side

38 (30)

23 (43)

12 (19)

3 (30)

.007

106 (84)

53 (98)

43 (69)

10 (100)

<.0001

Percutaneous femoral approach

.004

Unilateral

21 (17)

6 (11)

14 (23)

1 (10)

.10

Bilateral

85 (67)

47 (87)

29 (47)

9 (90)

<.0001

31 (25)

7 (13)

23 (37)

1 (10)

.003

5 (4)

1 (2)

3 (5)

1 (10)

.38

Iliofemoral conduit Iliac conduit

26 (21)

6 (11)

20 (32)

0

Amount of contrast material used, mL

Femoral conduit

167 6 61

155 6 56

183 6 63

122 6 51

.01

Total endovascular time, minutes

178 6 63

164 6 57

1906 66

189 6 64

.03

Total operation time, minutes

264 6 77

245 6 69

282 6 82

253 6 63

.006

Total fluoroscopy time, minutes

.006

90 6 34

89 6 32

90 6 37

96 6 31

.94

Total air kerma, mGy

2153 6 1815

1178 6 679

3131 6 2056

1255 6 1003

<.0001

Estimated blood loss, mL

522 6 545

426 6 408

622 6 653

400 6 241

.13

50 (40)

23 (43)

23 (37)

4 (40)

.55

Patient specific

93 (74)

42 (78)

48 (77)

3 (30)

Off-the-shelf t-Branch

33 (26)

12 (22)

14 (23)

7 (70)

9 (7)

5 (9)

4 (6)

0

Prior intentional stage repair Device design Device type

.96

Access scallop

.57

Low-profile device

65 (52)

37 (69)

25 (40)

3 (30)

.002

Preloaded catheters or wires

74 (59)

39 (72)

32 (52)

3 (30)

.02

Total vessels incorporated by directional branchesb Branches per patients

335

159

176

e

2.5 6 1.1

2.6 6 1.1

2.4 6 1.2

e

.40 <.0001

Stent length, mm

66 6 13

70 6 11

63 6 13

e

Stent diameter, mm

7.2 6 1.8

7.4 6 1.2

7.8 6 1.4

e

.001

Oversizing, mm

0.8 6 1.8

0.6 6 1.1

1.0 6 0.9

e

<.0001

Reline with self-expandable bare-metal stent

150 (45)

24 (15)

126 (72)

e

<.0001

35 (10)

29 (18)

6 (3)

e

<.0001

55

e

c

No. of the main stent >1 Celiac axisb

105

50

64 6 12

67 6 13

62 6 12

e

.35

Stent diameter, mm

8.4 6 0.9

8.1 6 0.7

8.7 6 0.9

e

<.0001

Oversizing,d mm

0.8 6 1.2

0.8 6 1.6

0.9 6 0.7

e

.03

57 (54)

15 (30)

42 (76)

e

<.0001

Stent length,c mm

Reline with self-expandable bare-metal stent Superior mesenteric arteryb

108

52

56

e

70 6 12

74 6 9

66 6 13

e

.09

Stent diameter, mm

8.3 6 0.8

8.0 6 0.8

8.6 6 0.7

e

<.0001

Oversizing,d mm

0.8 6 0.8

0.6 6 0.9

1.0 6 0.7

e

.0002

45 (42)

3 (7)

42 (75)

e

<.0001

Stent length,c mm

Reline with self-expandable bare-metal stent Right renal arteryb Stent length,c mm Stent diameter, mm

65

30

35

e

63 6 13

67 6 10

60 6 15

e

<.0001

6.2 6 0.8

6.2 6 0.8

6.2 6 0.7

e

.82

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

7

-

Table II. Continued. Overall (N ¼ 126)

BESGa (n ¼ 54)

SESGa (n ¼ 62)

SESG and BESG (n ¼ 10)

Oversizing,d mm

1.0 6 1.2

0.7 6 0.8

1.3 6 1.3

e

.01

Reline with self-expandable bare-metal stent

25 (38)

2 (7)

23 (66)

e

<.0001

Left renal arteryb

P valuea

57

27

30

e

65 6 13

71 6 10

61 6 14

e

Stent diameter, mm

6.2 6 0.8

6.2 6 .0.8

6.2 6 0.9

e

.96

Oversizing,d mm

0.7 6 0.9

0.5 6 0.6

0.9 6 1.0

e

.04

Stent length,c mm

Reline with self-expandable bare-metal stent Hospital stay Technical success

.0001

23 (39)

4 (13)

19 (66)

e

<.0001

9 6 11

864

11 6 14

11 6 8

.37

99% (335/339)

e

e

e

NA

NA, Not applicable. Categorical variables are presented as number of patients (%). Continuous variables are presented as mean 6 standard deviation. a Difference statistically significant between BESG and SESG. b Analysis per vessel. c Total length between the sealing zone of the target vessel and beginning of the cuff on centerline measurements. d Difference between the native diameter of the target vessel and nominal diameter of the stent used.

