One-year pivotal trial outcomes of the Nellix system for
endovascular aneurysm sealing
Jeffrey P. Carpenter, MD,aRobert Cuff, MD,bClifford Buckley, MD,cChristopher Healey, MD,d
Sajjad Hussain, MD,eMichel M. P. J. Reijnen, MD,fJose Trani, MD,aandDittmar Böckler, MD,gfor the Nellix Investigators,*Camden, NJ; Grand Rapids, Mich; Temple, Tex; Portland, Me; Indianapolis, Ind; Arnhem, The Netherlands; and Heidelberg, Germany
ABSTRACT
Objective: The Nellix EndoVascular Aneurysm Sealing (EVAS) System (Endologix, Inc, Irvine, Calif) is a novel approach to abdominal aortic aneurysm (AAA) treatment whereby polymer is used tofill the AAA sac. We report 1-year results of the investigational device exemption pivotal trial.
Methods: Eligible patients were treated at 30 sites in the United States and Europe. Inclusion criteria required an asymptomatic infrarenal AAA, with a neck length$10 mm and #60 angle, iliac artery blood lumen diameter 9 to 35 mm, access artery diameter$6 mm, and serum creatinine #2 mg/dL. Follow-up included computed tomography angiography scans at 30 days, 6 months, and 1 year that were evaluated by a core laboratory. The primary safety end point was 30-day major adverse events (MAEs), which were compared with a performance goal of<56% (the Society for Vascular Surgery open repair control group rate). The primary effectiveness end point was treatment success at 1 year, which was compared with a performance goal of>80%. Treatment success required procedural technical success and absence of AAA rupture during follow-up, conversion to open surgical repair, endoleak (type I or III) at 1 year, migration >10 mm causing complications or requiring secondary intervention, aneurysm enlargement, or secondary procedures through 1 year for resolution of endoleak, device obstruction or occlusion, or device defect.
Results: Of 150 treated patients, 149 (99.3%) completed 1-year follow-up. The MAEs rate at 30 days was 2.7% (95% con-fidence interval, 0.7%-6.7%), satisfying the primary safety end point (<56%). The 1-year treatment success was 94% (95% confidence interval, 88.6%-97.4%), achieving the primary effectiveness end point (>80%). At 1 year, key secondary outcomes included 6.7% MAEs, 4.7% serious device-related events, 1.3% AAA-related mortality, 3.7% secondary in-terventions, and 0.7% surgical conversions. MAEs through 1 year included death (n¼ 6), stroke (n ¼ 3), bowel ischemia (n¼ 2), renal failure (n ¼ 2), respiratory failure (n ¼ 2), and myocardial infarction (n ¼1). One iatrogenic AAA rupture occurred and one AAA rupture was reported during follow-up. AAA sac enlargement (>5 mm) was 1.5% at 1 year. Endoleaks were present in four patients (3.1%) at 1 year (1 type Ib and 3 type II). Migration>10 mm occurred in three patients (2.3%), but none required secondary intervention.
Conclusions: Outcomes with this novel endovascular therapy for AAA, the Nellix EVAS System, are encouraging. The primary safety and effectiveness end points have been met. Low morbidity, low mortality, and high procedural and treatment success were achieved despite the inevitability of a learning curve and unique risks associated with a new device and technique. Long-term follow-up is in progress. (J Vasc Surg 2017;65:330-6.)
The Nellix EndoVascular Aneurysm Sealing (EVAS) Sys-tem (Endologix, Inc, Irvine, Calif) offers a novel approach to the treatment of abdominal aortic aneurysm (AAA). In contrast to traditional prostheses for endovascular
aneurysm repair (EVAR), aneurysm sealing is achieved by means of polymer-filled endobags attached to covered stents that span the aortoiliac segment. The polymer-filled endobags function to seal the entire
From the Department of Surgery, Cooper Medical School of Rowan University, Camdena; the Division of Vascular Surgery, Spectrum Health,
Grand Rapidsb; the Department of Surgery, Scott and White Hospital,
Templec; the Department of Vascular Surgery, Maine Medical Center,
Portlandd; the Department of Vascular Surgery, St. Vincent’s Hospital,
Indianapolise; the Department of Surgery, Rijnstate Hospital, Arnhemf; and
the Department of Vascular Surgery, Heidelberg University Hospital, Heidelberg.g
*A complete listing of the Nellix Investigators is provided inSupplementary Table I
(online atwww.jvascsurg.org).
This work was funded by Endologix. Endologix had no involvement in the collection, analysis, and interpretation of data or the decision to submit the manuscript for publication.
Author conflict of interest: All authors have received research funding from Endologix for the performance of this study. J.P.C., M.M.P.J.R., and D.B. have received consulting fees from Endologix.
Presented at the 2016 Vascular Annual Meeting of the Society for Vascular Surgery, National Harbor, Md, June 8-11, 2016.
Additional material for this article may be found online atwww.jvascsurg.org. Correspondence: Jeffrey P. Carpenter, MD, 3 Cooper Plaza, Ste 411, Camden,
NJ 08103 (e-mail:carpenter-jeffrey@cooperhealth.edu).
