Early and Long-Term Performance of Stent Grafts Released in Dacron vs. Native Ascending Aorta During Hybrid Aortic Arch Repair

Abstract

Objective: To evaluate the impact of a prosthetic and native aorta proximal landing zone (PLZ) for hybrid thoracic aorta repair with a stent graft deployed in the Zone 0 ascending aorta. Methods: Retrospective analysis of 100 patients who underwent Zone 0 hybrid thoracic aorta repair between 2005 and 2022, at a single institution. Fifty-eight (58%) had a Dacron PLZ (ascending aorta replacement with debranching), and forty-two (42%) had a native ascending aorta PLZ (off-pump aortic arch debranching). All the surviving patients had at least one radiological follow-up. Results: Patients with a native aorta PLZ were older (p = 0.01) and had higher rates of pulmonary disease (p = 0.01) and chronic kidney disease (p = 0.01) and a higher Logistic EUROscore (p = 0.02). In-hospital mortality was similar between the two groups (2.3% vs. 1.7%; p = 0.68). Retrograde aortic dissection was observed in four (9.2%) of the native PLZ group. The mean follow-up was 37 ± 22 months in the Dacron PLZ group and 42 ± 20 months in the native PLZ group (p = 0.05). At 1,3, and 5 years, freedom from Type Ia endoleak was significantly higher in patients with a Dacron PLZ: 100% vs. 96 ± 0.3%, 100% vs. 85 ± 1%, and 100% vs. 66 ± 1.3% (p = 0.03). Despite this, the 1-, 3-, and 5-year survival was similar between groups: 100% vs. 96 ± 0.4%, 90 ± 1% vs. 89 ± 0.6%, and 66 ± 1% vs. 64 ± 2%. (p = 0.96). Conclusions: Endograft deployment in a prosthetic ascending aorta is a safe and durable option for aortic repair. A Dacron PLZ is associated with a lower rate of Type Ia endoleak and can prevent retrograde aortic dissection when compared with native aorta.

1. Introduction

In recent years, thoracic endovascular aortic repair (TEVAR) has emerged as a less invasive and alternative treatment over a standard surgical approach, for pathologies of the thoracic aorta [1]. Aortic arch aneurysms, with or without the involvement of the proximal thoracic aorta, are essentially aneurysms requiring a proximal deployment of the endograft in the ascending aorta (Zone 0 Criado classification). For this reason, a hybrid approach that combines aortic arch debranching followed by TEVAR has been proposed to safely relocate the origin of aortic arch’s vessel and allow these vessels to be covered by the stent graft [2]. Although theoretically attractive as a means to create a long and stable proximal landing zone, the impact of a prosthetic vs. native aorta for TEVAR has rarely been explored. Considering that Zone 0 is rapidly becoming the preferred landing zone for hybrid and total arch repair, understanding the mechanical interactions between the prosthetic or native aorta and the graft is mandatory. Nevertheless, this issue has rarely been investigated [3]. The aim of the present study is, therefore, to evaluate the early and long-term outcome of hybrid thoracic aorta repair in patients with native and prosthetic proximal landing zones (PLZs).

2. Materials and Methods

2.1. Patient Selection and Data Collection

This was a retrospective, observational, cohort study of prospectively collected data from consecutive patients who underwent hybrid thoracic aorta repair with a proximal landing zone (PLZ) in the ascending aorta (Criado Classification, Zone 0), at the G. Pasquinucci Heart Hospital between September 2005 and June 2022; no patients were excluded. The data collection form was entered into an institutional database and included ten sections that were filled in consecutively by anesthesiologists, surgeons, interventional cardiologists, intensive care units, and ward doctors. The resulting base sample contained detailed clinical information on 100 patients, 58 (58%) of whom had a Dacron PLZ (previously replaced ascending aorta with concomitant arch’s vessel debranching followed by TEVAR) and 42 (42%) patients with native ascending aorta PLZs (isolated off-pump aortic arch’s debranching followed by TEVAR).

The study was approved by the clinical audit committee of the G. Pasquinucci Heart Hospital to meet ethical and legal requirements, and individual consent was waived.

