Intraoperative touch imprint cytological analysis of sentinel lymph nodes for the presence of metastases in malignant melanoma.

2008 
Aim. Sentinel lymph node (SLN) biopsy has revolutionized lymph node staging in patients with malignant melanoma. Intraoperative evaluation is a new addition to the SLN procedure that allows for a one-step regional lymph node dissection to be performed when the SLN biopsy findings are positive. The discriminatory immunostaining pattern with the S-100 and HMB45 monoclonal antibodies allows intraoperative immunocytochemical evaluation of imprint smears of SLNs for melanoma metastases. Methods. One hundred twenty eight SLNs from a cohort of 52 patient-cases that had been identified using sulfur colloid as a radioactive tracer and isosulfan blue were bisected for rapid DiffQuick stained touch preparations. Intraoperative evaluation of sentinel node status by imprint cytology was correlated with the histopathological results of permanent sections. Tumor-negative nodes in routine paraffin sections were further investigated with the employment of the S-100 and HMB45 antibodies. Results. Thirty-six of all SLNs harbored metastases in paraffin sections, from which 32 were identified by imprint cytology (sensitivity 88.8%). Three SLNs were positive by imprint cytology and negative by histopathology of paraffin sections. Comparison of the results of the touch preparations with the final histopathology (hematoxylin-eosin and S-100/ HMB45 stains) demonstrated a sensitivity of 83.3% and a negative predictive value of 92.5%. The specificity and positive predictive value were 100% respectively. Conclusion. Touch imprint cytology is potentially useful for intraoperative evaluation of SLNs in malignant melanoma patients. Results can be improved if the surface sampled is appropriately enlarged and a rapid immunohistochemical S-100/HMB45 stain on the imprints is utilized.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    2
    Citations
    NaN
    KQI
    []