Fig 2. Kaplan-Meier estimates of primary patency in 355 directional branches targeted using self-expandable stent graft (SESG) or balloon-expandable stent graft (BESG).

Target artery outcomes Stent stenosis, kink, and occlusion. Twelve patients (10%) had 10 stenoses and 4 occlusions of directional branches. Of these, seven stenoses and one occlusion occurred in vessels targeted by BESGs (three mesenteric arteries, five renal arteries). Primary and secondary target artery patency for all directional branches at 1 year was 96% 6 1% and 98% 6 1%, respectively. Primary patency was lower for BESGs compared with SESGs (96%62% vs 97% 6 1%; P ¼ .004; Fig 2). For renal artery

directional branches, primary patency at 1 year was 95% 6 3% for BESGs and 94% 6 3% for SEGSs (P ¼ .09; Supplementary Fig 2, A, online only). For mesenteric arteries, primary patency was 96%62% and 100% at the same interval, respectively (P ¼ .04; Supplementary Fig 2, B, online only). Secondary patency at 1 year was similar for directional branches targeted by BESGs or SESGs (99% 6 1% vs 98% 6 1%; P ¼ .75), including renal (100% vs 94% 6 3%; P ¼ .22) and mesenteric arteries (99% 6 2% vs 100%; P ¼.19), respectively. Independent predictors of loss

Journal of Vascular Surgery

Tenorio et al

8

---

2019

Table III. Predictors by univariate and multivariate Cox regression analysis for target artery instability (TAI), target arteryrelated stenosis or occlusion, and type IC or type IIIC endoleak in 335 vessels incorporated by directional branches using balloon-expandable stent graft (BESG) or self-expandable stent graft (SESG) Confidence interval All vessels

Hazard ratio

Lower

Upper

P value

1.3

1.02

1.7

.037

TAI Univariate Baseline creatinine Celiac axis incorporation Renal artery incorporation Vessel diameter (large, þ1 mm) BESG: VBX SESG: Fluency

.163 4.4 .687 7.7 .159

.039 1.9 .526 2.9 .037

.687 10.1 .897 20.4

.014 <.0001 .006 <.0001

.677

.013

Stent diameter (large, þ1 mm)

.568

.429

.752

<.0001

Adjunctive self-expandable bare-metal stent

.215

.087

.528

<.0001

Multivariate Baseline creatinine Stent diameter (large, þ1 mm) BESG: VBX

1.5 .594

1.1 .446

1.9 .790

.005 <.0001

6.5

2.5

17.1

<.0001

7.0

1.9

25.1

.003

Target artery-related stenosis or occlusion Univariate Renal artery incorporation Vessel diameter (large, þ1 mm) BESG: VBX

.513 5.6

.335 1.6

.785 20.0

.002 .007

Stent diameter (large, þ1 mm)

.552

.375

.814

.003

Adjunctive self-expandable bare-metal stent

.320

.103

.997

.049

Multivariate Renal artery incorporation Vessel diameter (large, þ1 mm) BESG: VBX

6.4 .535 4.5

1.8 .353 1.3

23.1 .811

.004 .003

15.0

.016

Target artery-related type IC or type IIIC endoleak Univariate Renal artery incorporation BESG: VBX

3.6

1.2

10.5

.020

10.7

2.3

49.6

.002

Stent diameter (large, þ1 mm)

.565

.385

.829

.003

Oversizing (mm, þ1 mm)

.337

.125

.905

.031

Adjunctive self-expandable bare-metal stent

.160

.042

.611

.007

Multivariate Stent diameter (large, þ1 mm) BESG: VBX

of primary patency were renal artery target, small vessel diameter, and use of Viabahn VBX BESGs (Table III). TAI. Patients treated by BESGs had significantly (P > .05) higher rates of TAI, mostly due to higher rates of type IC endoleaks, which occurred in 8 of 54 patients (15%) or 9 of 159 vessels (6%) with BESGs and in 1 of 62 patients (2%) or 2 of 176 vessels (1%) with SESGs (Fig 3; Table IV). Directional branches targeted by BESGs had significantly lower 1-year freedom from TAI (88% 6 3% vs 96% 6 2%; P < .0001; Fig 4), freedom from type IC or type IIIC endoleak (92% 6 2% vs 98% 6 1%; P ¼ .0004; Fig 5),