The editors and reviewers of this article have no relevantfinancial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest.
0741-5214
CopyrightÓ 2016 by the Society for Vascular Surgery. Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.jvs.2016.09.024
aneurysmal segment and to distributefixation across the entire segment. This technique may address all potential sources (I, II, III, IV) of endoleak directly.
Normal and abnormal imaging appearances after stan-dard EVAS are unique and described in a consensus document based on the collective experience of>1000 clinical cases.1Results of multiple observational cohorts outside the United States with the commercially approved Nellix device have been reported and describe excellent real-world outcomes, with low mortality and very low rates of endoleaks and secondary procedures.2 We also recently reported that the 30-day safety end point of the investigational device exemption (IDE) trial of the Nellix device for EVAS was successfully achieved.3 We now report 1-year results of the IDE trial for safety and efficacy.
METHODS
Nellix system and EVAS procedure. The Nellix EVAS System and its method of implantation have been recently described in detail.2,3 In brief, after the pa-tient is anesthetized and given heparin for anti-coagulation, arteriography is performed to establish the specific device length needed on each side to ensure full coverage of the Nellix device from the nonaneurysmal infrarenal segment to the non-aneurysmal common iliac segments. The Nellix 17F outer-diameter catheter is introduced via each femoral artery. Each device is positioned at the lowest renal artery inferior margin and terminates in the common iliac artery. After the Nellix catheters are positioned below the most caudal renal artery, the outer sheaths are retracted and the catheter connectors are attached to the console connectors. The stent balloons are simultaneously inflated with contrast-enhanced saline to deploy the covered stents.
The endobags are prefilled with saline with pressure monitoring to determine the volume of polymer required to achieve adequate pressure and thus seal the aneurysm. The saline is withdrawn, and the volume aspirated is measured to determine the volume of poly-mer to be used. The polypoly-mer is injected into the endo-bags, and careful pressure manometry is used tofill the aneurysmal lumen to the target volume and pressure, thereby excluding and sealing the entire aneurysm. Angiography is performed to confirm stent patency and absence of endoleaks. The delivery systems are then detached from the implant and removed.
Clinical trial design. A prospective, multicenter, single-arm clinical study of the Nellix EVAS System was con-ducted at 29 sites under an IDE approved by the United States Food and Drug Administration. Local Institutional Review Board or Ethics Committee approval was obtained before patient enrollment at all sites (Supplementary Table I, online only).
After written informed consent was obtained, each po-tential participant underwent screening by the local site and an imaging core laboratory (Cleveland Clinic Periph-eral Vascular Laboratory, Cleveland, Ohio) as well as an in-dependent medical reviewer. High-resolution computed tomography angiography (CTA) scanning was required #3 months of screening. In addition to anatomical eligi-bility, each patient underwent a physical examination, a medical history review, and selected blood laboratory analyses. Inclusion and exclusion criteria are summarized inTable I. Patients were re-evaluated at 1 month, 6 months, and 1 year with clinical evaluation and CTA scanning. All patients will subsequently be evaluated annually through 5 years of follow-up. The first patient at each site was designated a“roll-in” patient. Subsequently enrolled pa-tients comprised the 150-patient trial cohort.
The primary safety end point definition is the incidence of major adverse events (MAEs) at 30 days as adjudicated by an independent Clinical Events Committee (CEC). MAEs included all-cause death, bowel ischemia, myocardial infarction, paraplegia, renal failure, respiratory failure, stroke, and procedural blood loss $1 L. Serious adverse events (SAEs) are defined as any adverse events resulting in death or serious deterioration in the health of the patient that resulted in life-threatening illness or injury, a permanent impairment, or required hospitalization or prolongation of an existing hospitalization. An independent Data Safety Monitoring Committee periodically reviewed safety data.
The primary effectiveness end point definition is the rate of treatment success at 1 year. Treatment success is defined as procedural technical success (intended deployment without unintentional coverage of both in-ternal iliac arteries or any visceral aortic branches and successful removal of the delivery system) and freedom from AAA rupture during follow-up, open conversion, type I or III endoleak, sac enlargement, or intervention for migration or lumen stenosis/occlusion at 1 year.
Statistical analyses. The analyses were performed on an intention-to-treat basis. The primary safety hypothe-sis was that the percentage of patients experiencing one or more MAEs through 30 days would be less than a performance goal of 56% (Society for Vascular Surgery [SVS] open surgery group),4based on a significance level of
a
¼ .025, using a one-tailed exact test. Effectiveness was tested using the exact binomial distribution based on the hypothesis that the treatment success rate at 1 year exceeds a performance goal of 80% at a signifi-cance level ofa
¼ .05. Power analysis assumptions have been previously reported.3Analyses were performed with SAS 9.2 software (SAS Institute Inc, Cary, NC).RESULTS
Compliance with follow-up
A total of 29 roll-in patients (1 per site) were enrolled for training purposes and were analyzed separately
(Supplementary Table II, online only), and 150 patients were enrolled in the primary arm of the study. Of these patients, 144 (96%) were eligible to return for the 1-year follow-up visit and 143 (99.3%) returned. Five patients died before the 1-year visit was due (see details of mortal-ity below). One additional patient withdrew consent before the 1-year follow-up. The core laboratory reviewed 100% of the images submitted by each site at each visit period.
