ABSTRACT
BACKGROUND: Sentinel lymph node (SLN) biopsy in patients with breast cancer has emerged as a conservative and promising procedure. One of the most important issues is the intraoperative evaluation of the SLN with a high degree of accuracy by frozen section and/or imprint cytology. The objective of this study was to test the ability of intraoperative touch imprint cytology (ITIC) to predict metastasis on SLN. METHODS: SLNs were freshly examined, bisected in <0.5 cm or serially sectioned at 2 mm intervals on the long axis. Each surface of the section was touched on the glass slide, and stained. Results of ITIC were compared with permanent sections. Sensitivity (Se), specificity (Sp), positive predictive value (PPV), negative predictive value (NPV), and accuracy (Acc) were calculated. False negatives were reviewed. RESULTS: We analyzed 179 SLN from 110 patients. The comparison between ITIC and final results of the SLN showed 139 (77.6%) true negative imprints, and 28 (15.6%) true positive. There were 12 (6.70%) false negative (FN) imprints which included 6 macrometastases, 3 micrometastases, and 3 isolated tumor cells. Re-screening after the definitive results of false negative imprints showed again 10 negative imprints, one with two groups of cells and one with multiple groups of cells. The overall Se was 70% (73.6% for micro/macrometastases and 82.3% for macrometastases), Sp and PPV were 100% in all cases. NPV was 92.1% overall (93.4% micro/macrometastases and 96% in macrometastases).Global accuracy was 93.3% (94.4% for micro/macrometastases and 96% for in macrometastases). CONCLUSIONS: ITIC is excellent to detect macrometastases, however, it fails to detect micrometastases. False negative imprints for macrometastases are mainly due to sampling error. The immediate availability, low cost, high Sp, PPV, preservation of the lymph node for histopathologic examination, avoiding of a second surgery are the major advantages of intraoperative evaluation of SLN.