.595 8.0

.400 1.9

.886 33.7

.011 .005

and freedom from target artery reintervention (88% 6 4% vs 98% 6 1%; P < .0001; Fig 6) compared with those targeted by SESGs. For mesenteric arteries, there was no difference in freedom from TAI (94% 6 3% vs 98% 6 1%; P ¼ .12), freedom from type IC or IIIC endoleak (98% 6 1% vs 98% 6 1%; P ¼ .83), and freedom from reintervention (96% 6 2% vs 98 6 1%; P ¼ .45) for BESGs and SESGs at the same interval, respectively. However, renal artery directional branches targeted by BESGs had lower freedom from TAI (78% 6 7% vs 90% 6 4%; P < .0001; Supplementary Fig 3, online only), freedom from type IC

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

9

-

Fig 3. Illustrations and computed tomography angiography (CTA) images showing type IC endoleak in patient with VBX balloon-expandable stent graft (BESG). A, The 30-day CTA image without endoleak. B, Displacement of the left renal stent leading to a type IC endoleak after 1-year follow-up. C, Reintervention with redo with another stent. (Reproduced by permission of Mayo Foundation for Medical Education and Research. All rights reserved.)

or IIIC endoleaks (83% 6 6% vs 98% 6 2%; P < .0001; Supplementary Fig 4, online only), and freedom from reinterventions (76% 6 8% vs 88% 6 5%; P < .0001; Supplementary Fig 5, online only) compared with SESGs. Independent predictors of TAI were large stent diameter and use of Viabahn VBX BESG. Downgoing renal artery configuration (negative angulation) was associated with lower risk of TAI by multivariate Cox regression (Supplementary Table III, online only).

DISCUSSION This is the first report on outcomes of Viabahn VBX BESG as a bridging stent for extent I to III TAAAs. Potential advantages of the Viabahn VBX BESG, based on benchside testing and the early feasibility TAMBE study, included flexibility, radial force, choice of stent length and diameter, ability to flare a stent up to 16 mm using an 8F delivery system, and decreased number of steps for construction of each branch, avoiding balloon exchanges that are often needed with SESGs. Based on these characteristics, we changed our practice in 2017 to use predominantly Viabahn VBX BESG for all directional branches. However, our initial enthusiasm was tempered by the unexpected findings of this study, including higher TAI, type IC and type IIIC endoleaks, and stent stenosis or kinks in a shorter follow-up compared with SESGs. Whereas the senior author has already overcome the F-BEVAR learning curve, it is possible that some of the sobering outcomes are due to inadequate selection or deployment of BESGs. Attention to patient selection and to choice of stent length, diameter, and oversizing and use of adjunctive self-expandable stents could have prevented or minimized some of the complications.

The concept of TAI was first coined by Tara Mastracci in 2013 to describe a composite of all relevant target artery outcomes. Analysis of TAI is particularly relevant in comparing fenestrations vs directional branches or in analyzing outcomes of different bridging stent choices, particularly for renal branches, which have been plagued by higher occlusion rates.7 Mastracci et al17 reported the outcomes of 940 vessels targeted by directional branches. In that study, renal branches had higher failure rates compared with mesenteric branches independent of stent choice, which was Fluency, Viabahn, or iCAST. A limitation of that report was bias toward selection of only fenestrations or only branches, with few opting for a balanced approach toward choice of incorporation. In contrast, this study included devices with combinations of fenestrations and directional branches. However, none of 13 types of stent configuration was associated with increase of the risk of TAI. Our study has shown improved freedom from TAI with use of SESG (96% at 1 year), rivaling the 94% rate reported by Mastracci et al for fenestrated stent grafts. Data from the United States F-BEVAR Research Consortium on 2428 targeted vessels also showed similar results for primary patency of fenestrations or branches at 2 years (97% vs 97%). Instead, it was not the type of incorporation but aneurysm extent and renal target that affected the outcome, independent of whether a fenestration or directional branch was selected.18 The ideal bridging stent should provide durable aneurysm exclusion while maintaining end-organ perfusion. Despite a significantly longer follow-up for SESGs (with more time for stent failures to occur), the Viabahn VBX BESG was associated with lower primary patency, more

10

Journal of Vascular Surgery

Tenorio et al

---

2019

Table IV. Rates of target artery instability (TAI) in 335 vessels incorporated by directional branches using balloonexpandable stent graft (BESG) or self-expandable stent graft (SESG) Overall (N ¼ 335)