2.2. Definitions

A native aorta PLZ consists of any endograft landing zone in the native ascending aorta, proximal to the origin of the brachiocephalic trunk. A prosthetic PLZ consists of any endograft released in the Dacron ascending aorta placed during a prior or concomitant procedure (Figure 1).

A Type Ia endoleak was defined as any leak originating at the level of the PLZ, documented with angiography and/or with an Angio-CT scan. Retrograde aortic dissection (RAD) was defined as any new acute aortic dissection involving the ascending aorta and/or the aortic root, after the TEVAR procedure. The diagnosis of RAD was confirmed with a transesophageal echo and/or Angio-CT scan. Postoperative renal dysfunction was defined as any increase of at least 50% of the serum creatinine from baseline. Our definitions of other postprocedural complications have been previously reported [4].

2.3. TEVAR Procedure

The patient selection for TEVAR and device delivery was performed by our Institutional Aortic Team (cardiac and vascular surgeons, interventional cardiologist, anesthetist, and radiologist). Preprocedural TEVAR planning was performed using three-dimensional curved multiplanar reconstruction (MPR) with centerline analysis on an Osirix three-dimensional workstation (Pixmeo SARL, Geneva, Switzerland). The procedures were performed percutaneously whenever possible; otherwise, the exposure of the femoral artery was obtained through a small surgical incision. Endograft delivery was performed under pharmacologically induced systemic hypotension and with rapid ventricular pacing from the femoral vein. In the native aorta PLZ, endograft oversizing was performed following device instructions for use. For the endografts deployed in the Dacron PLZ, a 15% to 20% oversizing was used. During the study period, the following thoracic endografts were used: Bolton Relay (Bolton Medical, Barcelona, Spain) in 95 patients; Gore TAG (W. L. Gore & Associates, Flagstaff, Ariz) in 2 patients, and Talent and Valiant (Medtronic Vascular, Santa Rosa, Calif) in 3 patients. A pre-discharge Angio-CT scan was performed on all the patients.

2.4. Follow-Up

All the patients underwent periodical follow-up visits with an Angio-CT scan at 3 and 12 months, and then annually. If during the follow-up a new onset endoleak was detected, radiological and clinical follow-up was tailored to the patient’s characteristics. Patient medical records and imaging studies were reviewed to identify patients experiencing Type Ia endoleak.

2.5. Statistical Analysis

Continuous data were expressed as mean ± SD, and categorical data as percentages. The Kolmogorov–Smirnov test was used to check for normality of data in the 2 groups before further analysis. Differences between Native and Dacron PLZ groups were compared with the use of a chi-square test or Fisher’s exact test for categorical variables and Student’s or Wilcoxon rank sum tests, as appropriate, for continuous variables. Event-free survival and freedom from Type Ia endoleak were estimated using the Kaplan–Meier method and compared between groups using the log-rank test. All reported p-values are 2-sided, and p-values of <0.05 were considered to indicate statistical significance. All statistical analysis was performed with SPSS version 19.0 (SPSS Inc., Chicago, IL, USA).

3. Results

3.1. Patient Characteristics

Baseline characteristics of the study population are shown in

Table 1. Preoperative characteristics.

Variable n (%)	Patients 100	Dacron PLZ (58 Patients)	Native PLZ (42 Patients)	p-Value
Age, mean (SD), yrs	71 ± 11	69 ± 14	74 ± 11	0.01
Female gender	12	9 (16)	3 (6.1)	0.12
Diabetes mellitus	12	6 (10)	6 (10)	0.59
Hypertension	76	45 (77)	31 (74)	0.72
Cerebrovascular Disease	7	4 (8)	3 (8)	0.65
Peripheral Artery Disease	39	19 (33)	20 (48)	0.17
COPD	12	2 (2)	10 (23)	0.01
Chronic Kidney failure	11	2 (2)	9 (20)	0.01
History of Smoke	25	11 (19)	14 (33)	0.24
Active Smoker	18	12 (20)	6 (10)	0.39
Urgent TEVAR	11	5 (8)	6 (15)	0.28
Aneurysm	49	25 (44)	24 (53)	0.41
PAU	6	2 (2)	4 (7)	0.33
Chronic Type A Dissection	21	21 (36)	0	0.001
Acute Type B Dissection	8	3 (5)	5 (10)	0.37
Chronic Type B Dissection	16	7 (11)	9 (17)	0.31

COPD = Chronic obstructive pulmonary disease; PAU = Penetrating aortic ulcer.

