Abstract
Keywords
Introduction
The use of fenestrated and branched endovascular aortic repair (F/BEVAR) devices for the treatment of thoracoabdominal aortic aneurysms (TAAA) is increasing worldwide, and recent studies reported encouraging perioperative and midterm results.1-5 However, the stent-graft designs, the procedural steps, and the materials employed are not standardized, and most decisions are left to physician preference and personal experience. Moreover, most of the implantable devices employed are not approved specifically for these procedures but are used off-label as part of clinical studies or investigational device exemption protocols.
A critical step in these F/BEVAR procedures is the placement of covered stents in renal and splanchnic vessels to mate the main stent-graft with the target vessels. No balloon-expandable or self-expanding covered stent has been specifically approved for use as a mating stent with F/BEVAR. About 2 years ago, a novel balloon-expandable covered stent (LifeStream; Bard Peripheral Vascular, Tempe, AZ, USA) was released on the market with a different design with respect to the most widely used bridging covered stent, the Atrium Advanta V12 (Maquet Getinge Group, Mijdrecht, the Netherlands). The aim of this study was to evaluate the perioperative and 1-year outcomes of the LifeStream covered stent as the bridging stent in F/BEVAR for TAAA treatment.
Methods
Study Design and Patient Sample
Between March 2015 and January 2017, 18 nonconsecutive, high-risk surgical candidates (median age 74.7 years; 14 men) with TAAA (Table 1) were treated using multistage F/BEVAR. Data were retrieved from the medical records for analysis. Results are reported in accordance with current reporting standards. 6 The study complied with the Declaration of Helsinki, and all subjects gave informed consent to the aortic procedure. According to national privacy laws, there was no need for ethical approval for a retrospective review of anonymized data.
Characteristics of the 18 Patients in the Study, Preoperative Aneurysm Morphology, and Stent-Graft Design. a
Abbreviations: AAA, abdominal aortic aneurysm; ASA, American Society of Anesthesiologists; BMI, body mass index; DTA, descending thoracic aorta; EVAR, endovascular aneurysm repair; SVS, Society for Vascular Surgery; TAAA, thoracoabdominal aortic aneurysm; TEVAR, thoracic endovascular aortic repair.
Continuous data are presented as the medians (first, third quartiles); categorical data are given as the counts.
According to Chaikof EL, Fillinger MF, Matsumura JS, et al. Identifying and grading factors that modify the outcome of endovascular aortic aneurysm repair. J Vasc Surg. 2002;35:1061-1066.
According to Fillinger MF et al. 6
The preoperative computed tomography angiography (CTA) scans were analyzed on an Aquarius workstation (TeraRecon, Foster City, CA, USA) in order to design the custom-made or off-the-shelf F/BEVAR devices (Cook Medical, Bloomington, IN, USA) on the basis of individual anatomy. The treatment plan called for the use of the LifeStream balloon-expandable covered stent in all fenestrated designs in these patients and selectively in the branched designs according to physician preferences. Fluency self-expanding stent-grafts (Bard Peripheral Vascular, Tempe, AZ, USA) were employed in all the branched designs not mated with the LifeStream covered stent.
Operative Technique
The bridging stent was sized according to the diameter of the target vessel (no oversizing) and the length from the fenestration/branch to within the target vessel so as to obtain a minimum 15-mm distal landing zone. All the procedures were performed in an angiographic suite equipped with a Philips Allura Xper FD 10 single-plane C-arm (Philips Medical, Best, the Netherlands). Unfractionated heparin was administered intravenously before and during the procedure to maintain an activated clotting time >250 seconds. After main graft deployment, the LifeStreams were advanced through a compatible Flexor sheath (Cook Medical) and deployed in the intended location, either protruding 5 to 10 mm within the main graft in fenestrated designs or overlapping into the target branch (18–21 mm). Proximal stent flaring was performed in all fenestrations with a 10×20-mm noncompliant balloon (Armada 35; Abbott Vascular, Santa Clara, CA, USA). In branched cases, proximal dilation was performed only if there was a diameter mismatch between the covered stent and the branch. In these instances, a 6- or 8-mm noncompliant Armada 35 balloon was used.
Patients were discharged on double antiplatelet therapy with aspirin and an adenosine diphosphate receptor antagonist for the first year. Following the predischarge CT, all patients were evaluated at 6 and 12 months and yearly thereafter. If a type I or III endoleak was seen on the predischarge CT scan, another scan was scheduled for 2 months after the index procedure.
Results
Delivery and deployment of the 43 LifeStream covered stents at the desired location was successful in all 32 fenestrations and 11 branches of the 18 F/BEVAR devices. Twenty-two Fluency bridging stents were used in other branches (total 65 target vessels). The LifeStream covered stent was implanted in 24 renal arteries, 11 celiac trunks, and 8 superior mesenteric arteries. Details of the procedures are presented in Table 2.