BESGa (n ¼ 159)

27 (8)

18 (11)

14 (4)

8 (5)

SESGa (n ¼ 176) Viabahn (n ¼ 102)

Fluency (n ¼ 74)

9 (5)

7 (7)

2 (3)

.04

6 (3)

6 (6)

0

.46

P valuea

All vessels Any branch instability Branch-related occlusion or stenosis Occlusion Stenosis/kink Branch-related disconnection Branch-related endoleak IC and IIIC

4 (1)

1 (1)

3 (2)

3 (3)

0

.37

10 (3)

7 (4)

3 (2)

3 (3)

0

.15

1 (1)

0

1 (1)

0

1 (1)

.34

15 (4)

11 (7)

4 (2)

2 (2)

2 (2)

.04

Type IC endoleak

11 (3)

9 (6)

2 (1)

1 (1)

1 (1)

.02

Type IIIC endoleak

4 (1)

2 (1)

2 (1)

1 (1)

1 (1)

.92

Branch-related rupture

0

0

0

0

0

Branch-related death Branch-related reintervention Renal arteries Any branch instability Branch-related occlusion or stenosis

0

0

0

0

0

20 (6)

14 (9)

6 (3)

4 (4)

2 (3)

NA NA .04

122

57

65

65

0

19 (16)

13 (23)

6 (9)

6 (9)

e

.04

11 (9)

5 (9)

6 (9)

6 (9)

e

.93

Occlusion

3 (1)

0

3 (5)

3 (5)

e

.10

Stenosis/kink

8 (7)

5 (9)

3 (5)

3 (5)

e

.35

0

0

0

0

e

NA

Branch-related disconnection Branch-related endoleak IC and IIIC

10 (8)

9 (16)

1 (2)

1 (2)

e

.004

Type IC endoleak

8 (7)

7 (12)

1 (2)

1 (2)

e

.02

Type IIIC endoleak

2 (2)

2 (4)

0

0

e

.13

14 (11)

11 (19)

3 (5)

3 (5)

e

.01

213

102

111

37

74

8 (4)

5 (5)

3 (3)

1 (3)

2 (3)

.40

3 (1)

3 (3)

0

0

0

.07

Branch-related reintervention Mesenteric arteries Any branch instability Branch-related occlusion or stenosis Occlusion

1 (1)

1 (1)

0

0

0

.30

Stenosis/kink

2 (1)

2 (2)

0

0

0

.14

Branch-related disconnection

1 (1)

0

1 (1)

0

1 (1)

.34

Branch-related endoleak IC and IIIC

5 (2)

2 (2)

3 (3)

1 (3)

2 (3)

.72

Type IC endoleak

3 (1)

2 (2)

1 (1)

0

1 (1)

.51

Type IIIC endoleak

2 (1)

0

2 (1)

1 (3)

1 (1)

.92

6 (3)

3 (3)

3 (3)

1 (3)

2 (3)

.92

Branch-related reintervention

Values are reported as number of vessels (%). a Difference statistically significant between BESG and SESG.

type IC and type IIIC endoleaks, and more branch-related reinterventions. This occurred in a relatively short followup of 8 months, which raises concern, given that stent stenosis, occlusion, and endoleaks are time-dependent end points. Moreover, the short follow-up may be insufficient to detect stent integrity issues, such as material fatigue, fabric tear, fractures, or disconnections. The Fluency stent graft was applied primarily for mesenteric arteries because of smaller sheath profile compared with self-expandable Viabahn stent grafts (9F vs 11F). The low rate of TAI (3%) observed in our study reflects the use in mesenteric vessels and is similar to other reports.

Perhaps the most significant finding of this study was the higher rate of stent-related endoleaks, particularly type IC endoleaks, among patients treated by BESGs. This is a new finding, given that in the TAMBE early feasibility study, there was only one patient treated for a type IC endoleak among 52 targeted vessels.19 However, a fundamental difference between the two studies resides in aneurysm extent. Whereas the TAMBE early feasibility study included only patients with PRAs and extent IV TAAAs (with narrower inner aortic diameters), our study focused on patients with extent I to III TAAAs (90%), many of whom had challenging target vessel anatomy and large inner aortic diameters. Nonetheless, our study

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

11

-

Fig 4. Kaplan-Meier estimates of target artery instability (TAI) in 355 directional branches targeted using selfexpandable stent graft (SESG) or balloon-expandable stent graft (BESG).

