Spatial Distribution of Abdominal Aortic Aneurysm Surface Growth and Correlation with Diameter and Volume Expansion

Abstract

Introduction - The management of Abdominal Aortic Aneurysms (AAAs) is currently based on maximum size and growth rate criteria. Large or rapidly growing AAAs are thought to be at an increased risk of rupture, thus being suitable for intervention. Size and growth are determined based on the maximum diameter alone. Nevertheless, this variable takes into account a single portion of the vessel, while expansion is non-uniform, characterized by significant spatial variability. We aim to quantify regional surface growth of a cohort of AAAs. Methods - This is a single institution study including 31 lesions. Each AAA should have at least one baseline and one follow-up CT scan. From those, 3-Dimensional AAA models were created by semi-automated segmentation using dedicated software. Maximum orthogonal diameter and total aneurysm volume were recorded and used to calculate annual growth rate. Regional surface growth was quantified using the VascForm algorithm, which is written in Matlab, adapted to aneurysm follow-up studies and has been recently validated.[1] In general it is based on non-rigid point cloud registration and the iterative closest point algorithm. Initially, surfaces are pre-aligned applying principal component analysis and a general deformation of the Source surface is performed in order to best match the Target. Then, the surfaces are finely matched through a non-rigid local deformation model and each point in the Source surface is placed closer to the Target surface. After the registration, the surface growth distribution is the discretized surface element growth distribution. Maximum and average surface growth were calculated and correlated with the diameter and volume growth rate. The Spearman rho correlation coefficients were calculated and the statistical significance was tested using a two-tailed test. In order to identify potential correlation between intraluminal thrombus deposition (ILT) and surface growth, the co-localization of maximum surface growth with initial ILT maximum thickness was examined. Results - Median annual surface growth was 6.7%(0%>68%) and maximum surface growth was 26%(12%>444%). Correlations between surface and diameter/volume growth are summarized in Table 1. In 48% of follow-ups (15/31), maximum surface growth occurred away from the site of maximum diameter and thus could not have been captured by the measurement of maximum diameter growth. Sixteen cases presented maximum surface growth away and fifteen at the region of maximum initial ILT thickness. AAAs in the former group had significantly lower initial ILT thickness (11.3 vs 19.5mm, p<0.001) compared to those in the latter.Tabled 1DmaxVolumeMaxSurfaceCorrelationCoefficient,465,544Sig. (2-tailed),014,009AvgSurfaceCorrelationCoefficient,389,866Sig. (2-tailed),045,000 Open table in a new tab Conclusion - Regional surface growth of AAAs is not uniform. Average and maximum surface growth are related to the diameter growth but the correlation is only moderate. Volume growth presents very strong correlation with average but only moderate correlation with maximum surface growth rate. Almost half of the lesions display maximum surface growth at a site different from that of maximum diameter. Regional growth is related to initial local amount of ILT. Surface growth provides a means to evaluate spatial distribution of AAA expansion which may be of value for rupture risk predictions. References Metaxa E, Iordanov I, Maravelakis E, Papaharilaou Y. A novel approach for local abdominal aortic aneurysm growth quantification. Medical & Biological Engineering. DOI 10.1007/s11517-016-1592-8.