Fatigue Resistance of the Advanta V12/iCast and Viabahn Balloon-Expandable Stent-Graft as Bridging Stents in Experimental Fenestrated Endografting

Introduction

The introduction of fenestrated endografts has enlarged the therapeutic options for treating patients with complex aortic pathologies.^1–3 Patients with multiple comorbidities or at high operative risk thus do not need to be excluded from treatment. The Achilles’ heel of this technique is the high rate of reinterventions^3,4 mostly due to occlusion, stenosis, endoleaks, fracture, or dislocation of the bridging stent. Durability issues have been mostly observed after use of single-layer polytertafluoroethylene (PTFE)-coated stent-grafts.^5,6 One of the most used bridging stents is the Advanta V12^7–9 (Getinge Deutschland GmbH, Rastatt, Germany).

Recently, the new Viabahn Balloon Expandable (VBX, W.L. Gore Inc, Flagstaff, AZ, USA) stent-graft has been introduced, with the potential to be used as bridging stent for fenestrated endografting.^10–13 In vitro studies have demonstrated an excellent impermeability and resistance to pull-out and shear forces in a fenestrated graft model.^9,14 However, fracture resistance to cyclic movement/loading over the time has not been evaluated so far. The aim of this study was to assess the morphological and radiological changes of 2 bridging stents under prolonged stress induction after overload with 50,000,000 cycles, simulating a life span of 75 months (6.25 years).

Materials and Methods

Endpoints and Procedure

Primary endpoint was to assess device safety, defined as absence of fractures detected on digital radiography as well as fabric failure with fragmentation and separation of the graft. Secondary endpoints were freedom from initial PTFE changes (FIC), and from PTFE breakpoint (FBP, all-layer defect).

Five 6×59 mm Viabahn Balloon Expandable stent-grafts (nonsterile, for in vitro use only) and five 6×59 mm Advanta V12 (sterile and for human use) were implanted in accordance with daily routine in clinical practice in a polyester sheet with nitinol wire enforced 6×6 mm fenestrations. In this manner we emulated stenting in fenestrated Cook Zenith endoprosthesis (Cook Medical, Bloomington, IN, USA).

After release, the stent-grafts were inflated again with a 10×20 mm PTA (percutaneous transluminal angioplasty) catheter (Powerflex, Cardinal Health, Dublin, OH, USA) for 30 seconds for the flaring process. All stent-grafts were implanted in the same way and without postdilatation with regard to the vessel side. Fatigue tests were conducted in ambient air at room temperature of 20°C to 26°C and humidity of 36% to 56%.

The fenestrated sheet with the implanted grafts was mounted in a frame. To create an in vitro setting simulating the in vivo situation of respiratory cyclic movement (15 cycles/min) the distal end of the stents was placed in silicone tubes (simulating the renal arteries as depicted in Supplementary Pictures a and c) and subjected to cranio-caudal angulations in a cyclic fashion with oscillating movements of ±15 mm by a dedicated testing machine. A controller was used to set the testing speed and count the performed cycles. The number of cycles was saved in the memory of the controller. The test setup is depicted in Figure 1 and Supplementary Picture b.

OPEN IN VIEWER

Figure 1.

Setup for cyclic movement fatigue testing on trial-related designed test bench, consisting of test machines (A) and their respective controller (B). A microscope camera (C) with a resolution of 5 megapixels was used for image recording.

Results

With regard to the primary endpoint, no fracture or separation of the stents was detected respectively on radiography and digital imaging after a testing period of 50,000,000 cycles. Digital radiographs showed minor changes in the endoskeleton of both graft brands. They were caused by the initial flaring process.

Regarding FIC, there was no significant difference between both stent types (3,400,000 cycles in the VBX vs 3,200,000 in the Advanta group). Slight wear of the stent covering was found at the peak of the stent strut (Figure 2A) or where the friction of the endoskeleton of the covered stent against the fenestration ring was particularly high (Figure 2B).

OPEN IN VIEWER

Figure 2.

(A) Initial polytetrafluoroethylene (PTFE) changes on digital microscope image of Viabahn ballon-expandable endograft after 3.400.000 cycles—outer surface on aortic side. (B) Initial PTFE changes on digital microscope image of Advanta V12 after 3,200,000 cycles—outer aortic end.

No material defects involving all PTFE layers (FPB) were observed in the tested VBX stents while the Advanta V12 revealed early defects of PTFE integrity in 3 of 5 tested stents. In 2 cases, the defects were due to early penetration of the nitinol ring through the PTFE fabric (Figure 3A). The Advanta V12 graft perforations evolved into large defects after an average of 8,500,000 cycles (Figure 3B). Accordingly, the FPB of VBX was significantly higher compared with Advanta (50,000,000 vs 33,400,000; p<0.01) as freedom from all-layer defects were maintained in the VBX samples through the completion of the full extent of the evaluation.

OPEN IN VIEWER

Figure 3.

