Abstract
Purpose
Bridging stent stability is crucial for efficacy and safety of branched aortic endovascular repair (bEVAR) of thoracoabdominal aortic aneurysms (TAAAs). In this study, we assess the performance of the new Viabahn Balloon-Expandable endoprosthesis (VBX) in bEVAR. Based on our learning curve we give recommendations for a safe and effective use of the device.
Materials and Methods
We prospectively collected the data of patients with TAAAs undergoing bEVAR between December 2017 and December 2019. All patients with implantation of at least 1 VBX stent-graft as bridging stent were included in our single-center analysis. Demographic, comorbidity, and computed tomography angiography (CTA) data of 112 patients were retrospectively evaluated. Primary endpoint was a composite of branch-related technical success and freedom from target vessel instability. Secondary endpoints were clinical and ongoing clinical success.
Results
Primary endpoint: technical success was achieved in all patients (100%) with a freedom from target vessel instability of 96.3% after a median follow-up of 18 months. Overall mortality was 13.4% (n=15) and 13 patients underwent secondary interventions, 12 of them are still alive and 1 suffered from aneurysm sac expansion, consequently an ongoing clinical success of 75.9% was reached. After modification of the implantation technique during the course of the study by selecting longer stent lengths after accurate estimation of vessel curvature and expected adaptation of the flexible endoskeleton to the specific anatomical conditions, no type Ic endoleaks were observed in the last 70 cases.
Conclusions
The VBX stent-graft can be safely used as bridging stent for branched thoracoabdominal repair. However, learning curve should be considered to avoid type Ic endoleak and edge stenosis. Based on this experience longer landing zones and 2-step deployment of VBX are useful for successful bridging also of challenging target vessels.
Keywords
Introduction
Since its initial description branched endovascular aortic aneurysm repair (bEVAR) has advanced to an established alternative technique in the treatment of thoracoabdominal aneurysms (TAAAs). 1
Compared with open repair, bEVAR is associated with lower perioperative morbidity and mortality. 2 However, secondary procedures are frequent and often caused by target vessel instability.3–5 Bridging stent-graft (BSG) occlusion, stenosis, type Ic, type IIIb, and type IIIc endoleaks have been observed.6,7 Therefore, performance of bridging stent-grafts is of paramount importance in bEVAR for TAAA. In contrast to patients with juxtarenal aneurysm the distance between the main endograft and the target vessels (TVs) is increased in TAAA. Thus, the risk of BSG kinking, migration, and loss of alignment between endograft branch and TV is elevated. Moreover, target vessels have variable directions and changing angulations during respiratory cycles. Therefore, the BSG should be flexible in order to accommodate variation of the anatomy and changes of aneurysm morphology over time.
Recently, a new-generation balloon-expandable stent-graft (Viabahn Balloon Expandable, VBX, WL Gore and Associates, Flagstaff, AZ) has been introduced on the market.
The VBX allows a similar flexibility compared with a self-expanding stent while maintaining the advantages of balloon-expandable stent as exact deployment and high radial force.8,9 Initial experience with this BSG has been described in combined fenestrated/branched EVAR (f/bEVAR) and chimney EVAR/fEVAR (chEVAR/fEVAR) cohorts with diverging results, which do not allow a discrimination of VBX stent-graft performance under different technical requirements.8–12
So far, the stent has not been extensively assessed in a solely TAAA cohort. The aim of this study is to evaluate the VBX as a bridging device in bEVAR for TAAAs.
Materials and Methods
Patient Population and Inclusion Criteria
All patients with TAAA undergoing bEVAR with implantation of at least 1 VBX stent-graft as BSG between December 2017 and December 2019 were included in this single-center study. Demographics, comorbidities, and clinical data available at the time of the intervention were recorded from all patients in a prospective vascular database. Retrospective analysis of prospectively collected data was performed.
