OBJECTIVETo analyze the ultrasonographic features of axillary lymph node metastasis in breast cancer patients, and to evaluate the accuracy of these features by Receiver Operating Characteristic (ROC) curve.

METHODSUltrasonography was carried out in 113 patients (mean age 56.5 yr, range 29 - 77 yr). The ultrasonographic features of the primary tumor and the axillary lymph node were observed. Univariate analysis was performed by Chi-squared test and Fisher exact test, and multivariate analysis to determine independent significant individual variables by multiple logistic regression analysis. ROC curve analysis was done to determine the sensitivity and specificity of individual and combined ultrasonographic features in distinguishing metastatic lymph node from the normal one.

RESULTSThe primary tumor size, abundance of blood supply in the primary tumor, longitudinal-transverse diameter ratio, cortex-hilum thickness ratio (on the longitudinal section), abnormal cortex thickness, distribution of intra-nodal vascularity of axillary lymph nodes were found to be statistically significant factors by univariate analysis. The primary tumor size, longitudinal-transverse ratio and cortex-hilum thickness ratio of lymph node were proved to be significant independent predictors of axillary lymph node metastasis by logistic regression analysis. Through ROC analysis, the combination of these independent ultrasonographic feature predictors was found to contribute significantly in differentiating metastatic lymph node from the normal with a sensitivity of 88.6% and specificity of 84.5%.

CONCLUSIONAxillary ultrasonography is helpful in staging the axillary lymph node in breast cancer patient. The primary tumor size, longitudinal-transverse diameter ratio and cortex-hilum thickness ratio of lymph node are the main features to determine whether metastatic involvement is present or not.