Reliability of intraoperative frozen section and imprint cytological investigation of sentinel lymph nodes in breast cancer

Aims: The sentinel lymph node procedure enables selective targeting of the first draining lymph node, where the initial metastases will form. A negati ve sentinel node (SN) predicts the absence of tumour metastases in the othe r regional lymph nodes with high accuracy. This means that in the case of a negative SN, regional lymph node dissection is no longer necessary Besides saving costs, this will prevent many side-effects of lymph node dissection . The aim of this study was to evaluate the reliability of intraoperative c ytological and frozen section investigation of the SN to detect metastases. This would allow the axillary lymph node dissection to be performed in the same session as the SN procedure and the excision of the primary tumour in case of a positive SN. Methods and results: Seventy-four SNs were detected by gamma probe detectio n of nanocolloid and visual localization of Patent Blue accumulations in 54 women with stage T1-2N0M0 invasive breast cancer. The identified SN were i mmediately investigated by frozen section and imprint cytological investiga tion. Diagnoses were confirmed on the paraffin material, and in case of neg ative frozen section and paraffin haematoxylin and eosin sections, skip sec tions and immunohistochemistry were performed. Thirty-one SNs (42%) contain ed metastases, of which 27 were detected by the frozen section procedure (s ensitivity 87%). There were no false positives (specificity 100%). The sens itivity of the imprints was 62% with a specificity of 100%. When evaluating the data per patient, for the frozen section procedure the sensitivity was 91% and the specificity 100%, and for the imprints, the sensitivity was 63 % and the specificity 100%. There were no SNs in which the imprints showed metastases and the frozen section did not. Conclusions: Intraoperative frozen section analysis is a reliable procedure by which a high percentage of sentinel lymph node metastases can be detect ed in breast cancer patients without false positive results. This allows th e surgeon to perform an immediate axillary lymph node dissection in case of positive SNs. In up to 10% of cases, the final paraffin sections will reve al micrometastases that were not detected by the frozen section, and in the se patients axillary lymph node dissection will have to be performed in a s econd session. The imprint method is significantly less sensitive than the frozen section but may be used as an alternative when frozen section is not possible.