Effect of increasing the surface sampled by imprint cytology on the intraoperative assessment of axillary sentinel lymph nodes in breast cancer patients.

As axillary sentinel nodes predict the nodal status and may allow dissection of the axilla on a selective basis we assessed the effects of increasing the surface sampled during intraoperative imprint cytology. Sentinel nodes from 110 patients identified with Patent blue and/or the high radioactivity due to the uptake of 99m-Tc-labeled colloidal albumin were analyzed via hematoxylin and eosin-stained touch preparations. Imprint cytology was performed either on bisected nodes (Protocol One; n = 55) or on sentinel nodes sliced into multiple pieces at 2- to 3-mm intervals (Protocol Two; n = 55). The sentinel nodes were submitted in toto to permanent step sectioning and immunostaining for cytokeratins. There were equal numbers of patients with involved nodes in the two groups assessed. With Protocols One and Two the imprints had sensitivities of 52 and 61 per cent, negative predictive values of 74 and 78 per cent, and false negative rates of 47 and 39 per cent, respectively. No macrometastasis missed by Protocol Two was absent from the surface sampled. These data suggest that increasing the surface sampled improves the proportion of involved sentinel nodes detected intraoperatively by imprint cytology, but a number of metastatic nodes still remain undetected by this method. The sampling of multiple surfaces is encouraged for a more accurate intraoperative assessment.