Touch imprint cytology and immunohistochemistry for the assessment of sentinel lymph nodes in patients with breast cancer

Aims: Accurate evaluation of sentinel nodes is of clinical importance to avoid further surgery for axillary node dissection. A prospective study was carried out to investigate the feasibility and accuracy of touch imprint cytology (TIC) and touch imprint immunohistochemistry (TIHC). Methods: Two hundred and five sentinel nodes from consecutive 118 patients with primary breast cancer were studied after successful identification of sentinel nodes. Sentinel nodes were sectioned at 2 mm intervals and imprint specimens prepared from all cut surfaces were subjected to Papanicolaou staining and immunohistochemical staining using anti-cytokeratin antibody. Results: Forty-nine sentinel nodes from 40 patients were positive by permanent section. The sensitivity of TIC was 84% (41/49) per sentinel node and 83% (33/40) on a per patient basis. The sensitivity of TIHC was 86% (42/49) per sentinel node and 83% (33/40) on a per patient basis. When the results of TIC and TIHC were combined, the sensitivity was 88% (43/49) per sentinel node and 85% (34/40) on a per patient basis. Among the 156 negative sentinel nodes, four sentinel nodes from four different patients were consistently positive by TIC and TIHC, but only one patient out of 78 node-negative patients was upstaged. Conclusions: Touch imprint cytology is sufficiently sensitive for intraoperative evaluation of sentinel nodes. A slight improvement in the sensitivity is expected when immunohistochemistry is used. The combination of these methods provides better sensitivity than either method alone. (C) 2003 Elsevier Ltd. All rights reserved.