Fig 5. Kaplan-Meier estimates of branch-related type IC or type IIIC endoleak in 355 directional branches targeted using self-expandable stent graft (SESG) or balloon-expandable stent graft (BESG).

showed low rates of type IC and type IIIC endoleaks for SESGs (2%). Interestingly, the use of SESGs was often coupled with greater degree of oversizing and more adjunctive bare-metal stents at the distal end, which may have provided better stent stability, preventing

retrograde stent migration and type IC endoleaks. The lower procedure metrics (eg, fluoroscopy, radiation, and contrast material volume) among patients treated by SEGSs is probably explained by incorporation of fusion imaging in patients treated by BESGs and the use of

12

Tenorio et al

Journal of Vascular Surgery ---

2019

Fig 6. Kaplan-Meier estimates of branch-related reintervention in 355 directional branches targeted using selfexpandable stent graft (SESG) or balloon-expandable stent graft (BESG).

adjunctive bare-metal stents and postdilation with SESGs, which required at least two additional guidewire exchanges. All type IC endoleaks occurred predominantly in patients with extensive aneurysms and wide inner aortic diameter, and most occurred on the left side with upgoing angulation (Supplementary Table IV, online only). There are several potential explanations for the occurrence of type IC endoleaks. First, it is possible that stent selection was not optimal and that a longer or larger stent, with greater oversizing and with use of bare-metal stents at the distal end, would have prevented some of the endoleaks. Second, BESGs assume a straight configuration between the branch cuff and the target vessel during balloon inflation. This straight configuration becomes curved once the balloon is deflated to accommodate tortuosity and changes in vector from longitudinal to transverse orientation. The difference in length may be accounted for by retrograde migration of the stent into the aneurysm sac. Conversely, during deployment of an SESG, the stent assumes a curved configuration at the time of deployment. Third, differences in radial force may affect continued enlargement of the target vessel, possibly leading to insufficient oversizing. Last, hemodynamics, respiratory movement, and aortic remodeling place unforeseen forces in directional branches that can lead to migration, kink, or loss of stent integrity. The properties of flexibility and radial force are variable among different types of BESGs, which should be

carefully considered before generalization is made to other types of BESGs. For the purpose of this study, all BESGs were VBX stent grafts. This is different from the study of Mastracci et al,17 which analyzed primarily Fluency vs iCAST stent grafts. In that study, there were only 17 Viabahn stent grafts and no Bentley BeGraft or BeGraft Plus stent grafts (Bentley Innomed, Hechingen, Germany). In general, one fundamental aspect is that balloon dilation tends to make the stent assume a straight configuration, which is then changed to curved configuration once the balloon is deflated. Stents that are more flexible in nature will tend to better adapt to curved anatomy. The effects of radial force on distal seal or progressive dilation of the native target vessel have not yet been determined. At present, there is a paucity of clinical data with directional branches targeted by other types of BESGs, and future studies should focus on describing the results in a critical manner. Several studies have shown that renal branches are more prone to late occlusion and have lower freedom from TAI compared with mesenteric branches or renal fenestrations.7,9,17,20,21 Factors that have been associated with branch failures include small vessel diameter, ostial disease, shortening of BESG, tortuous anatomy, and upward vessel angulation.7,22 We identified that downgoing renal artery configuration (negative angulation) was associated with lower risk of TAI, which is similar to other reports.23,24 Therefore, it is possible that some of the complications were due to poor selection of method of incorporation rather than the bridging stent itself. We

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

13

-

recommend careful analysis of renal angles in selecting the ideal method of incorporation, reserving branches for downgoing vessels in larger inner aortic diameters. The study strengths are the prospective design, carefully monitored data, and use of standardized protocols to design the aortic stent. Several limitations need to be mentioned. Stent selection was not randomized, introducing possible bias on selection of bridging stents; however, the change in practice to BESGs was nearly complete in 2017 (Fig 1). Follow-up was short in the BESG group, and rate of time-dependent outcomes, such as stent occlusion, type IC or type IIIC endoleaks, and branch reinterventions, still needs to be determined with longer follow-up. Overall, this study represents an analysis of specific changes in the choice of bridging stent for directional branches and shows some of the shortcomings of the BESG on early follow-up. Based on these results, we have changed our choice of bridging stent for the renal arteries from BESG (VBX) to SESG (Viabahn), especially in repair of extensive TAAAs.

CONCLUSIONS Directional branches were associated with high technical success and low rates of stent occlusion, independent of stent type. However, primary patency, freedom from TAI, and freedom from type IC and type IIIC endoleaks were lower for BESGs.