Baseline patient characteristics
The study population reflects the general AAA popula-tion, which consists of elderly adults with multiple surgi-cal risk factors. The average age was 73 years. Eight women were enrolled. The original version of the IDE study protocol only allowed subjects with a sac diameter of $5.5 cm. Because this criterion excluded most women, the protocol was amended to allow AAA with a sac diameter of$5 cm, or $4.5 cm if it had increased by$0.5 cm within the last 6 months, or which exceeded 1.5 times the transverse dimension of an adjacent
nonaneurysmal aortic segment. The average maximum sac diameter (long axis) was 58 mm (range, 44-82 mm). The American Society of Anesthesiologists Physical Sta-tus Classification for most patients was III or IV, indicative of extensive comorbidity (Table II).
Procedural characteristics
General anesthesia was used in 56%. Percutaneous ac-cess was used for 55% of the procedures. Mean polymer fill volume was 75 mL. Fluoroscopy time averaged 10 mi-nutes. The average procedure time was 88 minutes (range, 50-205 minutes). Procedural technical success was 100%. Hospital length of stay averaged 1.1 days.
Safety outcomes at 30 days
The MAEs rate at 30 days was 2.7% (95% confidence in-terval [CI], 0.7%-6.7%), which was significantly lower than the SVS performance goal of 56% (P< .0001). CEC adju-dication identified 4 patients (2.4%) with 6 MAEs by 30 days, including renal failure in 2, respiratory failure in 2, multiorgan failure in 1, and bowel ischemia in 1. Table I. Trial inclusion and exclusion criteria3
Inclusion criteria
1. Male or female at least 18 years old
2. Informed consent signed and agrees to all follow-up visits
3. AAA>5.0 cm, or >4.5 cm which has increased by >0.5 cm within the last 6 months, or which exceeds 1.5 times the transverse dimension of an adjacent nonaneurysmal aortic segment
4. Anatomically eligible for the Nellix System per instructions for use: a. Adequate iliac/femoral access (diameter$6 mm)
b. Aneurysm blood lumen diameter#60 mm c. Proximal non-aneurysmal aortic neck:
i. Length$10 mm
ii. Lumen diameter 18-32 mm iii. Angle#60to the aneurysm sac
d. Most caudal renal artery to each hypogastric artery length$100 mm e. Common iliac artery lumen diameter 9-35 mm
f. Ability to preserve at least one hypogastric artery Exclusion criteria
1. Life expectancy<2 years
2. Psychiatric or other condition that may interfere with the study 3. Participating in another clinical study
4. Known allergy or contraindication to any device material 5. Coagulopathy or uncontrolled bleeding disorder 6. Ruptured, leaking or infected aneurysm 7. Serum creatinine level>2.0 mg/dL
8. CVA or MI#3 months of enrollment/treatment 9. Aneurysmal disease of the descending thoracic aorta
10. Clinically significant mural thrombus within the proximal landing zone (minimum 10 mm) of the infrarenal non-aneurysmal neck (>5 mm thickness over >50% circumference)
11. Connective tissue diseases
12. Unsuitable vascular anatomy that may interfere with device introduction or deployment 13. Pregnant
The 30-day MAEs incidence was 1.4% (95% CI, 0.2%-5.0%) in men (n ¼ 142) and 25.0% (95% CI, 3.2%-65.1%) in women (n¼ 8). Complete details of the 30-day safety re-sults are reported elsewhere.3
Device-related SAEs
Device-related SAEs occurred#30 days of the proced-ure in two patients (1.3%). In thefirst patient, a renal ar-tery was inadvertently covered by a Nellix device. After the procedure, repeat review of the final angiogram identified delayed filling of the renal artery. The patient was returned to the operating room for attempted renal stenting, but recanalization could not be achieved. As a
result, the patient experienced renal failure, although he-modialysis was not required. The second patient sus-tained intraoperative iatrogenic AAA rupture during the prefill step. Upon evacuating the prefill from the endo-bags, the patient’s blood pressure was noted to be drop-ping as a result of presumed disruption of the aneurysm after pressurization of the sac during the prefill. The pro-cedure was completed uneventfully by injecting polymer into the endobags as planned, achieving seal with no ev-idence of endoleak. Although retroperitoneal hematoma was confirmed on the postoperative CT, no additional complications occurred and the patient required no additional treatments.
Device-related SAEs occurring>30 days were noted in five patients (3.4%). Limb occlusions were seen in two pa-tients (1.3%): one presented on day 285 and the otherfirst presented on day 41 with subsequent limb occlusion on days 195 and 244. No particular etiology was found for the two limb occlusions, and no stent fracture or kinking was reported by the site or by the core laboratory for these two individuals.
