Intraoperative touch imprint and frozen section analysis of sentinel lymph nodes after neoadjuvant chemotherapy for breast cancer

Objective: To evaluate the accuracy of touch imprint and frozen section analysis of sentinel nodes after neoadjuvant chemotherapy. Summary Background Data: Intraoperative evaluation of the sentinel node can determine the need for axillary dissection at the time of initial operation and therefore spare the patient a second operation. Little data, however, exists on the accuracy of intraoperative evaluation of sentinel nodes after neoadjuvant chemotherapy. Methods: A retrospective chart review was performed of all sentinel node procedures for breast cancer from 2004 to 2008 at a single institution. The sentinel node procedure was done before (no-NACT) chemotherapy in 107 patients and 37 had the procedure after neoadjuvant (NACT) chemotherapy. Intraoperative analysis of sentinel nodes was performed using touch imprint and frozen section techniques. Results: In the no-NACT group, intraoperative assessment by touch imprint analysis had 61% sensitivity, 100% specificity, and 87% accuracy. Frozen section analysis was similar with 74% sensitivity, 100% specificity, and 90% accuracy. In the NACT group, touch imprint analysis had 79% sensitivity, 100% specificity, and 90% accuracy. Frozen section analysis was again similar with 74% sensitivity, 100% specificity, and 83% accuracy. When the no-NACT group and the NACT group were compared, both frozen section and touch imprint analysis had similar sensitivity, specificity, and accuracy. Conclusions: Intraoperative evaluation of sentinel nodes with touch imprint and frozen section analysis in patients treated with neoadjuvant chemotherapy showed acceptable sensitivity, specificity, and accuracy. Nearly all of the misses in intraoperative evaluation were in patients with micrometastases.