Multiple sectioning and immunohistochemical staining of sentinel nodes in patients with breast cancer

Abstract

Background The aim of the present study was to investigate whether focused analysis of sentinel nodes is more useful than routine haematoxylin and eosin examination of axillary lymph nodes obtained by axillary lymph node dissection. Methods One hundred and fifty-two patients with breast cancer with clinically negative axillary nodes underwent successful sentinel node biopsy using a combination of dye and radioisotope, followed immediately by standard level I and II axillary lymph node dissection. Multiple sectioning, with haematoxylin and eosin and immunohistochemical analysis of sentinel nodes using cytokeratin antibody, was compared with single section and haematoxylin and eosin analysis of sentinel and non-sentinel nodes (routine examination). Results A mean of 1·9 (range 1–12) sentinel nodes and 11·2 (range 4–24) non-sentinel nodes were excised per patient. Metastases were detected in 44 patients (29 per cent) by single section and haematoxylin and eosin analysis of sentinel and non-sentinel nodes. An additional five patients (3 per cent) with metastases were detected by multiple sectioning and haematoxylin and eosin analysis of sentinel nodes. A further 20 patients (13 per cent) with metastases were identified by multiple sectioning and immunohistochemical analysis of sentinel nodes. Both haematoxylin and eosin and immunohistochemical analysis of sentinel nodes missed one patient with node metastases, which led to a false-negative rate of 1 per cent. Conclusion Multiple sectioning and immunohistochemical staining of sentinel nodes identified 16 per cent more patients with positive axillary lymph nodes than routine haematoxylin and eosin examination.