A type Ia endoleak was noted in one patient on a 30-day CT scan. This was treated with coils. The patient pre-sented 7 months after the procedure with abdominal and back pain. CTA revealed a type Ia endoleak and peri-aortic stranding. The leak was retreated with coils and the addition of glue, without resolution. The device was explanted, and a contained aneurysm rupture was iden-tified during the open conversion. The patient failed to thrive and died 1 month later in hospice care.
A type Ia endoleak was noted in another patient on the 6-month follow-up CTA. This resolved after being treated with coil and glue embolization and continues to be free of endoleak through the 1-year follow-up.
Thefinal patient was admitted to the hospital 9 months postprocedure for multiple infections and rectal bleeding. A device-related MAE, bowel (descending co-lon) ischemia, was adjudicated by the CEC. The patient sustained a stroke at 10.5 months and died w3 weeks later.
Outcomes at 1 year
Major adverse events. The MAEs rate at 1 year was 6.7% (95% CI, 1.0%-6.7%). The Kaplan-Meier freedom from MAEs estimate was 93.3% (Supplementary Fig 1, online only). There were 16 MAEs in 10 patients, including death in 6, stroke in 3, bowel ischemia in 2, renal failure in 2, respiratory failure in 2, and myocardial infarction in 1.
Mortality. All-cause mortality through 1 year was 4.0% (n ¼ 6). The Kaplan-Meier freedom from mortal-ity estimate was 96.0% (Supplementary Fig 2, online only). Four patients died after 30 days from complica-tions of lung cancer (n¼ 2), hypertrophic cardiomyop-athy (n ¼ 1), and stroke (n ¼ 1). Aneurysm-related mortality was 1.3% (n ¼ 2). The Kaplan-Meier freedom from aneurysm-related mortality estimate was 98.7% Table II. Baseline patient characteristics3
Characteristic Valuea(N
¼ 150)
Male gender 142 (95)
ASA Physical Status Classification
I/II 41 (27) III 92 (61) IV 17 (11) Medical historyb Hypertension 123 (82) Hyperlipidemia 108 (72) Hypercholesterolemia 81 (54) Smoking history 78 (52)
Coronary artery disease 76 (51) Peripheral vascular disease 44 (29)
Arrhythmia 42 (28)
Prior percutaneous coronary intervention 42 (28) Chronic obstructive pulmonary disease 41 (27)
Cancer 39 (26)
Myocardial infarction 38 (25)
Diabetes mellitus 26 (17)
Prior coronary artery bypass grafting 26 (17)
Renal insufficiency 24 (16)
Stroke/transient ischemic attack 18 (14)
Family history of AAA 19 (13)
Angina 16 (11)
Aortoiliac anatomy
Maximum AAA sac diameter, mm 586 6.2 (44-82) AAA sac volume, mL 1436 53 (63-379) Aortic neck length, mm 316 14 (10-103) Aortic neck diameter, mm 256 3.0 (19-32) Aortic neck angle, 306 14 (3.3-59) Access vessel diameter, minimum mm
Right 7.76 1.4 (3.6-13)
Left 7.86 1.4 (4.9-12)
AAA, Abdominal aortic aneurysm; ASA, American Society of Anesthesiologists.
aCategoric data are reported as number (%) and continuous data as
mean6 standard deviation (min-max).
(Supplementary Fig 3, online only), which included two early deaths and one AAA rupture reported in follow-up (Supplementary Fig 4, online only).
One patient underwent surgical conversion to open repair, which was performed for treatment of recurrent type Ia endoleak and aneurysm rupture. This patient sub-sequently died. Another patient with a history of conges-tive heart failure, emphysema, uncontrolled diabetes, morbid obesity, and prior pulmonary embolism died on postoperative day 4. The Nellix procedure was completed uneventfully, and she was discharged from the hospital on day 2. She suffered cardiac arrest at home and was found unresponsive; she was admitted emergently to the hospital where she continued to decline, developing multiorgan failure. CTA found the device intact and patent, with no evidence of endoleak or retroperitoneal hematoma. Because of the patient’s history and poor prognosis, the family elected no further intervention.
Treatment success. The treatment success rate was 94.0% (126 of 134; 95% CI, 88.6%-97.4%), highly significant against the performance criteria (80%;P< .0001). Treat-ment success at 1 year was 95.3% (95% CI, 90.0%-98.2%) in men (n ¼ 127) and 71.4% (95% CI, 29.0%-96.3%) in women (n ¼ 7). There were 10 events in 8 patients including 1 (0.7%) AAA rupture during follow-up, 1 (0.7%) conversion to open repair, 5 (3.7%) secondary in-terventions, 2 (1.5%) aneurysm enlargements, and 1 (0.7%) type Ib endoleak. No migration >10 mm causing com-plications or requiring secondary intervention, or type III endoleaks were noted at 1-year follow-up.
Stent thrombosis. Stent thrombosis was observed in two patients (1.3%). One of these patients, notably with a body mass index of 54 kg/m2, suffered three throm-botic events of unclear etiology, ultimately treated with relining of the affected Nellix limb with another endog-raft and adjunctive anticoagulation.
