Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 1.
Published in final edited form as: J Vasc Surg. 2022 Oct 28;77(3):731–740.e1. doi: 10.1016/j.jvs.2022.10.030

Mortality Analysis of Endovascular Aneurysm Sealing versus Endovascular Aneurysm Repair

Vinamr Rastogi 1,2, Thomas FX O’Donnell 1, Yoel Solomon 1,3, Rens RB Varkevisser 1,2, Priya B Patel 1, Jeffrey P Carpenter 4, Jorg L de Bruin 2, Michel MPJ Reijnen 5, Hence JM Verhagen 2, Marc L Schermerhorn 1
PMCID: PMC9974809  NIHMSID: NIHMS1846087  PMID: 36651654

Abstract

Background:

Endovascular aneurysm sealing (EVAS), using The Nellix Endovascular Aneurysm Sealing System has been found to be associated with high reintervention and migration rates. However, prior reports have suggested that EVAS may be related to a lower all-cause mortality compared with endovascular aneurysm repair (EVAR). In this study we examined 5-year all-cause mortality trends after EVAS and EVAR.

Methods:

We compared the 333 EVAS patients in the EVAS-1 Nellix United States Investigational Device Exemption (IDE) trial to 16,497 infrarenal EVAR controls from the Vascular Quality Initiative (VQI), treated between 2014 and 2016 after applying the exclusion criteria from the IDE (hemodialysis, creatinine >2.0 mg/dL, or rupture). As a secondary analysis, we stratified by aneurysm diameter (<5.5cm and ≥5.5cm). We calculated propensity scores adjusting for demographics, comorbidities and anatomical characteristics, and applied inverse probability weighting to compare risk adjusted long-term mortality using Kaplan-Meier and Cox-regression analysis.

Results:

After weighting, patients treated with EVAS experienced similar 5-year mortality compared with controls from the VQI (EVAS vs. EVAR; 18% vs. 14% / Hazard Ratio [HR]: 1.1[95%CI 0.71–1.7], p=.70). Subgroup analysis demonstrated that in patients with an aneurysm diameter <5.5cm, EVAS was associated with higher 5-year mortality compared with EVAR (19% vs. 11% / HR: 2.4 [95%CI 1.7–4.7], p=.013). In patients with an aneurysm diameter ≥5.5cm, EVAS was associated with lower mortality within the first two years (two-year mortality HR: 0.29 [95%CI 0.13–0.62], p=.002). However, compared with EVAR, EVAS was associated with higher mortality between two and five years (HR: 1.9 [95%CI 1.2–3.0], p=.005), with no mortality difference at five years (18% vs. 17% / HR: 0.82 [95%CI 0.4–1.4], p=.46).

Conclusion:

Within the overall population, EVAS was associated with similar 5-year mortality compared with EVAR. EVAS was associated with higher mortality in small aneurysms (<5.5cm). For larger aneurysms (≥5.5cm), EVAS was initially associated with lower mortality in the first two years, though this advantage is lost thereafter with higher mortality after two years. Future studies will be required to evaluate the specific causes of death, and to elucidate the beneficial mechanism behind sac obliteration that leads to this potential initial survival benefit. This could help guide development of future grafts with better proximal fixation and sealing, while incorporating sac obliteration.

Keywords: abdominal aortic aneurysm, endovascular aneurysm repair, endovascular aneurysm sealing, survival, mortality

Table of Contents Summary:

Although EVAS seems to be associated with lower mortality in the first two years in aneurysms ≥5.5cm, this association attenuated and is no longer significant at 5-years. Future studies will be required to elucidate the beneficial mechanism behind aneurysm sac sealing that leads to this potential initial survival benefit.

Introduction

Since its introduction in 1991,1 endovascular aneurysm repair (EVAR) has been widely adopted for treatment of abdominal aortic aneurysms (AAA) due to lower perioperative mortality and morbidity compared with open repair.2,3 However, large randomized clinical trials have shown that EVAR’s initial perioperative benefit may be lost over time.4–7 The EVAR-1 trial demonstrated that 8-years after their index procedure, EVAR patients were at higher risk for aneurysm-related mortality compared with patients who underwent open repair.4 Causes for durability loss include type-I/III endoleaks, aneurysm-related reinterventions and secondary rupture, as the aneurysm sac, though remodeled, is still intact.

Aside from secondary rupture, EVAR is associated with high rates of major adverse cardiac events (MACE), alongside a high cardiovascular related mortality burden of ~30%.8,9 Prior literature has demonstrated that pressure on the aneurysm sac leads to an inflammatory cascade, potentially explaining the high rates of MACE following EVAR, especially given the relation of thrombus formation with increased inflammation.10–12 Previous studies have demonstrated that the inflammatory response is secondary to aortic remodeling after EVAR,13–15 and reduction of this inflammatory response might secondarily reduce MACE. And, prior studies demonstrated that sac regression was associated with higher survival.16 Thus, given the issues with conventional EVAR, by minimizing tension and by preventing remodeling of the aneurysm sac, active sac management has been theorized to hold great potential for AAA management.

