Background
Abdominal aortic aneurysm (AAA) is a dilation of the abdominal aorta associated with
an increased risk of rupture. Two-dimensional ultrasound (2D-US) is the first-line
method for diagnosis and follow-up. Volume-based measurements using three-
dimensional ultrasound (3D-US) have been proposed as a potential complement,
although knowledge regarding the reliability of the method remains limited, particularly
concerning inter- and intra-observer reliability.
Aim
The aim was to investigate inter- and intra-observer reliability in volume-based
measurements of AAA using 3D-US, and whether agreement was affected by aneurysm
characteristics or differed between 2D- and 3D-based measurements.
Method
The study was conducted as a quantitative cross-sectional study based on previously
collected ultrasound material from the Oral and Vascular Health (OVH) research project.
A total of 25 patients with AAA were included. Two observers independently performed
measurements on two occasions. Inter- and intra-observer reliability were analyzed using
intraclass correlation coefficient (ICC) and Bland-Altman analysis. Linear mixed models
(LMM) were used to investigate the impact of aneurysm characteristics and differences
between 2D- and 3D-based measurements.
Results
ICC values for both inter- and intra-observer reliability were high. The Bland-Altman
analysis showed a low mean difference between observers, although a slight systematic
difference was observed. The LMM analysis showed that aneurysm size, morphology,
and location significantly affected measurement variability, whereas direction and
measurement occasion did not. No statistically significant difference was found between
2D- and 3D-based volume calculations.
Conclusion
The results demonstrated excellent inter- and intra-observer reliability for volume-based
measurements using 3D-US. The method showed good reproducibility, although intra-
observer results may have been influenced by memory effects. The validity of the
method could not be established; therefore, further studies are needed to evaluate clinical
applicability.