5.

6.

7.

8.

9.

10.

11.

12.

13.

AUTHOR CONTRIBUTIONS Conception and design: ET, BM, RD, TM, GO Analysis and interpretation: ET, JK Data collection: ET, JK, AD, JH Writing the article: ET Critical revision of the article: ET, JK, BM, RD, TM, AD, JH, GO Final approval of the article: ET, JK, BM, RD, TM, AD, JH, GO Statistical analysis: ET Obtained funding: Not applicable Overall responsibility: GO

REFERENCES 1. Oderich GS, Ribeiro M, Hofer J, Wigham J, Cha S, Chini J, et al. Prospective, nonrandomized study to evaluate endovascular repair of pararenal and thoracoabdominal aortic aneurysms using fenestrated-branched endografts based on supraceliac sealing zones. J Vasc Surg 2017;65:1249-59. e10. 2. Verhoeven EL, Paraskevas KI, Oikonomou K, Yazar O, Ritter W, Pfister K, et al. Fenestrated and branched stentgrafts to treat post-dissection chronic aortic aneurysms after initial treatment in the acute setting. J Endovasc Ther 2012;19:343-9. 3. Ferrer C, Cao P, De Rango P, Tshomba Y, Verzini F, Melissano G, et al. A propensity-matched comparison for endovascular and open repair of thoracoabdominal aortic aneurysms. J Vasc Surg 2016;63:1201-7. 4. O’Brien N, Sobocinski J, D’Elia P, Guillou M, Maioli F, Azzaoui R, et al. Fenestrated endovascular repair of type IV

14.

15.

16.

17.

18.

19.

20.

21.

thoracoabdominal aneurysms: device design and implantation technique. Perspect Vasc Surg Endovasc Ther 2011;23: 173-7. Haulon S, D’Elia P, O’Brien N, Sobocinski J, Perrot C, Lerussi G, et al. Endovascular repair of thoracoabdominal aortic aneurysms. Eur J Vasc Endovasc Surg 2010;39:171-8. Greenberg R, Eagleton M, Mastracci T. Branched endografts for thoracoabdominal aneurysms. J Thorac Cardiovasc Surg 2010;140(Suppl):S171-8. Mastracci TM, Greenberg RK, Eagleton MJ, Hernandez AV. Durability of branches in branched and fenestrated endografts. J Vasc Surg 2013;57:926-33; discussion: 933. Oderich GS, Ribeiro M, Reis de Souza L, Hofer J, Wigham J, Cha S. Endovascular repair of thoracoabdominal aortic aneurysms using fenestrated and branched endografts. J Thorac Cardiovasc Surg 2017;153:32-41.e7. Panuccio G, Bisdas T, Berekoven B, Torsello G, Austermann M. Performance of bridging stent grafts in fenestrated and branched aortic endografting. Eur J Vasc Endovasc Surg 2015;50:60-70. Bertoglio L, Loschi D, Cambiaghi T, Mascia D, Kahlberg A, Melissano G, et al. Preliminary outcomes of the LifeStream balloon-expandable covered stent in fenestrated and branched thoracoabdominal endovascular repairs. J Endovasc Ther 2018;25:230-6. Chuter T, Greenberg RK. Standardized off-the-shelf components for multibranched endovascular repair of thoracoabdominal aortic aneurysms. Perspect Vasc Surg Endovasc Ther 2011;23:195-201. Ribeiro M, Macedo T, Vrtiska T, Oderich G. Planning endovascular aortic repair with standard and fenestratedbranched endografts. J Cardiovasc Surg 2017;58:204-17. Mendes BC, Oderich GS. Selection of optimal bridging stents for fenestrations and branches. In: Oderich GS, editor. Endovascular aortic repair. Cham: Springer; 2017. p. 359-74. Mendes BC, Rathore A, Ribeiro MS, Oderich GS. Off-the-shelf fenestrated and branched stent graft designs for abdominal aortic aneurysm repair. Semin Vasc Surg 2016;29:74-83. Chaikof EL, Blankensteijn JD, Harris PL, White GH, Zarins CK, Bernhard VM, et al. Reporting standards for endovascular aortic aneurysm repair. J Vasc Surg 2002;35:1048-60. Fillinger M, Greenberg R, McKinsey J, Chaikof E; Society for Vascular Surgery Ad Hoc Committee on TEVAR Reporting Standards. Reporting standards for thoracic endovascular aortic repair (TEVAR). J Vasc Surg 2010;52:1022-33. Mastracci TM, Carrell T, Constantinou J, Dias N, MartinGonzalez T, Katsargyris A, et al. Editor’s choicedeffect of branch stent choice on branch-related outcomes in complex aortic repair. Eur J Vasc Endovasc Surg 2016;51:536-42. Schneider DB, Oderich GS, Farber MA, Schanzer A, Beck AW, Timaran CH, et al. SS03. Target artery outcomes after branched and fenestrated endovascular repair of pararenal and thoracoabdominal aortic aneurysms in the US investigational device exemption experience. J Vasc Surg 2018;67: e83. Oderich GS, Farber MA, Silveira PG, Tadros R, Marin M, Fillinger M, et al. Technical aspects and 30-day outcomes of the prospective early feasibility study of the GORE EXCLUDER Thoracoabdominal Branched Endoprosthesis (TAMBE) to treat pararenal and extent IV thoracoabdominal aortic aneurysms. J Vasc Surg 2019;70:358-68.e6. Martin-Gonzalez T, Mastracci T, Carrell T, Constantinou J, Dias N, Katsargyris A, et al. Mid-term outcomes of renal branches versus renal fenestrations for thoraco-abdominal aneurysm repair. Eur J Vasc Endovasc Surg 2016;52:141-8. Sylvan J, Brier C, Wolski K, Yanof J, Goel V, Kuramochi Y, et al. Impact of alterations in target vessel curvature on branch