Endovascular aneurysm sealing (EVAS) using the The Nellix Endovascular Aneurysm Sealing System (Endologix, Inc, Irvine, CA) provided a different concept of treating AAA with active sac management through endovascular means. In contrast to traditional devices used in EVAR, the concept of EVAS is to stabilize and completely seal the aneurysm sac with polymer-filled endobags attached to covered stents that span the aortic segment and maintain blood flow to the distal limbs. The Nellix device received their CE trademark in 2013, and initial reports showed promising results with the EVAS system with acceptable technical success, lower type-II endoleaks, and reduced post-implantation syndrome.17–19 Furthermore, all-cause mortality comparisons between EVAS and EVAR, including a prior study comparing the EVAS-1 Nellix United States Investigational Device Exemption (IDE) trial with EVARs from the Vascular Quality Initiative (VQI),20 demonstrated lower mid-term mortality following EVAS compared with EVAR.20,21 However, these findings were not confirmed in a study by Zoethout and colleagues,22 and later reports with longer follow-up displayed concerningly high rates of graft failure, type-IA endoleaks and stent migration,23,24 eventually leading to withdrawal of the device from the market.

Although the graft is no longer available, there is still value in understanding all-cause mortality trends following EVAS to understand if there is any benefit in sac obliteration. Thus, in this study we performed a 5-year all-cause mortality analysis comparing EVAS from the United Stated IDE trial with EVARs from the VQI.

Methods

Database and Patients:

EVAS:

We included EVAS patients from the EVAS-1 Nellix system investigational device exemption (IDE; 2014–2016) and pivotal trial which was a prospective, multicenter, single-arm clinical study that was conducted at 29 sites under an IDE approved by the United States Food and Drug Administration. The Nellix system IDE was approved by the local institutional review board (IRB) at each participating center prior to enrollment of any patients. The inclusion and exclusion criteria of this trial have previously been reported.18 Regarding the treatment threshold, it must be noted that patients included AAA >5.0cm, or AAA >4.5 cm that increased by >0.5cm times within the last 6 months, or which exceeded 1.5 times the transverse dimension of an adjacent nonaneurysmal aortic segment. Each patient provided written informed consent to participate in the trial and underwent preoperative eligibility assessment. Postoperatively, patients were followed at 1 month, 6 months, 1 year and annually thereafter with laboratory data and computed tomography angiography (CTA). Cause of death was determined by an independent Clinical Events Committee, and an independent Data Safety and Monitoring Board reviewed data on a periodic basis.

EVAR Controls:

We included EVAR patients from the Society for Vascular Surgery VQI as a control group. The VQI is a quality improvement registry established to improve patient care through the prospective collection of clinical data. Data include patient demographics, comorbidities, and perioperative complications. Although long-term outcomes such as re-intervention rates are not well documented in the VQI, mortality is well captured due to the linkage with the Social Security Death Index. More information about the VQI can be found at www.vascularqualityinitiative.org. To match the eligibility criteria from the Nellix IDE, we only included patients undergoing EVAR from 2014–2016, and excluded patients who presented with ruptured AAA and were on dialysis or with a creatinine >2.0 mg/dL. (Supplementary Figure)

The present report adhered to the applicable STROBE (strengthening the reporting of observational studies in epidemiology) and RECORD (Reporting of studies Conducted using Observational Routinely-collected Data) standards for observational studies.25,26 (Supplementary Table) The Beth Israel Deaconess Medical Center Institutional Review Board approved this study and waived the need for further patient consent because all patient data was deidentified.

Variables, Definitions and Outcomes

We calculated estimated glomerular filtration (eGFR) rate utilizing the CKD-EPI formula27, and categorized chronic kidney disease (CKD) into: none, eGFR 30–60, and eGFR <30. Coronary artery disease was defined as a history of documented coronary artery disease, previous myocardial infarction, angina, arrhythmia, coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI). Obesity was defined as a body mass index above 30 kg/m2.

Our primary outcome was 2- and 5-year all-cause mortality which was compared between EVAS and EVAR patients. In a secondary analysis, we compared 5-year all-cause mortality stratified by aneurysm size. As we hypothesized that aneurysm sac wall stress and remodeling were associated with higher mortality risk, we theorized that any mortality difference would be most pronounced in the subgroup of patients with larger aneurysms who would experience the most risk.20

Statistical Analysis

We compared demographics, comorbidities, and procedural characteristics between the study groups. Categorical variables were presented as counts and percentages and were compared using the Pearson’s χ2-test or Fisher’s exact tests where appropriate. For continuous variables, all were parametric and were presented as means and standard deviations, and compared using student t-tests.

Low outcome event rates within the EVAS group precluded robust conventional multivariable adjustment.28 Therefore we calculated propensity scores and used inverse probability weighting to compare outcomes after EVAR and EVAS. Thus, we generously introduced covariates into our model, including age, race, sex, aortic diameter, surgery year, peripheral vascular disease, hypertension, chronic kidney disease (CKD), prior coronary artery bypass graft (CABG), percutaneous coronary intervention (PCI), diabetes, chronic obstructive pulmonary disease, obesity, congestive heart failure (CHF), preoperative aspirin use, preoperative statin use, presence of one or more iliac aneurysm, family history of AAA, and smoking status. Mean standardized differences were less than 10% for all variables after weighting, indicating adequate balance. We then used these weights to compare the risk of 5-year mortality between the treatment modalities. We constructed weighted Kaplan- Meier survival curves and compared the EVAS patients and EVAR controls using weighted Cox regression analysis.