Endoleak. Endoleaks by type and follow-up interval are reported inTable III. At 1 year, endoleak incidence was 0.8% for type I, 2.3% for type II, and 0% for type III or IV. Migration. Migration of>10 mm was observed in three of 129 (2.3%) patients at 1 year. None of these individuals presented with aneurysm sac expansion or had an endo-leak or a secondary intervention.
Aneurysm sac morphology changes. Compared with the 1-month baseline, AAA sac diameter increased >5 mm in 2 patients (1.5%), remained stable (<5 mm change) in 129 patients (94.9%), and decreased in 5 pa-tients (3.7%). In the two papa-tients with sac enlargement (8.1 mm and 5.4 mm, respectively), no endoleaks were observed, and secondary intervention was not performed. Secondary procedures. Six secondary procedures were performed infive patients (3.3%) through 1 year for treat-ment of type Ia endoleak (n ¼ 2), AAA rupture during follow-up (n ¼ 1), limb occlusion (n ¼ 2), and inadver-tent renal artery coverage (n¼ 1; Supplementary Fig 5, online only).
Surgical conversion. One patient (previously described) required surgical conversion, which was a relatively straightforward procedure compared with the removal of a conventional endograft with suprarenalfixation. The Kaplan-Meier estimate of freedom from surgical conver-sion was 99.3% (Supplementary Fig 6, online only).
DISCUSSION
This report presents 1-year outcomes from the prospec-tive multicenter Nellix EVAS IDE study. The primary safety hypothesis tested in the Nellix EVAS IDE study was the rate of MAEs at 30 days relative to a performance goal of 56% (SVS open surgical control). The primary safety end point was satisfied with an MAEs rate of 2.7% at 30 days (95% CI, 0.7%-6.7%;P< .0001). The pri-mary effectiveness hypothesis was defined as treatment success>80%. The primary effectiveness end point was satisfied with treatment success at 1 year of 94.0% (95% CI, 88.6-97.4%;P< .0001).
Secondary safety and effectiveness outcomes provided additional supportive evidence of the clinical utility of the Nellix EVAS System. Through 1 year, key secondary outcomes included 6.7% MAEs, 4.7% device-related SAEs, 1.3% AAA-related mortality, 0.7% AAA rupture dur-ing follow-up, 3.3% secondary intervention, and 0.7% sur-gical conversion. These results indicate that the Nellix EVAS System is safe and effective for its intended use.
Generally, the results from this IDE trial are comparable to those reported in IDE trials of commercially available stent grafts. However, the incidence of type II endoleaks and secondary interventions is considerably lower in the Table III. Endoleaks by follow-up interval
Interval No. Type Ia, No. (%) Type Ib, No. (%) Type II, No. (%) Type III, No. (%) Type IV, No. (%)
1 month 142 1 (0.7) 0 (0) 8 (5.6) 0 (0) 0 (0) 6 months 138 2 (1.4) 0 (0) 2 (1.4) 0 (0) 0 (0) New 1 0 0 0 0 Persistent 1 0 2 0 0 1 year 133 0 (0) 1 (0.8) 3 (2.3) 0 (0) 0 (0) New 0 1 2 0 0 Persistent 0 0 1 0 0
Nellix EVAS Pivotal study compared with previous IDE studies. The incidence of type II endoleak at 1 year was 2.3% with the Nellix EVAS System vs 8.9% to 34.3% with other commercially available stent grafts.5-9 This is clini-cally meaningful, because individuals with persistent type II endoleak have a higher risk for AAA expansion and secondary intervention.10-12 In addition, secondary interventions through 1 year were 3.3% with the Nellix EVAS System compared with 5.1% to 11.1% for other devices.5-9,13,14These benefits are also noted in postmarket reports of experience with Nellix in real-world use, sug-gesting that these outcomes can be achieved outside the setting of an IDE trial among both straightforward and complex patient populations across a broad range of vascular anatomies.2
Proximal type I endoleak was reported in two patients within the 6-month follow-up window. Root-cause ana-lyses of the type Ia endoleaks as reported in contempo-rary literature suggest that most are related to patient selection, low device placement caudal to the optimum dealing zone, or underfilling of the endobags, or a com-bination of these. In the present series, the cause of the endoleaks is indefinitive; however, careful follow-up is advocated in case of endoleak progression, and treat-ment should include improving the proximal seal using coils in combination with liquid embolization.
Migration was observed in three patients (2.3%), using the SVS reporting standards definition of $10 mm of movement. Within the 1-year follow-up period, no new endoleaks occurred and no secondary procedures were performed in these three patients. The incidence of migration in this study is comparable to that in other EVAR IDE studies; however, the mechanisms of migration in EVAS are likely unique. In EVAR,fixation and seal are based on discrete points of contact within the aorta. Even small movement can result in loss offixation and seal. In EVAS,fixation and seal are distributed across the entire aneurysm. Movement of Nellix stents is not neces-sarily indicative of movement of the EVAS seal, because the latter is accomplished by the polymer-filled endo-bags, not the stents. Stent migration may occur in areas with inadequate distribution of polymer surrounding angulated stents, especially in small-flow lumens adjacent to large thrombus volume. If movement does occur, how-ever, the Nellix endobags continue to provide an opportu-nity forfixation and seal. Whether these initial movements of the Nellix represent a period of stabilization or whether it becomes progressive, is not clear.
