Purpose: The aim of this study was to evaluate the diagnostic accuracy of axillary lymph node sonography, if necessary in combination with US-guided large core biopsy, in the preoperative evaluation of breast cancer patients scheduled for quadrantectomy and sentinel lymph node excision.
Materials and methods: From July 2001 to December 2002, we evaluated 117 breast cancer patients with ultrasound and, where indicated, FNAB. Breast lesions had diameters between 4 and 26 mm (mean diameter 11 mm). Fifteen (13%) of the 117 patients were excluded from the series as they did not fulfil the criteria for preliminary sonography of the axilla: in 9 patients fewer than 4 nodes were detected and in 6 patients the breast lesions were intraoperatively found to be benign. Eleven patients (10.7 %) with sonographically suspicious axillary nodes were sampled by US-guided core biopsy using a 14 or 16 Gauge Tru-Cut needle.
Results: The ultrasound study aims to evaluate the dimensions and morphology of the breast lesion as well as detect and assess at least 4 axillary nodes. These were evaluated for hilar and cortical thickening and ratio between the sinus diameter and the total longitudinal diameter. Lymph nodes with hilar diameters equal to or greater than 50% of the longitudinal diameter were considered normal. Of the 102 patients evaluated, 77 (75.7%) had normal axillary nodes according to the US criteria adopted. Negativity was confirmed by histology in 56 cases (72.7%, true negative); 21 (27.3%, false negative) were found to be positive, in contrast with the sonographic appearance. The false negative cases were due to lymph node micrometastasis which probably did not cause morphologic alterations perceptible at ultrasound. The remaining 25 patients (24.5%) had axillary lymph nodes classified as suspicious. In 13 cases of (52%, true positive) there was agreement with histology, whereas in 12 cases (48%, false positive) the US suspicion was not confirmed at surgery. The most important sonographic alteration was the gradual reduction in hilar echogenicity (seen in 100% metastatic nodes); conversely, hilar denting or irregularities, as well as dimensional criteria, proved to be poorly specific.
Conclusions: The sonographic assessment of axillary and internal mammary chain nodes, possibly in addition to core biopsy, improves the preoperative evaluation of breast cancer patients scheduled for conservative surgery of the breast (quadrantectomy) and the axilla (sentinel node biopsy). US findings suspicious for metastatic involvement of axillary lymph nodes should be considered as an exclusion criterion for sentinel node biopsy. A more widespread use of axillary node biopsy and an accurate sonographic evaluation of the excised lymph node increase the specificity of the procedure, allowing a better correlation between sonographic findings and definitive histology.