Given the inherent differences between an IDE trial and data from a registry, we performed a sensitivity analysis in which we compared EVAS patients with EVAR patients who were treated inside of instruction for use for neck characteristics (neck diameter, neck length, infrarenal angulation, suprarenal angulation).

As a secondary analysis, we performed 5-year all-cause mortality analyses comparing EVAS and EVAR in groups stratified by aneurysm size (<5.5cm and ≥5.5cm), and this cutoff was used to remain in line with prior literature.20,22 For these analyses, new propensity scores were independently created within the respective stratified cohorts.

We plotted the number of deaths within the EVAS-1 IDE cohort by the number of years since repair, and displayed the causes of death which were stratified by AAA-related death, cardiovascular (CV)-related death and death due other cause. As the VQI only provides all-cause mortality, we could not reproduce this analysis for the EVAR controls.

Finally, we assessed the effect of sac and thrombus volume on mortality following EVAS. As the VQI does not report any information on this front, such an analysis was prohibited for EVAR patients.

All data analyses were performed using R version 4.0.3.

Missing data

All values aside of race (2.6%), iliac aneurysm (3.3%), and neck characteristics (~65%) contained <1% of missing data. Given the high proportion of missingness in neck characteristics, these were not included for adjustment in the primary analysis, but a sensitivity analysis was performed, comparing EVAS IDE patients with on-IFU EVAR patients as previously mentioned. Patients with missing values for any of the other variabels were excluded from risk-adjusted analysis.

Results

Patient Characteristics:

There were 333 EVAS patients and 16,497 patients in the EVAR control group. Baseline characteristics of the EVAS cohort and the EVAR controls are shown in Table 1. Compared with the EVAR controls, patients in the EVAS cohort had larger aortic diameters, more frequently had a family history of AAA, had higher rates of peripheral vascular disease, were more likely to have undergone prior PCI, and were more likely to be obese (all p<.05). However, compared with EVAR controls, EVAS patients less frequently were female, or had a history of smoking and CHF (all p<.05).

Table 1.

Baseline Characteristics of EVAS patients from the EVAS-1 Nellix IDE and EVAR controls from the Vascular Quality Initiative before weighting and mean standard difference after weighting (SD<0.10)

EVAS (N=333) EVAR Controls (N=16497) P-value Mean Standard Deviation After Weighting
Follow-up time, months Median (Interquartile Range) 59.4 (42.7 – 61.9) 62.1 (15.1 – 76.0) <.001 NA
Age 73 [SD 8.1] 73 [SD 8.8] .30 −0.091
Female Sex 21 (6.3%) 3140 (19%) <.001 0.087
AAA-Diameter 56.8 (SD 6.0) 55.8 (SD 11.7) .006 0.073
Non-Hispanic White 306 (92%) 14496 (88%) .50 0.030
Familiar AAA 43 (13%) 1360 (8.2%) .004 −0.016
Obesity 129 (39%) 5194 (32%) .006 −0.013
Hypertension 276 (83%) 13606 (83%) <.001 −0.036
Prior Percutaneous Coronary Intervention 92 (28%) 3606 (22%) .014 0.011
Prior Coronary Artery Bypass Graft 61 (18%) 3026 (18%) 1.0 −0.084
Prior CHF 24 (7.2%) 1904 (12%) .018 −0.10
Diabetes 62 (19%) 2775 (17%) .44 −0.081
Prior COPD 94 (28%) 5450 (33%) .073 −0.065
Smoking Status <.001
 Prior 85 (26%) 9017 (55%) −0.10
 Current 84 (25%) 5244 (32%) −0.016
Prior CKD 87 (26%) 5296 (32%) .024 −0.077
Peripheral Vascular Disease 85 (26%) 1230 (7.5%) <.001 0.096
Aspirin 226 (68%) 10672 (65%) .26 0.010
Statin 249 (75%) 11509 (70%) .058 0.10
Iliac Aneurysm 51 (15%) 4151 (25%) <.001 −0.042
Wide Neck (>30mm) * 387 (6.5%) 333 (6.9%) 1.0 NA
Short Neck (<15mm) * 735 (12%) 37 (11%) .29 NA
Infrarenal Angulation (>60 degr) * 391 (6.6%) 0 (0.0%) <.001 NA
Suprarenal Angulation (>45 degr) * 34 (0.6%) 0 (0.0%) <.001 NA
Open in a new tab

AAA = abdominal aortic aneurysm; CHF = congestive heart failure; COPD = chronic obstructive pulmonary disease; CKD = chronic kidney disease

*

Missing information in ~65% of patients

Five-Year All-Cause Mortality in the Full cohort:

Crude mortality at five years was 18% following EVAS, and 13% following EVAR controls (p=.27; Figure 1). After weighting, EVAS patients experienced similar five-year mortality compared with EVAR controls (EVAS vs. EVAR; 18% vs. 14%; Hazard Ratio [HR]: 1.1 [95% Confidence Interval (CI): 0.71–1.7], p=.70). There was no difference in two-year mortality between groups (two-year mortality HR: 0.77 [95%CI: 0.50–1.18], p=.22), however EVAS was associated with significantly higher mortality hazard between two and five years (HR: 2.0 [95%CI 1.4–2.9], p<.001)

Figure 1.