The instructions for use require a minimum neck length of 10 mm and recommend placement of the Nellix stents at the level of the renal arteries. In this present series, the infrarenal neck lengths in the three patients experiencing migration >1 cm were 27 mm, 32 mm, and 59 mm, respectively. Careful follow-up is necessary to determine the natural history of migration in Nellix patients and what developments would warrant further intervention.
As with other IDE EVAR studies, the number of female participants was small, perhaps due to the initial AAA diameter eligibility criterion, and the performance of EVAS in female patients will require further study in pro-tocols designed to examine this issue. Regarding out-comes in women with the Nellix EVAS System, two women experienced an MAE #30 days of treatment. The somewhat higher incidence of MAEs in women is not entirely unexpected because it is well established that women treated with EVAR have greater rates of mortality and morbidity.15-18Still, the benefit of EVAR in women likely remains greater than with open surgical AAA repair, although additional research is warranted.
It is important to understand that these data represent the initial experience with a device and technique for the endovascular therapy for aortic aneurysms that is signifi-cantly different from conventional EVAR therapies. Rather than a procedural end point focused on imaging, EVAS demands careful attention to pressuvolume re-lationships and the careful achievement of a suprasys-tolic polymer pressure to achieve the durable sealing and fixation envisioned as the primary mechanism of this therapy. Other differences involve the behavior of long balloon-expandable stents in often tortuous anat-omy and the interaction of the endobags and the stents, such that optimal stent position is maintained. It is also important to note that with EVAS, the AAA sac may shrink even whenfilled with polymer. Although the sac would not be expected to shrink beyond the volume of the endobag and stent structure, excluded intramural thrombus may still be resorbed, resulting in sac shrinkage.
This initial clinical study in the United States is chal-lenged by the same learning curve issues experienced in the application of EVAR in its early years. Despite these challenges, the results of this study demonstrate EVAS outcomes that appear comparable to those of contem-porary EVAR. Although only longer-term follow-up will determine the place of EVAS in clinical therapy, these early results are encouraging, and better technique and patient selection may contribute to improved outcomes in future studies.
A unique potential risk of EVAS noted in this trial relates to the single case of bowel ischemia, which may have been related to attempts to adjust the position of the de-vice after deployment of the balloon-expandable stents, with resulting atheroembolization. Repositioning the stents once they have been deployed can be chal-lenging. Attempts to reposition the relatively stiff stents can potentially cause atheroembolization, manifested as colon or lower extremity embolic events.
A second unique potential risk of EVAS is iatrogenic aortic rupture. The EVAS mechanism of action is an intentional pressurization of the aneurysm sac to exclude the entire treated segment. When performed according to protocol, the risk is small, with an observed incidence
of 0.7% in the current study. However, attention to detail is important during the endobag filling technique because the pressurization occurs over a relatively small volume increment.
This study had several limitations worth noting. First, given that this was a single-arm study that used safety and effectiveness performance goals, comparative per-formance of EVAS vs EVAR cannot be directly evaluated. Second, because only eight women (5%) were included, further characterizing EVAS performance and durability in women in future studies will be important. Third, pa-tient outcomes>1 year were not yet available, and there-fore, long-term outcomes with EVAS are currently unknown. Finally, because this clinical trial enrolled high-ly selected patients at high-volume centers, these out-comes may not be generalizable to the typical AAA patient in real-world practice.
CONCLUSIONS
Outcomes with the Nellix EVAS System are encour-aging, with very low morbidity and mortality and high procedural and treatment success. The primary safety and effectiveness end points have been achieved. Continued follow-up is in progress and will be essential for evaluation of the long-term results and durability of the device and the EVAS procedure.
AUTHOR CONTRIBUTIONS Conception and design: JC Analysis and interpretation: JC
Data collection: JC, CB, MR, CH, RC, SH, JT, DB Writing the article: JC
Critical revision of the article: JC, CB, MR, CH, RC, SH, JT, DB
Final approval of the article: JC, CB, MR, CH, RC, SH, JT, DB Statistical analysis: JC
Obtained funding: JC, CB, MR, CH, RC, SH, JT, DB Overall responsibility: JC
REFERENCES
1. Holden A, Savlovskis J, Winterbottom A, van den Ham LH, Hill A, Krievins D, et al. Imaging after Nellix endovascular aneurysm sealing: a consensus document. J Endovasc Ther 2016;23:7-20.
2. Böckler DM, Holden A, Thompson M, Hayes P, Krievins D, deVries JP, et al. Multicenter Nellix EndoVascular Aneurysm Sealing system experience in aneurysm sac sealing. J Vasc Surg 2015;62:290-8.
