Conical neck is strongly associated with proximal failure in standard endovascular aneurysm repair

Objective: Hostile proximal aortic neck (HN) challenges the suitability for standard endovascular aneurysm repair (EVAR) of patients at high risk for "open" repair. However, there has been little if any focus placed on the individual role of the "nonlength" HN features in EVAR outcomes. The aim of this study was to evaluate their individual and potentially predictive role in outcomes of EVAR under HN conditions. Methods: Data of 156 consecutive EVAR patients with short (<15 mm) HN, treated with the Endurant device (Medtronic Cardiovascular, Santa Rosa, Calif) at three European academic vascular centers between 2007 and 2015, were collected and retrospectively analyzed. All patients had at least one of the four well-known nonlength HN criteria (width >32 mm or bulge, angulation >60 degrees, reverse taper anatomy, and circumferential thrombus or calcification >50%) and underwent standard EVAR without additional techniques, such as use of chimney grafts or endoanchors. Primary end points were absence of type IA endoleak at 1 month and midterm follow-up and aneurysm sac stabilization or shrinkage. Secondary end points were 30-day mortality, overall survival, and secondary interventions related to EVAR. The study cohort was classified in two subgroups related to neck length (length <10mmand length between 10 and 14 mm) as well as in two subgroups according to on-label or off-label stent graft use. Results: Mean clinical and radiologic follow-up was 41.1 +/- 24.7 and 31.7 +/- 19.0 months, respectively. Overall EVAR-related mortality was 1.9% (n = 3). The total type IA endoleak rate was 5.8% (n = 9). In four patients, the type IA endoleak was detected intraoperatively and solved by endovascular means. A type IA endoleak was detected in three patients at 1month and in two patients at 2-year follow-up. During follow-up, five patients showed an increase of aneurysm diameter due to type II endoleak and were treated by secondary endovascular reinterventions. The total number of all EVAR-related secondary procedures in the midterm was 12 (7.7%). Univariate analysis showed that the center of treatment and the clinical or anatomic features were not associated with adverse outcomes. Multiple regression and Cox regression analysis of HN features revealed that reverse taper anatomy (conical neck) was the single and significantly associated predictor of proximal EVAR failure (P < .012). Width > 32 mm, angulation > 60 degrees, and calcification or thrombus were not associated with adverse outcomes. Analysis between HN length cohorts and between on-label and off-label subgroups revealed no difference in outcomes. Conclusions: A conical neck in hostile anatomies represents the single strongest factor associated with proximal failure of standard EVAR. This finding should be considered and highlighted apart from the length of the infrarenal neck to prevent midterm failure of standard EVAR.