Figure 1.

Open in a new tab

Adjusted 5-year Survival Curves. P = .61, all standard errors < 0.1. EVAS = Endovacular Aneurysm Sealing; EVAR = Endovascular Aneurysm Repair

Five Year All-Cause Mortality in Subgroups Stratified by Aneurysm Diameter

Diameter <5.5cm:

We identified 121 (36%) EVAS patients and 7,905 (48%) EVAR controls who were treated for aneurysms below 5.5cm. Inverse-probability weighted subgroup analysis in this patient cohort demonstrated that EVAS was associated with higher five-year mortality compared with EVAR controls (19% vs. 11%; HR: 2.4 [95%CI: 1.7–4.7], p=.013; Figure 2A). Within the first two years, there was no significant difference in mortality between groups (two-year mortality HR: 1.2 [95%CI: 0.47–3.1], p=.70), though the mortality was significantly higher following EVAS between two and five years (HR: 2.1 [95%CI: 1.1–4.0], p=.021).

Figure 2A.

Figure 2A.

Open in a new tab

Adjusted Long-term survival in the subgroup of patients with aneurysm diameters <5.5 cm.. P=.016 in this subgroup. All standard errors <0.1. EVAR: Endovascular Aneurysm Repair. EVAS: Endovascular Aneurysm Sealing

Diameter ≥5.5cm:

With regard to patients who were treated for aneurysms above or equal to 5.5cm, we identified 209 (64%) EVAS patients and 8,314 (52%) EVAR controls. Following inverse probability weighted analyses, there was no overall five-year mortality difference between groups (18% vs. 17%; HR: 0.82 [95%CI: 0.48–1.4], p=.46; Figure 2B). Within the first two years, compared with EVAR controls, EVAS was associated with lower mortality (two-year mortality HR: 0.29 [95%CI 0.13–0.62], p=.002). However, this association was lost over time and the curves crossed at 4 years with higher mortality following EVAS between two and five years (HR: 1.9 [95%CI 1.2–3.0], p=.005).

Figure 2B.

Figure 2B.

Open in a new tab

Adjusted Long-term survival in the subgroup of patients with aneurysm diameters ≥5.5 cm. P=.55 in this subgroup. All standard errors <0.1. EVAR: Endovascular Aneurysm Repair. EVAS: Endovascular Aneurysm Sealing

Sensitivity Analysis; All-cause mortality comparison between EVAS and on-label EVAR

We identified 4,781 on-label EVAR patients, of which 2,283 patients had a preoperative aneurysm diameter <5.5cm and 2,498 patients had a preoperative diameter ≥5.5cm. After weighting, within the overall cohort of the sensitivity analysis, EVAS was associated with higher 5-year mortality compared with on-label EVAR (1.36 [95%CI: 1.02–1.83], p=.034; Table 2). Among patients with aneurysms <5.5cm, following inverse-probability weighted analysis, EVAS was associated with a trend towards higher 5-year mortality compared with on-label EVAR (HR: 1.60 [95%CI: 0.99–2.58], p=.055). Among patients with aneurysms ≥5.5cm, following inverse-probability weighted analysis, EVAS was associated with a trend towards lower mortality within the first two years (HR: 0.58 [95%CI: 0.29–1.15], p=.12), though this association got lost at 5-years (HR: 1.09 [95%CI: 0.69–1.72], p=NS).

Table 2.

Adjusted long-term mortality outcomes of sensitivity analysis: Comparison between endovascular aneurysm sealing versus on-label instructions for use endovascular aneurysm repair

Overall Cohort Diameter <5.5cm Diameter ≥5.5cm
EVAS (n=333) vs. EVAR (n=4,781) Hazard Ratio EVAS (n=121) vs. EVAR (n=2,283) Hazard Ratio EVAS (n=212) vs. EVAR (n=2,498) Hazard Ratio
2-year mortality 0.91 (0.58–1.42); p=NS 1.49 (0.56–3.93); p=NS 0.58 (0.29–1.15); p=.12
2–5 year mortality 2.25 (1.50–3.39); p=<.001 2.43 (1.22–4.83); p=<.001 2.15 (1.30–3.58); p<.001
5-year mortality 1.36 (1.02–1.83); p=.034 1.60 (0.99–2.58); p=.055 1.09 (0.69–1.72); p=NS
Open in a new tab

EVAS = endovascular aneurysm sealing; EVAR = endovascular aneurysm repair

Causes of Death in EVAS-1 IDE cohort

Over the course of the five-year study period, 53 (16%) of EVAS-1 patients died. Of all death cases, 15% were AAA-related deaths and 21% were CV-related deaths. (Figure 3) There was a peak in AAA-related deaths in the first year post-implantation, of which one patient (0.3%) suffered AAA-related death within 30-days. There then was a decline in AAA-related mortality in the second and third postoperative year, again followed by an increase in AAA-related mortality after 3–4 years.

Figure 3.

Figure 3.