. Compared with the Dacron PLZ group, patients who had a native PLZ were older (p = 0.01) and were more likely to have severe COPD (p = 0.01), chronic kidney disease (p = 0.01), and a higher Logistic EUROscore (p = 0.02). The main indication for TEVAR was aneurismatic pathology in both groups. However, in 36% of the Dacron PLZ group, TEVAR was performed for the development of distal pathology after prior type A aortic dissection repair.

3.2. First Surgical Stage Details

In all the native aorta PLZ patients, the first surgical step consisted of an isolated ascending aorta to the innominate artery and left carotid artery bypass [5]. In 57 patients of the Dacron PLZ group, we performed an ascending aorta and partial arch replacement with concomitant debranching of the aortic arch, using a trifurcated vascular graft [6]. In one patient, a prior replaced ascending aorta was used as the proximal landing zone after the brachiocephalic bypass. Intrathoracic subclavian artery revascularization or extra-anatomic left carotid to left subclavian artery bypass was performed in 16 (38%) patients of the native aorta PLZ group and 19 (32%) of the Dacron PLZ group.

3.3. Procedural Data and Early Outcomes

TEVAR procedural data are reported in

Table 2. Procedural data.

Variable n (%)	Dacron PLZ (58 Patients)	Native PLZ (42 Patients)	p-Value
Mean PLZ length cm	3.8 ± 0.9	2.8 ± 0.7	0.001
Number of grafts	1.4 ± 0.5	1.3 ± 0.5	0.74
Diameter of grafts mm	38.2 ± 3.1	38.8 ± 3.9	0.83
Length of covered Aorta mm	191 ± 31	186 ± 26	0.68
Percutaneous Access	34 (94.4)	33 (84.6)	0.18

PLZ = Prosthetic landing zone.

.

The interval between the first surgical step and TEVAR was 3.1 ± 2.8 months in the native aorta PLZ and 6.3 ± 5.9 months in the Dacron PLZ group (p = 0.001). An urgent procedure with concomitant surgical and endovascular steps was necessary in 15% and 8% of the native and Dacron PLZ patients, respectively. There were no significant differences in terms of the number and length of endografts used. In-hospital results are reported in

Table 3. Early Outcomes.

Variable n (%)	Dacron PLZ (58 Patients)	Native PLZ (42 Patients)	p-Value
In-hospital mortality	1 (1.7)	1 (2.3)	0.64
Stroke	0	0	-
TIA	2 (3.4)	2 (4.6)	0.66
Retrograde Aortic Dissection	0	4 (9.2)	0.15
Transient Paraplegia	1 (2.8)	2 (5.1)	0.53
Major Bleeding	1 (2.8)	1 (2.6)	0.73
Renal Dysfunction	1 (2.8)	1 (2.6)	0.73
Pulmonary complications	2 (5.6)	0	0.22

TIA = Transient ischemic attack.

. The overall in-hospital mortality was 2.6% without significant differences between the two groups (2.3% vs. 1.7%; p = 0.64). A transient ischemic attack was observed in two native aorta PLZs and two Dacron PLZ patients (4.6% vs. 3.4%; p = 0.66). The incidence of postprocedural retrograde aortic dissection was observed in four (9.2%) of the native PLZ group, whereas no patients of the Dacron PLZ group had RAD. The diameter of the ascending aortas of these four patients were 39, 40, 41, and 42 mm, respectively. Patients with RAD underwent emergent surgical repair. No differences were observed between the two groups for other complications.

3.4. Long Term Outcomes

The mean follow-up was 37 ± 22 months in the Dacron PLZ group and 42 ± 20 months in the native PLZ group (p = 0.05). Radiological follow-up was 100% complete. Four patients in the native aorta PLZ group developed a Type Ia Endoleak at 6, 12, 15, and 32 months, whereas no Type Ia endoleak was observed in the Dacron PLZ group. At 1, 3, and 5 years, freedom from Type Ia endoleak was significantly higher in patients with a Dacron PLZ: 100% vs. 96 ± 0.3%, 100% vs. 85 ± 1%, and 100% vs. 66 ± 1.3% (p = 0.03) (Figure 2). At follow-up, eight patients required a reintervention: four patients (9.5%) in the native aorta PLZ group and four (6.8%) in the Dacron PLZ. At 1, 3, and 5 years (p = 0.36), freedom from aortic reintervention was not statistically different between the two groups (Figure 3).

