Touch preparation or frozen section for intraoperative detection of sentinel lymph node metastases from breast cancer.

BACKGROUND: The preferred technique for intraoperative evaluation of the sentinel lymph node has not been determined. The purpose of this study was to compare the sensitivity and accuracy of intraoperative evaluation of the sentinel lymph node by touch preparation cytology and frozen section.

METHODS: A total of 117 patients with clinically node-negative breast cancer or ductal carcinoma-in-situ undergoing sentinel lymph node biopsy had intraoperative evaluation of the sentinel node by touch preparation, frozen section, or both. The results of the intraoperative evaluation were compared with the final histological results of hematoxylin and eosin (H&E) paraffin section and immunohistochemistry (IHC).

RESULTS: Twenty-six (57%) of the 46 patients with nodal involvement had metastases detected during surgery. The sensitivity of touch preparation for detecting macrometastases was 78%; for detecting all H&E metastases, including micrometastases, was 57%; and for detecting all metastases, including those seen on IHC, was 40%. The sensitivity of frozen section for detecting macrometastases was 83%; for detecting all H&E metastases, including micrometastases, was 78%; and for detecting all metastases, including those seen on IHC, was 64%. Both have a low sensitivity for micrometastases seen by H&E paraffin section: 57% and 78%, respectively. Neither detected micrometastases diagnosed by IHC only.

CONCLUSIONS: Both touch preparation and frozen section seem to be accurate in detecting macrometastases, but not micrometastases. Intraoperative evaluation of the sentinel lymph node by touch preparation allows for a quick evaluation of the node without wasting significant tissue and without detecting occult microscopic metastases, which may be beneficial because the clinical importance of these has yet to be elucidated.