Three-dimensional intravascular ultrasound assessment of abdominal aortic aneurysm necks

Purpose: To document the accuracy of an automated analysis system for measu ring lumen diameter and neck lengths of abdominal aortic aneurysms (AAAs) f rom intravascular ultrasound (IVUS) images and to describe additional featu res associated with 3-dimensional (3D) IVUS imaging. Methods. Twenty-two aortic aneurysms were studied with IVUS. Lumen diameter s obtained using the automated analysis system were compared with manual me asurements from axial IVUS scans, as were neck lengths obtained using autom ated analysis versus those measured with the aid of a displacement sensing device. Automated analyses were repeated by a second observer Agreement was expressed as the coefficient of variation (CV). Results: Twenty proximal aortic, 6 distal aortic, and 3 iliac necks were av ailable for analysis. Comparison between automated analysis and manual meas urements for lumen diameter revealed a difference of 0.45 +/- 0.42 mm (mean +/- SD, Pearson's r = 0.99, p < 0.001, CV = 2.1%) and a difference of 0.05 +/- 0.12 cm (r = 0.99, p = 0.04, CV = 4.1%) for neck length. Interobserver difference for lumen diameter was 0.13 +/- 0.66 mm (r = 0.99, p < 0.001, C V = 3.4%) and 0.05 +/- 0.11 cm for length measurements (r= 0.99, p = 0.02, CV = 3.5%). The 3D IVUS imaging facilitated the identification of neck conf iguration. Conclusions: Automated analysis of IVUS images allows accurate measurement of the lumen diameter of proximal and distal AAA necks and gives length mea surements comparable to those of manual analysis. Longitudinal display of I VUS images aids in the elucidation of neck anatomy.