3. Carpenter JP, Cuff R, Buckley C, Healey C, Hussain S, Reijnen MM, et al. 30-day results of the Nellix System IDE pivotal trial for endovascular aneurysm sealing. J Vasc Surg 2016;63:23-31.
4. Zwolak RM, Sidawy AN, Greenberg RK, Schermerhorn ML, Shackelton RJ, Siami FS, et al. Lifeline registry of endovas-cular aneurysm repair: open repair surgical controls in clin-ical trials. J Vasc Surg 2008;48:511-8.
5. U.S. Food and Drug Administration. Summary of safety and effectiveness data. EXCLUDER bifurcated endoprosthesis
(report no. P020004b); 2002. Available at:http://www.access data.fda.gov/cdrh_docs/pdf2/P020004b.pdf. Accessed May 17, 2016.
6. U.S. Food and Drug Administration. Summary of safety and effectiveness data. Talent Abdominal Stent Graft System (Report No. P070027b); 2008. Available at: http://www. accessdata.fda.gov/cdrh_docs/pdf7/P070027b.pdf. Accessed May 17, 2016.
7. U.S. Food and Drug Administration. Summary of safety and effectiveness data. Endurant Stent Graft System (report no. P100021b); 2010. Available at:http://www.accessdata.fda.gov/ cdrh_docs/pdf10/P100021b.pdf. Accessed May 17, 2016. 8. U.S. Food and Drug Administration. Summary of safety and
effectiveness data. Ovation Abdominal Stent Graft System (report no. P120006b); 2012. Available at: http://www. accessdata.fda.gov/cdrh_docs/pdf12/P120006b.pdf. Accessed May 17, 2016.
9. U.S. Food and Drug Administration. Summary of safety and effectiveness data: Aorfix Flexible Stent Graft System (Report No. P110032) 2013. Available at: http://www.accessdata.fda. gov/cdrh_docs/pdf11/P110032b.pdf. Accessed May 17, 2016. 10.Jones JE, Atkins MD, Brewster DC, Chung TK, Kwolek CJ,
Lamuraglia GM, et al. Persistent type 2 endoleak after endovascular repair of abdominal aortic aneurysm is asso-ciated with adverse late outcomes. J Vasc Surg 2007;46:1-8. 11. Nolz R, Teufelshauer H, Asenbaum U, Beitzke D, Funovics M,
Wibmer A, et al. Type II endoleaks after endovascular repair of abdominal aortic aneurysms: fate of the aneurysm sac and neck changes during long-term follow-up. J Endovasc Ther 2012;19:193-9.
12. vanMarrewijk CJ, Fransen G, Laheij RJ, Harris PL, Buth J; EUROSTAR Collaborators. Is a type II endoleak after EVAR a harbinger of risk? Causes and outcome of open conversion and aneurysm rupture during follow-up. Eur J Vasc Endovasc Surg 2004;27:128-37.
13. Arko FR, Hill BB, Reeves TR, Olcott C, Harris EJ, Fogarty TJ, et al. Early and late functional outcome assessments following endovascular and open aneurysm repair. J Endovasc Ther 2003;10:2-9.
14. U.S. Food and Drug Administration. Summary of safety and effectiveness data: Endologix Powerlink System (Report No. P040002) 2004. Available at:http://www.accessdata.fda.gov/ scripts/cdrh/cfdocs/cfTopic/pma/pma.cfm?num¼P040002. Accessed May 24, 2016.
15. Abedi NN, Davenport DL, Xenos E, Sorial E, Minion DJ, Endean ED. Gender and 30-day outcome in patients un-dergoing endovascular aneurysm repair (EVAR): an analysis using the ACS NSQIP dataset. J Vasc Surg 2009;50:486-91. 16. Chung C, Tadros RO, Torres M, Malik R, Ellozy S, Faries P, et al.
Evolution of gender-related differences in outcomes from two decades of endovascular aneurysm repair. J Vasc Surg 2015;61:843-52.
17. Dillavou ED, Muluk SC, Rhee RY. Does hostile neck anatomy preclude successful endovascular aortic aneurysm repair? J Vasc Surg 2003;38:657-63.
18. Mehta M, Byrne WJ, Robinson H, Roddy SP, Paty PS, Kreienberg PB, et al. Women derive less benefit from elective endovascular aneurysm repair than men. J Vasc Surg 2012;55:906-13.
Submitted May 24, 2016; accepted Sep 20, 2016.
Additional material for this article may be found online atwww.jvascsurg.org.