At follow-up, seven patients died: five patients (11.9%) in the native aorta PLZ and four (6.8%) in the Dacron PLZ. The overall 1-, 3-, and 5-year survival was 98 ± 1%, 92 ± 4%, and 74 ± 12% with no significant differences between groups: 100% vs. 96 ± 0.4%, 90 ± 1% vs. 89 ± 0.6%, and 66 ± 1% vs. 64 ± 2% (p = 0.96) (Figure 4).

4. Discussion

Our study has two main results: (1) hybrid thoracic aorta repair can be performed safely with low procedural mortality and morbidity, and (2) using a Dacron PLZ reduces the incidence of Type I endoleak and RAD.

Type I or attachment-site endoleak is the most frequent complication associated with stent-graft treatment of aortic pathology [7]. The fate of Type I endoleaks is still a matter of debate. Despite some studies reporting spontaneous resolution (1 week to 8 months), if endoleak persists, progressive aneurysmal dilatation is likely and requires conversion to open surgery and close follow-up [8].

Predictive characteristics that may represent risk factors for endoleak formation are a short proximal landing zone, angulation of the aortic arch, incorrect sizing, and procedural mistakes. Whatever the etiology, the final mechanism is poor contact between the stent graft and the proximal landing zone. This raises the issue of the ideal PLZ, whose quality and stability largely determine the technical success of the procedure. To create a stable and long PLZ, ascending aorta replacement during the first surgical step has been proposed by many authors as an alternative to isolated aortic arch debranching [9]. Despite the creation of a prosthetic ascending aorta requiring the use of cardiopulmonary bypass (CPB) and hypothermic circulatory arrest, the theoretic benefit consists of mechanical stability and caliber uniformity, which allow an increase in the percentage of stent oversizing, thus resulting in a more homogenous sealing and less probability of developing endoleak. In addition, a Dacron PLZ is potentially less susceptible to aortic remodeling than a native aorta and, therefore, may be more resistant to delayed endoleak. Moreover, the vast majority of endovascular prostheses currently in use are not designed for the treatment of arch pathologies. Therefore, using the native ascending aorta as the proximal landing zone may result in incomplete alignment and thus in the development of endoleaks. Nevertheless, nowadays, only a few studies have investigated the role of a prosthetic PLZ vs. native aorta for hybrid thoracic aorta repair. In 2012, Esposito and colleagues reported the mid-term results of type A aortic dissection repair with preventive arch debranching [10]. During the emergent surgical procedure, a prosthetic landing zone was created using a specifically designed vascular prosthesis. The mean length of the PLZ was 3.1 ± 0.5 cm. This prophylactic proximal landing zone was used for second-stage TEVAR in 34 patients with ongoing dilatation of the residual dissected thoracic aorta. The authors reported 100% technical success and no patients developed Type I endoleak at follow-up. In a recently published paper, Kotha et al. addressed the issue of the complications of TEVAR at the level of the prosthetic PLZ [11]. They identified PLZ complications in 4 of 20 patients who had undergone aortic arch replacement with arch debranching and subsequent TEVAR. The complications reported comprised two Type Ia endoleaks, one graft migration, and one graft infolding. Out of these patients, three required a second proximal intervention.

In 2014, Ganapathi and colleagues published a study primarily focused on the incidence of attachment-site endoleaks in Dacron versus native aorta landing zones [12]. They performed a retrospective analysis of 319 patients with different landing zones, undergoing TEVAR at the Duke University Medical Center, over 10 years. The authors reported a similar rate of endoleak between native aorta, Dacron, and endograft landing zones (p = 0.44). Interestingly, all the Type Ia endoleaks with Dacron landing zones occurred in the first tertile of their TEVAR experience in patients who had a relatively short proximal landing zone. However, they did not observe any endoleak after switching to a policy of increasing the length of the Dacron PLZ to superior to 4 cm. In their paper, therefore, the authors emphasize the importance of obtaining a long prosthetic landing zone to avoid endoleak.