Open in a new tab

Rates of Death and proportion of Abdominal Aortic Aneurysm (AAA)-related death and Cardiovascular (CV)- related death by year since Repair in the EVAS-1 Investigational Device Exemption trial

Effect of Thrombus on Death following EVAS

Among EVAS patients, compared with patients with aneurysms <5.5cm, patients with aneurysms with a diameter ≥55cm had larger sac volumes (153 [±49] cc vs. 108 [±27] cc, p<.001), alongside a higher thrombus volume (79 [±38] cc vs. 49 [±21] cc, p<.001). Furthermore, the proportion of thrombus volume to sac volume was higher in patients with aneurysms ≥55mm too (51% [±16%] vs. 45% [±14%], p=.003). Nevertheless, following EVAS, neither sac volume (HR/100cc: 1.13;[95%CI: 0.67–1.90]), thrombus volume (HR/100cc: 0.93;[95%CI: 0.45–1.94]), or the proportion of thrombus to sac volume were associated with an increased hazard of mortality (HR/%: 0.67;[95%CI: 0.13–3.55]).

Discussion

In this observational study comparing five-year all-cause mortality between EVAS from the EVAS-1 US-IDE trial and EVAR from the VQI, we found no difference in all-cause mortality between the repair modalities at five years. However, sensitivity analysis including EVAR patients who were treated within IFU for neck characteristics, demonstrated higher 5-year mortality following EVAS compared with on-IFU EVAR. Furthermore, following stratification by aneurysm diameter, EVAS was associated with higher five-year all-cause mortality compared with conventional EVAR in aneurysms <5.5cm. However, in aneurysms ≥5.5cm, EVAS was associated with similar five-year all-cause mortality risk compared with EVAR, with lower mortality with EVAS in the first two postoperative years, but higher mortality with EVAS in the period between two to five years. Stratification by aneurysm diameter within the sensitivity analysis demonstrated similar trends.

We found a similar five-year all-cause mortality between EVAR and EVAS. Although various baseline characteristics such as female sex, PCI, smoking, and CHF were different between groups, these differences were eliminated following risk adjusted analysis as was demonstrated with a mean standard deviation of <0.10, indicating adequate balance. Our findings corroborated outcomes from another EVAS vs. EVAR study by Zoethout et al., who also reported similar five-year mortality or survival between EVAR and EVAS (n=514).22 In a prior analysis by O’Donnell et al.,20 EVAS was associated with lower mid-term mortality compared with EVAR. Several theories have been proposed for the mechanism behind this observation, including the theory regarding reduced thrombosis due to sac obliteration, leading to lower inflammation.20 However, with extended follow-up, these data of our current study did not find a similar association in the overall cohort, and demonstrated that the all-cause mortality after EVAS was significantly higher after two years. Furthermore, sensitivity analysis in which we only included EVAR patients who were treated within IFU for neck characteristics, demonstrated higher 5-year mortality following EVAS compared with on-IFU EVAR patients. These findings are in line with expectations, as prior data showed high reintervention and migration rates following EVAS from 2-year onwards.23,24,29 A series of EVAS patients by Quaglino et al. reported a rapid increase in aneurysm-related mortality in this timeframe (two-years: 3.1%, five years: 10.1%), alongside an increase in technical failure over time, especially after 3–4 years.30 Thus, though uncertain, we speculate that the increased all-cause mortality after the two-year timeframe may be driven by device failure. Despite a small sample size, all-cause mortality analysis of the EVAS cohort demonstrated a peak in AAA-related mortality following 3-years, potentially supporting this argument.

Among patients with larger aortic diameter, EVAS was associated with lower mortality within the first 2-year after repair compared with EVAR. However, between two and five years after repair EVAS was associated with higher mortality in patients with larger aortic diameter. Along with our findings of higher five-year mortality following EVAS compared with EVAR in aneurysms <5.5cm, these data suggest that EVAS is comparatively more beneficial in larger aneurysms, but not in smaller aneurysms, confirming prior findings by O’Donnell et al. that EVAS seemed to be more beneficial in larger aneurysms.20 And although this outcome lost significance in our sensitivity analysis comparing EVAS with EVAR patients who were on-IFU for neck characteristics, probably due to the low EVAS sample size, the hazard ratio followed a similar trend. These findings may support the argument that the initial lower mortality following EVAS may be related to the reduction of an inflammatory response which is mediated through active sac obliteration. Furthermore, we found that the proportion of CV-related death was around 20%, which is lower than previously reported CV-related death following EVAR (~30%), though interpretation of our data was difficult due to the limited sample size of the EVAS cohort.31 Alongside our findings, the prior study by Zoethout et al. also found similar five-year mortality between EVAS and EVAR in aneurysms ≥5.5cm, though the authors did not display a Kaplan-Meier curve or data on mortality differences at an earlier point within this subset of patients. This limited any comparison with other studies on this front, and due to high missingness in VQI’s long-term follow-up module, limited comparison of causes of death with this study’s EVAR cohort. Nevertheless, to see whether active sac management strategies do indeed lead to reduced cardiovascular related cause of death compared with EVAR remains an interesting point for future investigation.