Intraoperative and 30-Day Outcomes. a
Continuous data are presented as the medians (first, third quartiles); categorical data are given as the counts.
Acute Kidney Injury Network criteria.
Median procedure time was 340 minutes with a median 98 minutes of fluoroscopy. The technical success of the entire repair was 83% owing to sac perfusion demonstrated on the intraoperative completion angiogram in 3 cases. The source was not identifiable, and the endoleaks were left untreated in order to reassess them in a second step. On the predischarge CT, another 2 patients had endoleaks. Further investigation determined that all 5 patients with intraoperative or predischarge endoleaks had type IIIc perifenestration endoleaks secondary to inadequate sealing of the LifeStream stent-graft in 7 (22%) of 32 fenestrations (Figure 1, Table 3). None of the target vessels with endoleaks had received previous stenting, and no type IIIc endoleak was recorded in the branched cases.
(A) Intraoperative placement and deployment of an 8-mm LifeStream covered stent through a fenestration into the superior mesenteric artery. Note that the stent protruded properly within the stent-graft and was flared with a 10-mm angioplasty balloon. (B) Postoperative computed tomography (CT) highlighted a type IIIc endoleak from the superior mesenteric artery with aneurysm perfusion. (C) Intraoperative placement and deployment of an 8-mm balloon-expandable bare stent within the previous covered stent and subsequent reflaring. (D) Postoperative CT observed complete resolution of the endoleak and thrombosis of the aneurysm sac.
Distribution of the Type IIIc Endoleaks.
All 5 patients received a second CT scan at 2 months, which confirmed persistence of the endoleaks. Four patients underwent a secondary endovascular reintervention under local anesthesia. A 10×20-mm balloon was inflated at the level of the fenestration while aortography was performed from the contralateral groin to confirm the type IIIc endoleak. In 3 of the 4 cases, contrast-enhanced ultrasound imaging was performed while the balloon was inflated to confirm the findings and rule out other endoleak sources. The IIIc endoleaks were sealed in all 4 cases with proximal relining using a bare balloon-expandable stent at the perifenestration transition area [median procedure duration 46 minutes, interquartile range (IQR) 40, 57]. The endoleaks were completely resolved in all 4 cases at 6 and 12 months from the secondary procedure. One patient refused intervention for the type IIIc endoleak; the patient is alive at 2 years with a 5-mm growth in the aneurysm sac.
In the last 4 fenestrated repairs in this series, prophylactic perifenestration relining was performed with no evidence of type IIIc endoleaks at CT imaging. Retrospective analysis of the predischarge CT scans of all cases in which a LifeStream covered stent was deployed in a fenestration without prophylactic reinforcement (26 of 32 fenestrations) was done to measure the distance between the fenestration and the target vessel. The incidence of type IIIc endoleak was higher in the cases in which the fenestration was not against the aortic wall (58% vs 0%; p=0.008).
Over a median follow-up of 14.1 months (IQR 11, 22), 2 patients died within 6 months and another 2 within a year, all with patent target vessels at the last imaging session. Of the 16 patients evaluated at 6 months, all 61 target vessels (42 LifeStreams) were patent. At 1 year in 12 patients evaluated (2 lost to follow-up), all 47 target vessels (38 LifeStreams) were patent.
Discussion
Current F/BEVAR devices rely on a durable connection between the main aortic stent-graft and the target vessels. To mate the main graft with the target vessels, 2 systems have been proposed, one involving directional branches and one with reinforced fenestrations, both to be bridged with covered stents. While the sealing between a branch and its covered stent is based on their overlap, the sealing within fenestrations is achieved by flaring the proximal end of the covered stents inside the aortic stent-graft. For this reason, the branches can be combined with either a self-expanding or a balloon-expandable covered stent, while the fenestrations exclusively require balloon-expandable covered stents.
At present worldwide, the most commonly used covered stent for F/BEVAR has been the iCast/Advanta V12, and it seems to be performing well, with low rates of occlusion, dislodgement, migration, fracture, and tears.7,8 Different articles reported high patency rates for this covered stent.7,8 The only factor affecting renal target vessel patency was the design of the main graft. 9 Independent of the covered stent employed, the patency rate was lower when branched designs were used to target the renal arteries.9,10 The rate of modular type IIIc endoleaks reported in the literature ranges from 1.7% to 3.3% in reinforced fenestration designs.7,9 Interestingly, perioperative type IIIc endoleaks are described only as a midterm adverse event secondary to late implant failures or to material fatigue.7,9
Although the performance of the iCast/Advanta V12 covered stent with Cook devices is considered durable,7-10 different limitations have been observed. First, there is no iCast/Advanta V12 stent available that measures ~30-mm long, forcing operators to use the short 22-mm stent when employing fenestrated designs, as the 38-mm stent is often too long for target vessels with a proximal origin of collaterals or bifurcations. Second, this balloon-expandable covered stent is rigid and does not conform easily to vessel tortuosity or changes induced by the aneurysm. 11 For these reasons, different studies reported extensive use of distal relining with bare self-expanding stents in order to aid the transition between the rigid covered stent and the target vessel to prevent kinking.7,8 Third, most of the available iCast/Advanta V12 stent sizes are compatible with a 7-F sheath (Table 4).
