Pre-operative Neck Growth and Size are Associated with Type 1a Endoleak after Endovascular Aneurysm Repair

INTRODUCTION: Type 1a endoleak (T1a) is seen in up to 15% of endovascular aortic aneurysm repair (EVAR) within instructions for use (IFU). Involvement of neck by abdominal aortic aneurysm (AAA) pathology may increase risk of T1a although current methods based on diameter/volume, may miss slow/sub-maximal pre-op growth in the neck. Vascular Deformation Mapping (VDM) is an emerging method for quantifying aortic wall growth in 3D with sub-millimeter accuracy. We sought to determine the association of pre-EVAR neck growth and size with T1a using VDM. METHODS: We identified patients with infra-renal AAA who underwent EVAR between 2010-2024. Patients were included if they had ≥2 pre-operative CTAs completed ≥6 months apart with the last CTA being ≤3 months of EVAR. Patients with non-contrast CT scans, off-IFU, and those treated with prophylactic endo-anchors were excluded. Patient sex and device type were collected. AAA neck diameters and volumes were measured from both pre-operative CTAs using PRAEVAorta (Nurea, Bordeaux, France) and changes over time were calculated. VDM, which involves a multi-step deformable image registration, was used to quantify 3D-growth (3D- defined as degree of deformation in the direction normal/perpendicular to the aortic wall) and 3-D size (defined by local centerline radius) of the aneurysm neck. Statistical shape modeling was utilized to compare mean 3D-growth and 3D-size of the neck and four neck segments (anterior, posterior, left lateral, and right lateral) between patients with and without T1a. RESULTS: The cohort included 97 patients (90.7% male, average age: 73.0±9.6 years). Twenty-one patients (21.6%) developed T1a. T1a endoleak was not associated with patient demographics, neck diameter, neck length, neck volume, or changes in neck diameter and/or volume (p>0.05, for all). However, patients with T1a had larger 3D-growth in the aneurysm neck (3.3±0.8 vs 2.4±0.6 mm/year, p<0.001) and across all neck segments (p<0.001, for all). Receiver operating characteristic analysis identified a mean neck 3D-growth ≥ 2.7 mm/year as a predictor of T1a (area under the curve=0.78, sensitivity=0.81, specificity=0.76). Neck 3D-size was larger in T1a patients compared to controls (12.5 ± 0.4 vs 11.5 ± 0.6 mm, p<0.001) and across all neck segments (p<0.001, for all). ROC analysis identified an optimal cut-point for neck 3D-size of 11.9 mm as a predictor of T1a (AUC: 0.85, Sensitivity: 86%, Specificity: 84%). CONCLUSIONS: Pre-operative 3D-growth and 3D-size of the aneurysm neck are strong predictors of T1a, despite devices being on-IFU. VDM can provide important information about aneurysm growth that is not captured via current standard methods. Pre-operative aneurysm neck growth/size should be considered for optimal EVAR planning.