The study was approved by the local Ethics Committee (approval number 2020-764-f-S) and was performed in accordance with the Declaration of Helsinki. Patient consent was obtained prior to inclusion. The clinical follow-up included the findings of the most current clinical and radiological investigation either in our outpatient department or by an external medical institution. Causes of death were clarified by contacting the treating physician.
All patients underwent CTA of the aorta prior to discharge, after 6 months and yearly thereafter when glomerular filtration rate (GFR) >60 mL/min/1.73 m2. Patients with chronic or acute renal disease (GFR<60 mL/min/1.73 m2) underwent duplex ultrasound and noncontrast CT in the same intervals.
All measurements prior to bEVAR were performed by the operating vascular surgeon. In the framework of this study, the postoperative imaging studies were reanalyzed from an independent investigator with advanced experience in endovascular aortic surgery and a radiologist specialized in vascular imaging and endovascular treatment.
Materials
All procedures were performed with thoracoabdominal branched endografts based on the Zenith stent-graft platform (Cook Medical, Cook Inc, Bloomington, IN), including custom-made devices and off-the-shelf side-branched device (T-Branch, Cook Medical, Cook Inc, Bloomington, IN). The procedure and specific materials have been described elsewhere.2–5,9 The endografts were manufactured with caudally directed cuffs and connected with the target vessels using covered stents as bridging stent-graft. The most frequently used covered stent was the VBX. In case the VBX was not available in the desired size, an Advanta V12-iCast (Getinge Maquet, Rastatt, Germany) mated with a self-expanding Viabahn stent-graft (WL Gore and Associates, Flagstaff, AZ) was alternatively utilized.
Endpoints
The primary endpoint of the study was a composite of branch-related technical success, defined as successful introduction and deployment of the bridging stent-graft in the target vessel and freedom from target vessel instability. Target vessel instability was defined as any death or rupture related to BSG complications or any secondary intervention indicated to treat a branch-related complication, including endoleak, disconnection, kink, stenosis, occlusion, or rupture.
Secondary endpoints were clinical and ongoing clinical success. In accordance with the reporting standards of the Society for Vascular Surgery clinical success is defined as successful deployment and implantation of the aortic modular components and side branches, absence of death, aneurysm rupture, graft infection, conversion to open surgical repair, permanent paraplegia, disabling stroke and permanent dialysis. 13 Furthermore, the ongoing primary clinical success was evaluated, which is defined as clinical success with freedom from any unplanned secondary surgical or endovascular procedure targeting the aortic pathology, which was initially treated with the complex endovascular aortic repair.
If multiple items of composite endpoints applied to the same patient, they were counted as 1, for example, if a patient was treated for type Ia endoleak and had aneurysm expansion. Additionally, procedural, and postoperative clinical/radiological follow-up data were analyzed.
Statistical Analysis
Statistical data analysis was performed with SPSS Statistics (version 25, IBM Corp, Armonk, NY, USA). Categorical variables are expressed as frequency and percentage. The distribution of continuous variables was explored by Shapiro-Wilk test and variables with normal distribution are presented as mean ± SD. Variables with skewed distribution presented as median and interquartile range (IQR). Univariate analysis of categorical variables between subgroups was performed by Fisher’s exact test. Continuous variables were compared by the non-parametric Wilcoxon signed rank test. Analysis of endpoints was performed by Kaplan-Meier survival analysis and comparison between subgroups with log rank (Mantel-Cox) test. Differences were considered significant at p<0.05.
Results
During the study period, a total of 112 patients underwent endovascular treatment with VBX as bridging stent-graft in 355 target vessels (celiac trunk: 80, superior mesenteric artery: 81, renal arteries: 194). Procedural and postoperative details are shown in Tables 1 and 2.
Patient and Aneurysm Characteristics.
Abbreviations: AAA, abdominal aortic aneurysm; CMD, custom-made device; TAAA, thoracoabdominal aortic aneurysm.
Procedural and Postoperative Data.
Abbreviations: ICU, intensive care unit; IMC, intermediate care unit.