(A) Penetration of Advanta V12 by the nitinol ring of the fenestration after 1,000,000 cycles causing a defect of the polytetrafluoroethylene (PTFE) coverage—inner lumen. (B) Image of penetrated Advanta V12 with all-layer defect on aortic side—taken after 40,000,000 cycles.

The intraclass correlation coefficient for FIC was 0.961 (95% CI 0.886 to 0.989) and for FPB = 1 revealing consistent interobserver agreement.

Signs of incipient brownish discoloration became visible on the microscopic images after a high cycle number on both devices (Figure 4A). These changes were located at the level of the uncovered stent struts in the Advanta group (Figure 4B).

OPEN IN VIEWER

Figure 4.

(A) Microscope image of VBX showing obvious signs of brownish discoloration on outer portion of the aortic side after 40,000,000 cycles. (B) Microscope image of Advanta V12 showing obvious signs of brownish discoloration on outer portion of the aortic side after 40,000,000 cycles.

Discussion

An important reason for reinterventions after fenestrated and branched endovascular aneurysm repair is bridging stent instability. ⁴ The stents are exposed to different stress factors like cardiac cycles, respiratory movements of the target vessels, postoperative angulation and diameter changes of the aorta as well as remodeling of the aortic endograft inside the aneurysm during the follow-up. The aim of this study was to understand the response of fabric and endoskeleton of two stent used for this indication to cyclic loading over a long time period.

In the presented fatigue test, VBX and Advanta V12 performed equally well in terms of fracture resistance and freedom from initial PTFE changes after stress induction of 50,000,000 cyclic movements.

However, microscopy showed all-layer defect in 3 of 5 Advanta stents at the end of 50,000,000 cycles. In 2 of 5 Advanta stents the defects were due to early penetration of the nitinol ring through the PTFE coverage at the level of the fenestration. The reason for this unexpected outcome is not clear. The Advanta stent-graft has a more rigid stent design. The rigidity of the stent may have caused a displacement of the nitinol wire of the fenestration during flaring. After flaring the open end of the nitinol wire might have punctured the PTFE layer causing defects of the fabric. For this specific application, a stent design combining a highly rigid endoskeleton with ultrathin PTFE coating should be further discussed.

Since the underlying mechanism of early PTFE penetration exclusively into the Advanta endografts could not be clarified in these in vitro tests, further investigations should be conducted. In particular, changes of flared covered stents implanted in fenestrated aortic endografts of other manufacturer and with a thicker PTFE fabric should be evaluated.

As the PTFE defects were detected in the flared portion of the stent and therefore at the transition from graft to the target vessel, this flaw may have a clinical impact on stability of the bridging stent during the follow-up.

Regarding wear and discoloration, also these changes could be documented in the region around the penetrated area near the fenestration. These changes could suggest initial rust. Our lay understanding is that stainless steel should not rust at all. However, according to the name, steel is “stainless” and not “stain free.”

Knowledge of the material science beneath is critical to correctly interpret this finding. Stainless steel is an alloy commonly used for manufacturing medical devices. Steel is a product of carbon and iron. To prevent it from rusting, 316L stainless steels contain chromium, nickel, and molybdenum. Chromium has a strong affinity for oxygen. When the surface of stainless steel is exposed to oxygen, it forms chromium oxide (Cr₂O₃). If the oxide layer wades off or is removed by scratch, the steel exposed may begin to form iron oxide (rust) under certain conditions. The process of corrosion can be accelerated among others by chloride, by direct contact between 2 different metals with a common electrolyte, or by external stress.^16,17 However, exposition to chloride or humidity of the environment did not play a role in our experimental setting.

In this study, the stents were exposed to extreme load for a long time. However, no radiological signs of metal erosion were detected by digital radiography. Nevertheless, these changes should be further evaluated after larger amount of cycles.

This report has several limitations. With this fatigue test we try to emulate the consequences of the breathing motions for stents implanted in the renal arteries. However, the in vivo conditions could only partially be replicated. The biomechanical forces on stent-grafts implanted in the human body are also altered by intraluminal pressure due to the cardiac cycle and by the environmental temperature and fluid composition. Therefore, as the results of this study can only partially estimate what occurs after using stents in humans, further fatigue tests should emulate pulsatile flow of blood simulating fluids at body temperature. A further limitation of this study is that other aortic stent-grafts like Anaconda or the Jotec custom made devices were not tested. Furthermore, only 6×6 mm fenestrations were used.

Conclusion

After a life span of 75 months, cyclic fatigue tests emulating breathing motion of the kidney show no fracture or fragmentation of VBX and Advanta V12 bridging stents. However, penetration of the nitinol ring wire caused all-layer defects of PTFE after an average of 33,400,000 cycles in Advanta V12, while the VBX PTFE remained free from all-layer damage after 50,000,000 cycles. Initial signs of a brownish discoloration were observed in both stent types without consequences for the endoskeleton. A longer follow-up of these changes is required and should be further evaluated after continued stress induction. As in vivo conditions are only partially emulated, future fatigue tests should be performed in a pulsatile flow model and with fluids simulating temperature and components of the human blood.

Supplemental Material

Supplemental Material

Supplemental Material