As the Achilles heel of conventional EVAR remains its durability, due to which EVAR has minimal effect on long-term mortality compared with open repair, EVAS was designed to reduce/eliminate the risk of complications such as endoleaks through active sac management. Despite the initial promising outcomes, a sudden increase in type-IA endoleaks and migration from two years onwards led to withdrawal of the device from the market. Root-cause analysis reported technical errors including low device placement and underfilling of the endobag which led to refinements of the instruction for use (IFU). The Nellix system is currently undergoing further study in the EVAS-2 IDE trial. Future investigation, including outcomes from the EVAS-2 IDE trial shall demonstrate whether the improved device alongside the new IFU-criteria shall lead to lower risk of device failure. However, given the known problems with EVAR, this exploratory analysis may support the argument that while this first-generation device has issues that led to withdrawal of the device, the underlying concept of active sac management may hold potential in future AAA management. And at present, various different methodologies of active sac management are undergoing trials including AAA-SHAPE32,33 and Aneufix.34 Thus, this concept warrants future research to examine the actual potential.

Our study needs to be interpreted in the context of its retrospective study design. This is a retrospective analysis, not a randomized trial, and therefore we can only show association, not causation, and the results should be interpreted with caution. As the EVAS cohort was small, this may have resulted in an inability to detect significant differences in the overall cohort. Otherwise, it is possible that the mortality differences we found in the subcohorts was merely the product of statistical noise, again especially given the small size of the EVAS group. Specifically, the lower 2-year mortality following EVAS among patients with preoperative diameters ≥5.5cm might be the results of comparing an IDE study to the broader EVAR population. This in an important consideration as it is likely to bias against EVAR. However, we performed a sensitivity analysis to only include on-IFU EVAR patients, and though few outcomes lost significance (probably due to the reduced population size), effect sizes remained similar. Furthermore, the IDE cohort is a highly selected group being treated by a highly selected group of high-volume surgeons. However, hospital participation in the VQI is voluntary too, leading to higher proportion of high volume vascular centers.35 Over time, fewer states provide SSDI data due to which survival estimates may become less reliable for the EVAR cohort. However, this would underestimate the EVAS group. Nevertheless, again, this exploratory analysis shows that while this first-generation device has issues due to which it was withdrawn, the underlying concept may hold promise, and a second-generation device with updated IFU-criteria is currently undergoing further study in the EVAS-2 IDE trial. Future investigation with longer follow-up will have to report outcomes of these devices, to assess whether these benefits of sac obliteration persist.

Conclusion

Within the overall population, EVAS was associated with similar 5-year mortality compared with EVAR. However, EVAS was associated with higher mortality in small aneurysms (<5.5cm). For larger aneurysms (≥5.5cm), EVAS was initially associated with lower mortality in the first two years, though this advantage is lost thereafter with higher mortality after two years. Future studies will be required to evaluate the specific causes of death, and to elucidate the beneficial mechanism behind sac obliteration that leads to this potential initial survival benefit. This could help guide development of future grafts with better proximal fixation and sealing, while incorporating sac obliteration.

Supplementary Material

1

ARTICLE HIGHLIGHTS.

Type of Research:

Retrospective 5-year All-Cause Mortality Comparison using prospectively collected data from both the Nellix system investigational device exemption trial and the Vascular Quality Initiative.

Key Findings:

In this retrospective inverse probability-weighted analysis comparing 333 EVAS with 16,497 EVAR patients, EVAS was associated with higher 5-year mortality in smaller aneurysms. However, in aneurysms ≥5.5cm, EVAS was associated with lower mortality in the first two years, though this association rapidly attenuated, losing its significance at 5-years.

Take home Message:

Although EVAS may be associated with lower mortality in the first two years in aneurysms ≥5.5cm, this association attenuated thereafter and is no longer significant at 5-years. Future studies are required to elucidate a potential beneficial mechanism behind aneurysm sac sealing that leads to this potential initial survival benefit.

Acknowledgments

PP is supported by the Harvard-Longwood Research Training in Vascular Surgery NIH T32 Grant 5T32HL007734.

Footnotes

Conflict of Interest:

MR is consultant for Terumo Aortic, Medtronic and W.L. Gore and associates. HV is a consultant of Medtronic, WL Gore, Terumo, Endologix, Philips.

