Agreement between computed tomography and ultrasound on abdominal aortic aneurysms and implications on clinical decisions.

ABSTRACT

OBJECTIVES: The United Kingdom abdominal aortic aneurysm (AAA) screening programme refers aneurysms with ultrasound (US) diameters of ≥5.5 cm to vascular services for consideration of computed tomography (CT) and intervention. We investigated the discrepancy between US and CT, implications on clinical decisions and question at which stage CT be used. DESIGN/METHODS: AAA USs over 5 years were retrospectively analysed. Patients included had aneurysms measuring ≥5 cm on US with subsequent CT within 2 months (n = 123). Based on maximum US diameters, 44 patients had aneurysms between 5 and 5.4 cm (group I) and 79 patients ≥5.5 cm (group II). Results were cross-referenced. Correlation and limits of agreement were calculated. Two radiologists re-measured 44 pairs of CT/US scans and the inter-observer bias in determining discrepancies between imaging modalities calculated. RESULTS: Mean difference between imaging modalities was 0.21 cm (±0.39 cm, p < 0.001). Limits of agreement were -0.55 to 0.96 cm, exceeding clinical acceptability. Mean difference was higher and significant in group I (0.39 cm, p < 0.001) compared to group II (0.10 cm, p > 0.05). Seventy-percent of group I patients had CT scans revealing diameters of ≥5.5 cm. Inter-observer bias was not significant. CONCLUSION: Significant differences between imaging modalities, more in US diameters of below 5.5 cm, exist. We recommend AAAs measuring ≥5 cm on US should undergo earlier referral to a vascular service and CT.