Measurement of maximum diameter of native abdominal aortic aneurysm by angio-CT: reproducibility is better with the semi-automated method.

OBJECTIVES: We sought to identify the technique yielding the best reproducibility from among various measures of native maximum abdominal aortic aneurysm (AAA) diameter with computed tomography angiography (CTA). METHODS: Ten parameters of maximum diameter in 68 native AAA were measured double-blind by three radiologists on orthogonal planes, curved multiplanar reconstructions, and, finally, using semi-automated software. The semi-automated software creates the AAA lumen centreline and automatically provides cross sections perpendicular to this centreline. The maximum diameter in any direction is automatically calculated once the slice of interest has been selected. Intra- and inter-observer reproducibility and discordance >5 mm were analysed. RESULTS: Intra-observer reproducibility was high. The limits of agreement were within the clinically accepted range [-5; +5 mm] in 27/30 (90%) comparisons. The method common to all three observers that yielded the lowest values was the semi-automated method. Inter-observer reproducibility was poorer. The limits were outside the clinically accepted range in 26/30 (87%) comparisons. The semi-automated method led to lower intra- (0%) and inter-observer (5.88%) discordances rates. CONCLUSION: Even using precise methodology, the reproducibility of maximum diameter measurements of native AAA on CTA may exceed recommended thresholds. The semi-automatic method yielded the lower discordance rates and provided a more relevant anatomical approach for measuring the maximum diameter of an AAA.