Prior Presentation: This work was presented in an interactive poster session at the 2022 Vascular Annual Meeting of the Society for Vascular Surgery, Boston, MA, June 15–18, 2022.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • 1.Parodi JC, Palmaz JC, Barone HD. Transfemoral intraluminal graft implantation for abdominal aortic aneurysms: Two decades later. Ann Vasc Surg. 1991; [DOI] [PubMed] [Google Scholar]
  • 2.Dua A, Kuy S, Lee CJ, Upchurch GR, Desai SS. Epidemiology of aortic aneurysm repair in the United States from 2000 to 2010. J Vasc Surg [Internet]. 2014;59(6):1512–7. Available from: 10.1016/j.jvs.2014.01.007 [DOI] [PubMed] [Google Scholar]
  • 3.Dansey KD, Varkevisser RRB, Swerdlow NJ, Li C, de Guerre LEVM, Liang P, et al. Epidemiology of endovascular and open repair for abdominal aortic aneurysms in the United States from 2004 to 2015 and implications for screening. J Vasc Surg [Internet]. 2021;74(2):414–24. Available from: 10.1016/j.jvs.2021.01.044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Patel R, Sweeting MJ, Powell JT, Greenhalgh RM. Endovascular versus open repair of abdominal aortic aneurysm in 15-years’ follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial 1): a randomised controlled trial. Lancet [Internet]. 2016;388(10058):2366–74. Available from: 10.1016/S0140-6736(16)31135-7 [DOI] [PubMed] [Google Scholar]
  • 5.De Bruin JL, Baas AF, Buth J, Prinssen M, Verhoeven ELG, Cuypers PWM, et al. Long-term outcome of open or endovascular repair of abdominal aortic aneurysm. Vasc Med. 2010;15(6):515–6. [DOI] [PubMed] [Google Scholar]
  • 6.Lederle FA, Freischlag JA, Kyriakides TC, Matsumura JS, Padberg FT, Kohler TR, et al. Long-Term Comparison of Endovascular and Open Repair of Abdominal Aortic Aneurysm. N Engl J Med. 2012;367(21):1988–97. [DOI] [PubMed] [Google Scholar]
  • 7.Becquemin JP. The ACE trial: A randomized comparison of open versus endovascular repair in good risk patients with abdominal aortic aneurysm. J Vasc Surg [Internet]. 2009;50(1):222–4. Available from: 10.1016/j.jvs.2009.04.074 [DOI] [PubMed] [Google Scholar]
  • 8.Baigent C, Blackwell L, Emberson J, Holland LE, Reith C, Bhala N, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: A meta-analysis of data from 170000 participants in 26 randomised trials. Lancet [Internet]. 2010;376(9753):1670–81. Available from: 10.1016/S0140-6736(10)61350-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Brown LC, Thompson SG, Greenhalgh RM, Powell JT. Incidence of cardiovascular events and death after open or endovascular repair of abdominal aortic aneurysm in the randomized EVAR trial 1. Br J Surg. 2011;98(7):935–42. [DOI] [PubMed] [Google Scholar]
  • 10.Aday AW, Ridker PM. Antiinflammatory Therapy in Clinical Care: The CANTOS Trial and Beyond. Front Cardiovasc Med. 2018;5(June):1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Ridker PM, Danielson EMIA, Fonseca F, Genest J, Gotto AM, Kastelein JJP, et al. Rosuvastatin to Prevent Vascular Events in Men and Women with Elevated C-Reactive Protein. N Engl J Med. 2008;359(21):2195–207. [DOI] [PubMed] [Google Scholar]
  • 12.Ridker PM, Everett BM, Thuren T, MacFadyen JG, Chang WH, Ballantyne C, et al. Antiinflammatory Therapy with Canakinumab for Atherosclerotic Disease. N Engl J Med. 2017;377(12):1119–31. [DOI] [PubMed] [Google Scholar]
  • 13.Sangiorgi G, D’Averio R, Mauriello A, Bondio M, Pontillo M, Castelvecchio S, et al. Plasma levels of metalloproteinases-3 and −9 as markers of successful abdominal aortic aneurysm exclusion after endovascular graft treatment. Circulation. 2001;104(SUPPL. 1). [DOI] [PubMed] [Google Scholar]
  • 14.Lorelli DR, Jean-Claude JM, Fox CJ, Clyne J, Cambria RA, Seabrook GR, et al. Response of plasma matrix metalloproteinase-9 to conventional abdominal aortic aneurysm repair or endovascular exclusion: Implications for endoleak. J Vasc Surg. 2002;35(5):916–22. [DOI] [PubMed] [Google Scholar]
  • 15.Kano M, Nishibe T, Dardik A, Iwahashi T, Ogino H. Association of High-Sensitivity C-Reactive Protein With Aneurysm Sac Shrinkage in Patients Undergoing Endovascular Abdominal Aneurysm Repair. J Endovasc Ther. 2021;6–7. [DOI] [PubMed] [Google Scholar]
  • 16.O’Donnell TFX, Deery SE, Boitano LT, Siracuse JJ, Schermerhorn ML, Scali ST, et al. Aneurysm sac failure to regress after endovascular aneurysm repair is associated with lower long-term survival. J Vasc Surg [Internet]. 2019;69(2):414–22. Available from: 10.1016/j.jvs.2018.04.050 [DOI] [PubMed] [Google Scholar]
  • 17.Thompson MM, Heyligers JM, Hayes PD, Reijnen MMPJ, Böckler D, Schelzig H, et al. Endovascular Aneurysm Sealing: Early and Midterm Results from the EVAS FORWARD Global Registry. J Endovasc Ther. 2016;23(5):685–92. [DOI] [PubMed] [Google Scholar]