14

Tenorio et al

Journal of Vascular Surgery ---

durability after endovascular repair of thoracoabdominal aortic aneurysms. J Vasc Surg 2016;63:634-41. 22. Suh GY, Choi G, Herfkens RJ, Dalman RL, Cheng CP. Respiration-induced deformations of the superior mesenteric and renal arteries in patients with abdominal aortic aneurysms. J Vasc Interv Radiol 2013;24:1035-42. 23. Conway BD, Greenberg RK, Mastracci TM, Hernandez AV, Coscas R. Renal artery implantation angles in thoracoabdominal aneurysms and their implications in the era of branched endografts. J Endovasc Ther 2010;17:380-7.

2019

24. Sugimoto M, Panuccio G, Bisdas T, Berekoven B, Torsello G, Austermann M. Tortuosity is the significant predictive factor for renal branch occlusion after branched endovascular aortic aneurysm repair. Eur J Vasc Endovasc Surg 2016;51:350-7. Submitted Feb 5, 2019; accepted Jul 16, 2019.

Additional material for this article may be found online at www.jvascsurg.org.

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

14.e1

-

Supplementary Table I (online only). Type of device configuration of 93 patients treated with customized patient-specific device Overall (N ¼ 93), No. (%)

Configurations 2 directional branches/2 fenestrations

34 (37)

4 directional branches

23 (25)

1 directional branch/3 fenestrations

11 (12)

3 directional branches

7 (8)

1 directional branch/2 fenestrations

4 (4)

2 directional branches

4 (4)

2 directional branches/1 fenestration

2 (2)

2 directional branches/3 fenestrations

2 (2)

3 directional branches/1 fenestration

2 (2)

1 directional branch/1 fenestration

1 (1)

3 directional branches/2 fenestrations

1 (1)

1 directional branch/4 fenestrations

1 (1)

5 directional branches

1 (1)

Supplementary Table II (online only). Major adverse events (MAEs) and 30-day mortality in 126 patients treated by fenestrated and branched endovascular aneurysm repair (F-BEVAR) using balloon-expandable stent graft (BESG) or selfexpandable stent graft (SESG) for directional branches

Early death Any MAE

Overall (N ¼ 126)

BESGa (n ¼ 54)

SESGa (n ¼ 62)

SESG and BESG (n ¼ 10)

P valuea

1 (1)

1 (2)

0

0

.28

40 (32)

19 (35)

13 (26)

5 (50)

.27

Estimated blood loss >1000 mL

15 (12)

6 (11)

9 (15)

0

.59

Acute kidney injury by RIFLE criteria

15 (12)

7 (12)

4 (8)

4 (17)

.58

New-onset dialysis

1 (1)

0

1 (2)

0

.35

Myocardial infarction

3 (2)

0

3 (5)

0

.10

Respiratory failure

4 (3)

1 (2)

3 (5)

0

.38

Any spinal cord injury

10 (8)

4 (7)

4 (6)

2 (20)

.84

Grade 1-2

3 (2)

1 (2)

2 (3)

0

.64

Grade 3a-3c (paraplegia)

7 (6)

3 (6)

2 (3)

2 (20)

.54

Permanent paraplegia

3 (2)

2 (4)

0

1 (10)

.13

Any stroke

4 (3)

1 (2)

3 (5)

0

.38

0

0

0

0

NA

4 (3)

1 (2)

3 (5)

0

.38

1 (1)

0

0

1 (4)

NA

Major stroke Minor stroke/TIA Bowel ischemia

NA, Not applicable; RIFLE, Risk, Injury, Failure, Loss of kidney function, and End-stage; TIA, transient ischemic attack. Values are reported as number of patients (%). a Difference statistically significant between BESG and SESG.