Eighteen patients (16.0%) experienced a total of 27 major adverse events whereof 3 patients suffered from a combination of respiratory insufficiency and spinal cord ischemia and 3 others from spinal cord ischemia only (Table 3). Four out of 6 patients with spinal cord ischemia were treated for a ruptured or symptomatic aneurysm.
Major Adverse Events: 24 Complications Occurred in 18 of 112 Patients.
The 30-day mortality-rate was 4.5% (n=5) and a total of 10 patients died before the 6-month follow-up. Eight of these 10 patients who died in the early phase were ruptured or highly symptomatic cases and had to be treated in emergency situations. Two of them were treated for mycotic thoracoabdominal aneurysms. Two others had severe comorbidities and died from cardiac or pulmonary reasons.
During a mean follow-up of 17.7±8.1 months, the mortality rate was 13.4% (n=15) and for estimation of survival please refer to Kaplan-Meier curve in Figure 1.
Kaplan-Meier estimation of survival. Overall survival rate at 18 months was 86.73% (95% CI: 80.11 to 93.34).
Efficacy and Safety Results of VBX
The primary endpoint was reached in 96.3% with a technical success of 100%. For cumulative freedom from BSG instability and reintervention see Kaplan-Meier estimation (Figure 2).
Kaplan-Meier estimation of cumulative freedom from target vessel instability of VBX stent-grafts in included patients. Freedom from primary endpoint events at 18 months was 95.83% (95% CI: 93.31 to 98.31).
The reasons for the 13 BSG instabilities were type Ic endoleaks (n=6; 1.7%), TV stenosis (n=2; 0.6%) and TV occlusion (5 occlusions; 1.4% in 4 patients). All endoleaks, occlusions, and stenoses were corrected by endovascular means. The reason for type Ic endoleak was an insufficient distal landing zone or insufficient dilatation of the stent at the level of the left renal artery in three cases, in celiac artery in 2 cases and in superior mesenteric artery in one. No type Ic endoleak was detected in the last 70 cases of our patient cohort.
The analysis of freedom from primary endpoint events between the first and second half of the patient population showed a significant correlation between early procedural experience with VBX and target vessel instability (Figure 3). Freedom from all primary endpoints in late versus early experience was estimated by Kaplan-Meier (p=0.05).
Kaplan-Meier estimation of cumulative freedom from target vessel instability in early vs late phase. Freedom from primary endpoint events at 18 months was 94.39% (95% CI: 90.78 to 97.94) in the early phase vs 98.84% (95% CI: 97.22 to 100) in the late phase.
With regard to the vessel patency, there was no significant difference in freedom from occlusions and stenosis of the target vessels between early and late experiences (p=0.38).
Clinical Outcome/Secondary Endpoint
The clinical success as secondary endpoint of our study amounted to 86.6% (15/112) at 18 months. As 13 patients underwent secondary interventions, 12 of them are still alive and 1 suffered from aneurysm sac expansion, the ongoing clinical success was 75.9%.
No aneurysm rupture or conversion to open repair occurred during follow-up. Graft infection or migration >10 mm was not detected by follow-up CT.
Two (1.7%) patients showed an increase (2 and 6 mm) of the aneurysm sac and in 110 (98.1%) the sac was stable (n=53, 48.2%) or decreased (n=57, 51.8%). The median decrease was 9.0±8.0 mm. The difference between preoperative (61±10 mm) and postoperative (57±16 mm) aneurysm sac maximum diameters was statistically significant with p<0.001.
Discussion
To our knowledge, this is the first study evaluating the performance of VBX as BSG in a large, homogeneous cohort of patients with TAAA treated with bEVAR.
In contrast to other reports, the clinical and technical performance of VBX were acceptable. 10 No TAAA ruptures were observed during the surveillance period. The technical success was 100% and the freedom from target vessel instability reached a favorable outcome of 96% at 18 months.
Analyzing in details the results of our study, several points should be addressed.
First, the marked reduction of bridging stent-graft-related complications and reinterventions during the last half of our experience reflecting our learning curve in the handling of this device.
