The incidence and effect of noncylindrical neck morphology on outcomes after endovascular aortic aneurysm repair in the Global Registry for Endovascular Aortic Treatment

Abstract

BackgroundThe Gore Global Registry for Endovascular Aortic Treatment (GREAT) was designed to evaluate real-world outcomes after treatment with Gore aortic endografts used in a real-world, global setting. We retrospectively analyzed the GREAT data to evaluate the incidence and effects of noncylindrical neck anatomy in patients undergoing endovascular aortic aneurysm repair. MethodsThe present analysis included patients with data in the GREAT who had been treated with the EXCLUDER endograft from August 2010 to October 2016. A noncylindrical neck was defined when the proximal aortic landing zone diameter had changed ≥2 mm over the first 15 mm of the proximal landing zone, indicating a tapered, conical, or hourglass morphology. Cox multivariate regression analyses were performed for any reintervention (including reinterventions on aortic branch vessels), device-related reinterventions, and reintervention specifically for endoleak. Independent binary (cylindrical vs noncylindrical necks) and continuous (percentage of neck diameter change) variables were assessed. The abdominal aortic aneurysm (AAA) diameter, proximal neck length, maximal infrarenal neck angle, gender, and use of aortic extender cuffs were also assessed. ResultsOf 3077 GREAT patients with available proximal aortic landing zone diameter measurements available, 1765 were found to have cylindrical necks and 1312 had noncylindrical necks. The noncylindrical neck cohort had a significantly greater proportion of women (17.4% vs 12.6%; P < .001) and more severe infrarenal angulation (33.8° vs 28.4°; P < .001). A total 14.7% of noncylindrical neck patients and 11.2% cylindrical neck patients underwent implantation outside of the EXCLUDER instructions for use regarding the anatomic inclusion criteria (P = .004). The procedural characteristics were similar between the two cohorts; however, noncylindrical neck patients required significantly more aortic extender cuffs (P = .004). The average follow-up was 21.2 ± 17.5 months and 17.8 ± 15.8 months for the cylindrical and noncylindrical cohorts, respectively (P < .001). The Cox multivariate regression models demonstrated female gender and maximum AAA diameter were significant risk factors for subsequent reintervention (overall, device-related, and endoleak-specific). Women were 2.2 times as likely to require device-related intervention during the follow-up period compared with men (P < .001). Neck shape morphology was not a significant predictor, except for device-related intervention, for which cylindrical necks (binary definition) resulted in a slightly elevated risk (1.5 times; P = .03). ConclusionsNoncylindrical neck morphology was more common in women and was associated with an increased use of aortic extender cuffs but did not increase the risk of intervention. Female gender and AAA diameter were associated with an increased need for reintervention.