14.e2

Journal of Vascular Surgery

Tenorio et al

---

2019

Supplementary Table III (online only). Predictors by univariate and multivariate Cox regression analysis for renal artery instability in 122 renal arteries incorporated with directional branches using balloon-expandable stent graft (BESG) or selfexpandable stent graft (SESG) Confidence interval Renal arteries

Hazard ratio

Lower

Upper

P value

Univariate Renal angulation (downgoing oriented)

.963

BESG: VBX

.930

7.7

.998

2.6

22.5

.036 <.0001

SESG: Viabahn

.130

.044

.382

<.0001

Adjunctive self-expandable bare-metal stent

.194

.063

.598

.004

.994

.024

Multivariate Renal angulation (downgoing oriented)

.954

BESG: VBX

.915

6.8

1.4

33.5

.019

Supplementary Table IV (online only). Anatomic characteristics of all patients with type IC endoleak targeted by balloonexpandable stent graft (BESG) VBX Vessel Inner aortic Renal Aneurysm Target diameter, diameter, angulation, extent vessel mm mm degrees

Stent used, mm

Adjunctive selfFinal diameter Timing for expandable after dilation, Seal zone reintervention, mm length, mm days stent

1 Extent 4

LRA

6

30

þ1.4

6  59

No

6.3

15.3

2 Extent 2

LRA

7

29

þ3.2

7  79

No

7.8

7.1

3 Extent 2

LRA

5

46


13.6

5  59/6  59

No

6.7

5.25

4 Extent 3

LRA

5

43


3.9

5  59

No

6.5

5 Pararenal

LRA

5

37


5.7

5  79

Yes

6.7

7.64 12.5

147 30 229 139 462

6 Extent 2

LRA

5.5

49

þ38.4

5  79

No

5.8

8.1

115

7 Extent 2

CA

9

34

e

8  59

No

8.7

13.8

125

8 Extent 2

SMA

8.7

64

e

8  79

No

8.8

18.7

6

Extent 2

RRA

6

65

þ10.6

6  79

No

6.3

8.3

6

CA, Celiac axis; LRA, left renal artery; SMA, superior mesenteric artery; RRA, right renal artery.

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

-

Supplementary Fig 1 (online only). Illustration showing selection of alignment or bridging stents. A, Ideal configuration of bridging stent for directional branches would combine different radial force, flared and baremetal stents. B, Types of alignment and bridging stents for fenestrations and branches. C, Viabahn selfexpandable stent graft (SESG) for the renal arteries and Fluency SESG for the mesenteric arteries. D, VBX Viabahn balloon-expandable stent graft (BESG), which has been used under clinical investigation for the TAMBE device. (Reproduced by permission of Mayo Foundation for Medical Education and Research. All rights reserved.)

14.e3

14.e4

Tenorio et al

Journal of Vascular Surgery ---

Supplementary Fig 2 (online only). A, Kaplan-Meier estimates of primary patency in 122 renal artery directional branches targeted using self-expandable stent graft (SESG) or balloon-expandable stent graft (BESG). B, KaplanMeier estimates of primary patency in 213 mesenteric artery directional branches targeted using SESG or BESG.

2019

Journal of Vascular Surgery Volume

-,

Number

Tenorio et al

-

Supplementary Fig 3 (online only). Kaplan-Meier estimates of target artery instability (TAI) in 122 renal artery directional branches targeted using self-expandable stent graft (SESG) or balloon-expandable stent graft (BESG).

Supplementary Fig 4 (online only). Kaplan-Meier estimates of branch-related type IC or IIIC endoleak in 122 renal artery directional branches targeted using self-expandable stent graft (SESG) or balloon-expandable stent graft (BESG).

14.e5

14.e6

Tenorio et al

Journal of Vascular Surgery ---

Supplementary Fig 5 (online only). Kaplan-Meier estimates of branch-related reintervention in 122 renal artery directional branches targeted using self-expandable stent graft (SESG) or balloon-expandable stent graft (BESG).

2019