Aortic Endograft Sizing and Endoleak, Reintervention, and Mortality After Endovascular Aneurysm Repair

Abstract

Aortic endograft sizing for endovascular aneurysm repair (EVAR) is not consistent despite published instructions for use (IFU). By analyzing data obtained from the Global Registry for Endovascular Aortic Treatment (GREAT), we sought to identify factors associated with oversizing and undersizing; to determine sex influence on sizing; and to examine sizing effects on endoleak, reintervention, and mortality. All patients enrolled in GREAT undergoing EVAR were included for analysis. Proximal and distal aortic landing zones were compared with device implanted to assess sizing as related to IFU. The χ2 and Fisher exact tests were used to evaluate associations between IFU sizing and demographics. Logistic regression modeling was used to identify predictors of outside IFU sizing. Cox proportional hazards regression analyzed the relationship between sizing and endoleak, device-related reinterventions, and all-cause/aortic mortality. There were 3607 participants enrolled in GREAT as of March 2020. Of these, 1896 (53%) were within IFU for sizing, 791 (22%) were oversized, 540 (15%) were undersized, and 380 (10%) had both oversized and undersized components. Factors predictive of use outside of IFU included female sex (P = .001), nonwhite race (P = .0003), decreased proximal neck length (P < .001), and smaller proximal aortic (P < .0001) or larger iliac diameters (P < .0001). Women were more likely than men to have proximal neck undersizing and iliac limb oversizing, and men were more likely to have iliac limb undersizing. On multivariate analysis, undersizing of the proximal graft was associated with endoleak (hazard ratio [HR], 1.8) and aortic (HR, 65.7) and all-cause (HR, 18.0) mortality. Undersizing of iliac limbs was associated with endoleak (HR, 1.5) and device-related reintervention (HR, 1.3). Proximal and distal oversizing was not associated with adverse outcomes. Female sex was associated with mortality on univariate but not on multivariate analysis. Women undergoing EVAR with Gore Excluder endoprosthesis (W. L. Gore & Associates, Flagstaff, Ariz) are more likely to have proximal stent graft undersizing and iliac limb oversizing, whereas men are more likely to have undersized iliac limbs. Proximal aortic graft undersizing is associated with endoleak and all-cause/aortic mortality, whereas undersizing of iliac limbs is associated with endoleak and device-related reintervention.