Comparing touch imprint cytology, frozen section analysis, and cytokeratin immunostaining for intraoperative evaluation of axillary sentinel lymph nodes in breast cancer

Abstract

Background: Sentinel lymph node (SLN) biopsy has been applied to the management of breast carcinoma inorder to decrease postoperative complication and morbidity. Touch imprint cytology (TIC), frozen section (FS), scrape cytology, or combination of these methods are used as intraoperative diagnostic methods. However, the sensitivity of these intraoperative modalities for detecting metastatic disease in SLNs is not equivalent to permanent histopathologic examination as a gold standard method.

Objectives: The aim of this study was to review our department's results with SLN biopsy using touch imprint and frozen section for intraoperative diagnosis of breast cancer metastasis. Immunohistochemistry for cytokeratin was used on permanent sections. The sensitivities and specificities of TIC with those of FS analysis and IHC were also compared.

Materials and methods: A total of 100 consecutive SLN biopsies from 49 patients performed. The TIC and subsequently frozen were stained using hematoxylin and eosin. The cytological and frozen findings were compared and results were reported to the surgeon during operation. Final pathologic evaluation was performed on the formalin-fixed, paraffin-embedded tissue sections. Analysis of the permanent tissue included evaluation of three-step sections of the lymph node by H&E and immunohistochemical (IHC) staining. The sensitivities, specificities, positive and negative predictive values of TIC, FS and IHC for the detection of metastatic tumor in the SLNs were determined with the corresponding 95% confidence intervals (CIs).

Results: One hundred SLNs were examined from 49 patients with invasive breast carcinoma with mean age of 45.29 ° 10.6 years. Intraoperative TIC and FS failed to show metastatic involvement in 10 examined lymph nodes from three patients. No false positive results for TIC and FS was identified. The sensitivity of TIC compared with the final histopathological result, considered the gold standard, was 90% (CI, 68.49-98.81%). Similarly, the sensitivities of frozen sections and permanent were the same respectively. The specificities of TIC, FS, and permanent were 100% (CI, 94.95-100.00). The sensitivity of touch imprint cytology compared with the final histopathological result, considered the gold standard, was 90% (CI, 68.49--98.81%). Similarly, the sensitivities of frozen sections and permanent were the same respectively. The specificities of TIC, FS, and permanent were 100% (CI, 94.95-100.00).

Conclusions: Our experience with TI and FS for the intraoperative evaluation of SLNs is similar to the findings from previously reported studies. We detected the same sensitivities for these two methods; however lower sensitivity of TI in detecting metastasis with higher false-negative rate has been addressed in the published literature. The 90% sensitivity of TI and FS with permanent histopathologic examination as the gold standard falls within the range of reported sensitivities: 33-96% for TI and 44-100% for FS. However, variations in patient selection criteria, experience of the pathologist, skill of the technician submitting specimen for intraoperative evaluation, and tumor size are important variables that influence the results.