In 2005, we started our TEVAR program, and from that moment, the number of patients who had a hybrid thoracic aorta repair has increased significantly. During the study period, we progressively moved toward zone 0 repair, and from 2006, partial aortic arch replacement with aortic arch debranching has been our preferred strategy for proximal landing zone creation. Isolated aortic arch debranching with brachiocephalic bypass is used only in those patients who have an absolute contraindication to hypothermic circulatory arrest. In addition, considering that at least one-third of patients with prior type A aortic dissection repair will need an additional intervention for late-developing downstream pathology after surgery, from 2006 we have routinely performed a prophylactic debranching during acute type A aortic dissection repair, in the vast majority of patients [6].

Our results reconfirm the positive effect of prosthetic PLZ in reducing the incidence of proximal landing zone complications. Indeed, long-term radiological follow-up demonstrated that patients with a Dacron PLZ have a significantly lower probability of developing Type Ia endoleak. We speculate that our strategy of obtaining a Dacron PLZ of at least 3.5 cm may have contributed to these encouraging results. Indeed, the length of stent-wall apposition at the proximal landing zone has been shown to have a significant effect on device stability, with more than 18 mm of proximal landing zone apposition reducing the risk of endoleak [13]. Interestingly in all the patients with a native PLZ who have developed a proximal endoleak, a various grade of bird beak effect was evident immediately after the procedure. The bird beak effect, which can be defined as a gap between the aortic wall and the stent, with stent protrusion into the aortic lumen of more than 5 mm, may lead to incomplete sealing, resulting in a risk of attachment site endoleak or stent graft migration.

Even though the native PLZ group of patients was older and had more comorbidities, we speculate that this did not affect the interaction between the graft and landing zone.

In addition to attachment-site endoleak, a Dacron PLZ confers absolute protection against the life-threatening complication of post-TEVAR retrograde aortic dissection. RAD after TEVAR is a serious complication that requires emergency surgery and is associated with high operative mortality and morbidity [14]. Different mechanisms for post-TEVAR RAD have been described, and the cause is usually multifactorial [14,15]. Pre-existing diseases of the ascending aorta, such as atheromatic plaques or calcification, may predispose to RAD after TEVAR. Also, the mechanical stress applied by semirigid stent grafts on the ascending aorta might produce intimal injuries and subsequent RAD.

In our series, we observed RAD in four patients with a native PLZ (in these patients, the PLZ diameter ranged between 39 and 42 mm). Among these, one patient had RAD 1 day after the TEVAR procedure, while the other two were at 1 and 2 months, respectively. Ascending aorta dilatation probably represents the main risk factor for RAD after TEVAR. Releasing a rigid stent graft on a fragile aortic wall associated with aortic dilatation increases the risk of intimal injury and dissection. Also, direct aortic manipulation with catheters and sheaths may create minimal injury. In an effort to reduce the incidence of RAD, we prefer to replace the ascending aorta if its size exceeds 40 mm, and we perform the procedure under pharmacologically induced systemic hypotension. In addition, delivery of the stent graft is performed under rapid ventricular pacing. Despite some authors suggesting combining the surgical and endovascular procedures in the same operation, we prefer to perform TEVAR in a second session one month after the aortic debranching to avoid tension on a “fresh” ascending aortic anastomosis.

This study has several limitations. It is based on a retrospective analysis of a small cohort of patients. The possibility of selection bias between native and Dacron PLZ is the main limitation of our study. In addition, our series encompassed our initial learning curve with hybrid aortic repair, and this might have influenced the results. Another important limitation is the clinical disparity between the two groups.

5. Conclusions

In conclusion, hybrid thoracic aorta repair can be performed safely. Patients undergoing Zone 0 TEVAR are at risk for Type I endoleak and RAD, and the current study shows that Dacron landing zones might represent an option to reduce the incidence of these complications. In an era where total endovascular aortic arch repair is rapidly appearing, the importance of creating a lengthy and stable proximal landing zone should not be underestimated.