Intraoperative examination of axillary sentinel lymph nodes in breast carcinoma patients

BACKGROUND: Routine histologic examination of axillary sentinel lymph nodes predicts axillary lymph node status and may spare patients with breast carcinoma axillary lymph node dissection. To avoid the need for two separate surgical sessions, the results of sentinel lymph node examination should be available intraoperatively. However, routine frozen-section examination of sentinel lymph nodes is liable to yield false-negative results. This study was conducted to ascertain whether extensive intraoperative examination of sentinel lymph nodes by frozen section examination would attain a sensitivity comparable to that obtained by routine histologic examination without intraoperative frozen section examination. METHODS: In a consecutive series of 155 clinically lymph node negative breast carcinoma patients, the axillary sentinel lymph nodes were examined intraoperatively, before complete axillary lymph node dissection. The frozen sentinel lymph nodes were sectioned subserially at 50-microm intervals. For each level, one section was stained with hematoxylin and eosin and the other section immunostained for cytokeratins using a rapid immunocytochemical assay. RESULTS: Sentinel lymph node metastases were detected in 70 of the 155 patients (45%). In 37 cases the sentinel lymph nodes were the only axillary lymph nodes with metastases. Immunocytochemistry did not increase the sensitivity of the examination. Five patients had metastases in the nonsentinel axillary lymph nodes despite having negative sentinel lymph nodes. The general concordance between sentinel and axillary lymph node status was 96.7%; the negative predictive value of intraoperative sentinel lymph node examination was 94.1%. CONCLUSIONS: The intraoperative examination of axillary sentinel lymph nodes is effective in predicting the axillary lymph node status of breast carcinoma patients and may be instrumental in deciding whether to spare patients axillary lymph node dissection.