Background: The sentinel lymph node biopsy (SLNB) is a diagnostic or staging option in the treatment of patients with cutaneous malignant melanoma (CMM) and is investigated intensively. A positive SLNB has appeared to identify patients who might have benefited from a lymph node dissection (LND). Intraoperative frozen section analysis (FSA) of the sentinel lymph node (SLN) during surgery would allow SLNB and LND to be performed in the same procedure. In the current study, we tested the reliability of FSA on the sentinel lymph node in patients with CMM.
Methods: Before definitive treatment of their melanomas began, FSA was performed on the SLNBs of 58 patients, whose median age was 56 (22-81) years, and who were 55% male and 45% female. Serial sections (500 micrometer interval), stained with routine hematoxylin and eosin and immunohistochemistry (S-100 and HMB-45), obtained definitive histology of the sentinel lymph node.
Results: Detection of the sentinel lymph node was possible in 56 patients (97%). Sixty-one SLNBs were performed in these patients. FSA detected metastases in 5 of 108 SLN (5%) in 5 patients. This was upgraded after definitive histology to 13 SLN (12%) in 11 patients (20%). Sensitivity of the FSA was 38%. After a median follow-up of 35 (range: 24-54) months, the false-negative rate of the SLN was 4% (2 patients).
Conclusion: The combination of the low sensitivity of FSA and a finding that only 12% of the SLNBs contained metastases does not justify routine use of FSA on the SLN of patients with CMM.
Copyright 2000 American Cancer Society.