No BSG-related type III endoleak was observed during the 18-month follow-up period. In contrast to experiences with other devices no loss of stent-graft integrity was observed and no fracture.14,15 Because of its flaring property, good visibility, and exact placement, this covered stent improves sealing at the level of the cuff reducing the risk of migration. Considering the vessels stented with VBX 6 type Ic endoleaks were found within the first 42 cases but no type Ic endoleak was detected in the last 70 cases. With increasing experience, we learned that careful intraprocedural estimation of BSG length, diameter, and oversizing is essential to avoid this complication.
Type Ic endoleaks were described by Tenorio et al 10 in their comparative SESG Viabahn vs BESG as one of the most prominent shortcomings of VBX. They obtained comparable results for self-expanding covered stents but not with VBX. However, their experience was short and VBX was used in target vessels at risk for type Ic endoleak. A closer look at their supplementary data revealed that 4 of the 9 affected target vessels had a smaller diameter than 6 mm and the distal landing zone was <15 mm in 7 out of 9 cases.
Our group learned how crucial it is to avoid such shortcomings and also to anticipate the adaptation of the very flexible stent to the curved course of the target vessel. In aneurysms with large inner aortic diameters and severe angulated TV the stent needs to accommodate a long distance between main graft and target artery and gets a more curved and swirled course. The degree of the curvature is obviously more pronounced using flexible stents and becomes more manifest after deflation of the catheter balloon during deployment.
For the choice of the BSG length we take now into consideration not only the nominal stent length but also the expected foreshortening after flaring at various diameters. We now deploy the VBX by dilating it to approximately 8 atm and then retract the deflated balloon of the VBX for 5 to 10 mm and inflate it again to nominal diameter pressure (11–14 atm). With this technique, we did not observe any case of the so-called “edge stenosis” at the distal level of the BSG due to over-dilating and thus causing intimal damage of the TV. Additionally, as aforementioned, no type 1c endoleaks were observed after we have modified our implantation technique. Taking this finding into consideration, we suggest choosing a longer distal landing zone (20 mm) and a slightly longer device as measured by graduated catheters.
As reported in other studies, the most bridging-related complications were observed in the renal branches, specifically in the left one.10,17
During respiratory movement the renal artery on the left side experiences a significant curvature change, which results in a greater deformation of the left renal artery as compared with the right renal artery. 16 This might be based on the anatomical asymmetric structure and the course/passage of the vena cava. According to their computational analysis the left renal artery showed a significant change of curvature and changes during breathing, while the curvature changes on the right renal arteries were not significant.
We observed 5 branch occlusions in the VBX cohort, whereof the left renal artery was affected in 3 cases. In one case with a bilateral renal involvement a heparin-induced thrombocytopenia was the leading cause. After mechanical thrombectomy renal perfusion and function could be restored. In our opinion the role of heparin bonding needs to be established in a large, multicenter study.
Although the present study represents the largest series reporting outcomes of VBX in bEVAR for TAAA to date, there are several limitations that need to be acknowledged. First of all, the retrospective nature of this study and the relatively short follow-up period. The study also contains self-reported data and no core laboratory or clinical events committee adjudicated endpoints or clinical events. Furthermore, as the focus of the study was on the performance of VBX no meaningful analysis could be made concerning the performance of other devices within the same approach. Moreover, the limited number of stent-graft occlusions does not allow us to draw robust conclusions concerning the role of heparin bonding in BSG patency.
Conclusion
VBX used as a bridging stent-graft in BEVAR appears to be safe and effective. However, one should take the learning curve into account to avoid type 1c endoleak and edge stenosis. Based on our experience, one should opt for the longer stent, longer distal landing zones, and perform the deployment in 2 steps, with low pressure within the target vessel and higher pressure inside the bridging stent.
Footnotes
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Gore Inc, the manufacturer of the VBX endografts, financially supported this clinical evaluation until a follow-up of 6 months without providing any procedural materials.