  • 18.Carpenter JP, Cuff R, Buckley C, Healey C, Hussain S, Reijnen MMPJ, et al. One-year pivotal trial outcomes of the Nellix system for endovascular aneurysm sealing. J Vasc Surg [Internet]. 2017;65(2):330–336.e4. Available from: 10.1016/j.jvs.2016.09.024 [DOI] [PubMed] [Google Scholar]
  • 19.Berg P, Stroetges RA, Miller LE, Schoefferle J. A Propensity Score-Matched Analysis of Inflammatory Response with Endovascular Aneurysm Sealing vs Endovascular Aneurysm Repair. J Endovasc Ther. 2017;24(5):670–4. [DOI] [PubMed] [Google Scholar]
  • 20.O’Donnell TFX, Carpenter JP, Lane JS, Trani J, Hussain S, Healey C, et al. Endovascular Aneurysm Sealing is Associated with Higher Medium-Term Survival than Traditional EVAR. Ann Vasc Surg. 2020;63(5):145–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Aleksandra CZ, Iris H, Maarten J van der L, de de V, Michel MPJR, Clark JZ. Systematic Review on the Mid-Term Outcomes of Elective Endovascular Aneurysm Sealing in Comparison to Endovascular Aneurysm Repair. J Endovasc Ther. 2021; [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Zoethout AC, Sheriff A, Zeebregts CJ, Hill A, Reijnen MMPJ, Holden A. Survival After Endovascular Aneurysm Sealing Compared With Endovascular Aneurysm Repair. J Endovasc Ther. 2021;28(5):788–95. [DOI] [PubMed] [Google Scholar]
  • 23.van den Ham LH, Holden A, Savlovskis J, Witterbottom A, Ouriel K, Reijnen MMPJ, et al. Editor’s Choice – Occurrence and Classification of Proximal Type I Endoleaks After EndoVascular Aneurysm Sealing Using the Nellix™ Device. Eur J Vasc Endovasc Surg. 2017;54(6):729–36. [DOI] [PubMed] [Google Scholar]
  • 24.Harrison SC, Winterbottom AJ, Coughlin PA, Hayes PD, Boyle JR. Editor’s Choice – Mid-term Migration and Device Failure Following Endovascular Aneurysm Sealing with the Nellix Stent Graft System – a Single Centre Experience. Eur J Vasc Endovasc Surg [Internet]. 2018;56(3):342–8. Available from: 10.1016/j.ejvs.2018.06.031 [DOI] [PubMed] [Google Scholar]
  • 25.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–9. [DOI] [PubMed] [Google Scholar]
  • 26.Nicholls SG, Quach P, von Elm E, Guttman A, Moher D, Petersen I, et al. The REporting of Studies Conducted Using Observational Routinely-Collected Health Data (RECORD) Statement: Methods for Arriving at Consensus and Developing Reporting Guidelines. PLoS One. 2015;10(5):1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Levey AS, Lesley SA, Schmid CH, Zhang Y, Castro AF, Feldman HI, et al. A New Equation to Estimate Glomerular Filtration Rate. Ann Intern Med. 2009;150(9):604–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Desai RJ, Franklin JM. Alternative approaches for confounding adjustment in observational studies using weighting based on the propensity score: A primer for practitioners. BMJ. 2019;367:1–10. [DOI] [PubMed] [Google Scholar]
  • 29.Stenson KM, Patterson BO, Grima MJ, De Bruin JL, Holt PJE, Loftus I. Midterm results of endovascular aneurysm sealing to treat abdominal aortic aneurysm. J Vasc Surg [Internet]. 2019;69(1):53–62.e1. Available from: 10.1016/j.jvs.2018.04.016 [DOI] [PubMed] [Google Scholar]
  • 30.Quaglino S, Mortola L, Ferrero E, Ferri M, Cirillo S, Lario CV, et al. Long-term failure after endovascular aneurysm sealing in a real-life, single-center experience with the Nellix endograft. J Vasc Surg [Internet]. 2021;73(6):1958–1965.e1. Available from: 10.1016/j.jvs.2020.11.029 [DOI] [PubMed] [Google Scholar]
  • 31.Wibmer A, Nolz R, Teufelsbauer H, Kretschmer G, Prusa AM, Funovics M, et al. Complete ten-year follow-up after endovascular abdominal aortic aneurysm repair: Survival and causes of death. Eur J Radiol [Internet]. 2012;81(6):1203–6. Available from: 10.1016/j.ejrad.2011.03.092 [DOI] [PubMed] [Google Scholar]
  • 32.ClinicalTrials.gov NI of HNL of M. Abdominal Aortic Aneurysm Sac Healing and Prevention of Endoleaks (AAA-SHAPE) [Internet]. 2020. [cited 2022 May 26]. Available from: https://clinicaltrials.gov/ct2/show/NCT04227054
  • 33.ClinicalTrials.gov NI of HNL of M. Abdominal Aortic Aneurysm Sac Healing and Prevention of Endoleaks - Netherlands (AAA-SHAPE_NLD) [Internet]. 2021. [cited 2022 May 26]. Available from: https://clinicaltrials.gov/ct2/show/NCT04751578
  • 34.ClinicalTrials.gov NI of HNL of M. AneuFix - Prophylactic Sac Filling. 2020.
  • 35.Dansey KD, de Guerre LEVM, Swerdlow NJ, Li C, Lu J, Patel PB, et al. A comparison of administrative data and quality improvement registries for abdominal aortic aneurysm repair. J Vasc Surg [Internet]. 2021;73(3):874–88. Available from: 10.1016/j.jvs.2020.06.105 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1846087-supplement-1.pdf (151.8KB, pdf)

RESOURCES