Main Characteristics of Commercially Available, 0.035-Inch–Compatible, Balloon-Expandable Covered Stents.
Abbreviations: PTFE, polytetrafluoroethylene.
Maquet Getinge Group, Mijdrecht, the Netherlands.
Bentley Innomed GmbH, Hechingen, Germany.
Bard Peripheral Vascular, Tempe, AZ, USA.
W.L. Gore & Associates Inc, Flagstaff, AZ, USA.
The 7-F requirement of this covered stent may be a limiting factor for different reasons. First, the need for a bigger sheath might complicate vessel cannulation, especially in small renal arteries. Second, the bigger the sheaths required for each target vessel the larger the contralateral femoral sheath. In our experience, a 16-F introducer sheath is required to accommodate two 6-F sheaths and an 18-F sheath is required for two 7-F sheaths. This aspect might play a role in spinal cord ischemia prevention because the contralateral sheath can impair pelvic and lower limb circulation. 2 Finally, new preloaded custom-made devices have been designed for use with 6-F sheaths, and at present, the only 6-F–compatible size of the iCast/Advanta V12 covered stent is the 6×22 mm (Figure 2). 12 These limitations support the use of the more recent covered stents released on the market (Table 4).
Sheath compatibility (green = 6-F, orange = 7-F, blue = 8-F), according to the diameter and length of commercially available, 0.035-inch–compatible, balloon-expandable covered stents. V12, Advanta V12 (Maquet Getinge Group, Mijdrecht, the Netherlands); BE, BeGraft peripheral (Bentley Innomed GmbH, Hechingen, Germany); LIF, LifeStream (Bard Peripheral Vascular, Tempe, AZ, USA); VBX, Viabahn VBX (W.L. Gore & Associates Inc, Flagstaff, AZ, USA).
In our single-center experience, the Advanta V12 covered stent has been used extensively, and the new BeGraft covered stent (Bentley Innomed GmbH, Hechingen, Germany) is being investigated. The LifeStream covered stent came into use with F/BEVAR in our practice when the Advanta V12 became temporarily unavailable in 2015. Though the follow-up period in our small cohort was limited to 1 year, the patency rate was 100%. However, the occurrence of type IIIc endoleaks with the fenestrated designs was concerning. While the LifeStream covered stent was flared with the same method used for the iCast/Advanta V12, and correct flaring was confirmed on the predischarge CT scan, perifenestration sealing was still inadequate.
According to our experience during the index procedures and the relining solution adopted during reinterventions, the LifeStream covered stent seems to recoil excessively at the peri-fenestration level, possibly due to a lack of sufficient radial force to obtain/maintain an effective seal after flaring within the reinforced fenestration of custom-made Cook devices. Different causative mechanisms can be postulated. LifeStream and iCAST/Advanta V12 covered stents are both fabricated with 316L stainless steel, but they have different stent designs. The LifeStream is composed of W-shaped stents, which might explain the different behavior. Further studies and bench tests would be needed to explain the mechanisms involved and to guide possible changes in stent design. Moreover, further studies should address whether the reported behavior of this covered stent is to be expected with other fenestrated devices (ie, Anaconda, Bolton, and Jotec).
It remains unclear why a type IIIc endoleak was not evident in all fenestrated implants, but presumably the limited room present between the fenestration and the aortic wall could prevent formation or detection of a type IIIc endoleak. The analysis of the preoperative CT scans of all cases in which a LifeStream covered stent was deployed in a fenestration without preventive reinforcement found that type IIIc endoleaks were more frequent in those cases in which there was room between the ostium of the target vessel and the fenestration, thus corroborating our hypothesis.
Because of this preliminary experience, the LifeStream covered stent is no longer employed in fenestrations in our practice; it is used only when branched designs are required. In the future, other covered stents are going to be released on the market, but since most of them will not be specifically tested for F/BEVAR, any off-label application will require a careful initial evaluation. In the forthcoming years, it is hoped that companies producing covered stents will design bridging components specifically dedicated to fenestrated/branched devices.
Conclusion
This single-center preliminary experience with a novel balloon-expandable covered stent used in F/BEVAR for TAAA repair demonstrated good patency; however, an unexpectedly high rate of type IIIc endoleaks was observed. These endoleaks were resolved with reintervention or during the index procedure by proximal relining with a bare balloon-expandable stent, achieving adequate perifenestration sealing. The LifeStream covered stent is no longer deployed in fenestrations in our practice.
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) received no financial support for the research, authorship, and/or publication of this article.
