Re-evaluating the efficacy of axillary ultrasound in the detection of nodal metastasis and its impact on clinical practice.

Abstract

4 Background: Recent reports have indicated a lack of overall survival benefit for axillary node dissection versus sentinel lymph node biopsy in early breast cancer. To study this further, we wished to assess the accuracy and effectiveness of ultrasound guided fine needle aspiration (FNA) cytology in detecting lymph node involvement in breast cancer patients, in order to refine and evaluate our current clinical pathways as newly diagnosed invasive breast cancer patients routinely undergo pre-surgical axillary ultrasound. Methods: An FNA was taken from nodes of consecutive patients, which appeared abnormal on ultrasonography based on size, morphology, fatty hilum and cortical thickness measurements. Ultrasound and FNA cytological findings were correlated with histology following axillary node dissection. Of 260 cases, 123 (47.3%) had metastatic nodal involvement. Of these cases, only 66 (53.7%) were reported as positive on US findings. Results: The overall positive predictive value (PPV) of ultrasound for detecting metastatic nodal involvement measured 0.82, and the negative predictive value (NPV) was 0.60. The sensitivity was 0.54, specificity measured 0.85 and the accuracy was 0.68. The ultrasound morphological nodal features with the greatest correlation with malignancy were absence of a fatty hilum (p=0.003) and an increased cortical thickness (p=0.03). Cases with a metastatic nodal burden density of a least 20% were also more likely to be detected as abnormal on axillary ultrasound. (p=0.009). Conclusions: Axillary ultrasound has a low NPV and negative sonographic results do not exclude node metastases with sufficient sensitivity in most cases, to justify its routine clinical use. [Table: see text]