Is the Axilla Really a Blind Spot for Automated Breast Ultrasound? A Comparative Multimodality Study

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background: Automated Breast Ultrasound (ABUS) is primarily used for standardized volumetric breast imaging but has been widely considered inadequate for axillary assessment due to technical constraints. This study aims to quantitatively evaluate this assumption by comparing the axillary lymph node (ALN) detection performance of ABUS with that of Mammography (MG), Digital Breast Tomosynthesis (DBT), and Magnetic Resonance Imaging (MRI).Methods: This retrospective, single-center study screened 339 women undergoing imaging between May 2023 and December 2025; after excluding 35 patients (imaging interval >14 days or insufficient data), 304 valid axillary evaluations were included in the analysis. Patients were categorized into three cohorts based on paired imaging of the same axilla within a two-week interval: MG-DBT-ABUS (n = 304), ABUS-MRI (n = 51), and a comprehensive MG-DBT-ABUS-MRI subset (n = 16). A specialized ABUS axilla protocol involving arm hyper-abduction was utilized. ALN counts were compared using paired samples t-tests and Wilcoxon Signed-Rank tests.Results: In the MG-DBT-ABUS group, ABUS detected significantly more lymph nodes (mean 4.52 ± 1.80; median 4.00) compared to the median values of 0.00 (interquartile range (IQR): 0.0–2.0) for both MG and DBT (p < 0.001). In the ABUS-MRI group, MRI detected significantly more nodes (6.51 ± 2.70) than ABUS (4.41 ± 1.94) (p < 0.001). However, ABUS visualized approximately 68% of the lymph nodes detected by MRI. In the four-modality group (n = 16), non-parametric testing of median detection hierarchy demonstrated that MRI [6.00 (IQR: 4.0–8.0)] and ABUS [5.00 (IQR: 4.0–5.5)] were significantly superior to DBT [0.00 (IQR: 0.0–2.0)] and MG [0.00 (IQR: 0.0–1.0)]; no direct statistical comparison between DBT and MG was performed in this subset.Conclusion: ABUS is not a blind spot for the axilla. While MRI remains the current imaging reference standard, ABUS demonstrates superior quantitative detection performance compared to mammography and tomosynthesis. When performed with a dedicated protocol, ABUS serves as a valuable, non-ionizing supplementary tool for axillary lymph node assessment. While this study primarily evaluates quantitative detection performance, pathological correlation and morphological characterization warrant further investigation in future studies.