Supplementary Table I (online only). Investigational sites and principal investigators (PIs)3
Investigative site Location PI
Cooper University Hospital Camden, NJ Jeffrey P. Carpenter, MD (Global PI); Jose Trani, MD (Site PI)
Addenbrooke’s Hospital Cambridge University Cambridge, United Kingdom Paul Hayes, MD
Allegheny General Hospital Pittsburgh, Pa Satish Muluk, MD
Baylor Heart Hospital Plano, Tex Javier Vasquez, MD
Baylor Scott and White Healthcare System Temple, Tex Clifford Buckley, MD
Baystate Medical Center Springfield, Mass Neal Hadro, MD
Carolinas Health Care Charlotte, NC Steven Lalka, MD
Christiana Hospital Wilmington, De Ralph Ierardi, MD
Cleveland Clinic Cleveland, Ohio Daniel Clair, MD
Froedtert Memorial Lutheran Hospital (Medical College of Wisconsin)
Milwaukee, Wisc Cheong Jun Lee, MD
Inova Hospital Fairfax, Va Homayoun Hashemi, MD
Maine Medical Center Portland, Me Christopher Healey, MD
MedStar Health Research Institute D.C. Nelson Bernado, MD
Miami Vascular Institute Miami, Fla James Benenati, MD
Nebraska Heart Hospital Lincoln, Neb Steve Tyndall, MD
Ohio Health Research Institute Columbus, Ohio Mitchell Silver, DO Providence Sacred Heart Medical Center Spokane, Wash Stephen Murray, MD Rijnstate Hospital Arnhem, The Netherlands Michel Reijnen, MD, PhD
Sacred Heart Hospital Pensacola, Fla Stuart Harlin, MD
San Diego VA Hospital San Diego, Calif John Lane, MD
Spectrum Health Grand Rapids, Mich Robert Cuff, MD
St. Elizabeth’s Hospital Brighton, Mass Nikhil Kansal, MD
St. Luke’s Episcopal Hospital Houston, Tex Zvonimir Krajcer, MD
St. Vincent Healthcare Billings, Mont Kevin Bruen, MD
St. Vincent’s Heart Center of Indiana Indianapolis, Ind Sajjad Hussain, MD Tucson Medical Center/PIMA Vascular Tucson, Ariz Luis Leon, MD
University Hospital Heidelberg, Germany Dittmar Böckler, MD, PhD
University of Pittsburgh Medical Center Pittsburgh, Pa Michel Makaroun, MD
Yale New Haven Hospital New Haven, Conn Jeffrey Indes, MD
Supplementary Table II (online only). One-year out-comes in“roll-in” patients
Outcome Percentage (n¼ 29)
Technical success 100
Major adverse event 14
AAA rupture 0 Surgical conversion 3.4 Secondary intervention 6.9 AAA-related mortality 0 All-cause mortality 3.4 Type I endoleak 4.0
Type III endoleak 0
Migration 3.8
AAA enlargement 3.8
150 146 146 144 144 143 143 143 143 141 141 139 136 0 2 4 6 8 10 12 Month 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Supplementary Fig 1 (online only). Freedom from major adverse events (MAEs) through 1 year. Kaplan-Meier estimate: 93.3% (standard error: 2.0%). The blue shaded region represents the 95% confidence limit band (87.9%, 96.3%) at 1 year. Values above thex-axis are the number of patients at risk at each follow-up interval.
150 149 149 149 148 147 147 147 147 145 145 143 140 0 2 4 6 8 10 12 Month 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
Supplementary Fig 2 (online only). Freedom from all-cause mortality through 1 year. Kaplan-Meier estimate: 96.0% (standard error: 1.6%). Theblue shaded region represents the 95% confidence limit band (91.3%, 98.2%) at 1 year. Values above thex-axis are the number of patients at risk at each follow-up interval.
150 149 149 149 148 147 147 147 147 145 145 143 140
0
2
4
6
8
10
12
Month
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Supplementary Fig 3 (online only). Freedom from abdominal aortic aneurysm (AAA)-related mortality through 1 year. Kaplan-Meier estimate: 98.7% (standard error: 0.9%). Theblue shaded region represents the 95% con fi-dence limit band (94.7%, 99.7%) at 1 year. Values above thex-axis are the number of patients at risk at each follow-up interval. 150 149 149 149 148 147 147 147 146 145 145 143 140
0
2
4
6
8
10
12
Month
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Supplementary Fig 4 (online only). Freedom from abdominal aortic aneurysm rupture through 1 year. Kaplan-Meier estimate: 99.3% (standard error: 0.7%). Theblue shaded region represents the 95% confidence limit band (95.3%, 99.9%) at 1 year. Values above thex-axis are the number of patients at risk at each follow-up interval.
150 148 147 147 146 144 144 143 143 142 141 139 136
0
2
4
6
8
10
12
Month
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Supplementary Fig 5 (online only). Freedom from device-related secondary intervention through 1 year. Kaplan-Meier estimate: 96.6% (standard error: 1.5%). Theblue shaded region represents the 95% confidence limit band (92.1%, 98.6%) at 1 year. Values above thex-axis are the number of patients at risk at each follow-up interval. 150 149 149 149 148 147 147 147 146 145 145 143 140
0
2
4
6
8
10
12
Month
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Supplementary Fig 6 (online only). Freedom from surgical conversion through 1 year. Kaplan-Meier estimate: 99.3% (standard error: 0.7%). Theblue shaded region represents the 95% confidence limit band (95.3%, 99.9%) at 1 year. Values above thex-axis are the number of patients at risk at each follow-up interval.
