Sentinel lymph node biopsy in patients with Stage I/II melanoma: Clinical experience and literature review.

BACKGROUND: The sentinel lymph node (sN) represents one of the most powerful predictors of the outcome of patients with Stages I and II cutaneous melanoma, and may be relevant for the therapeutic planning of early-stage melanoma patients. Since adopting the technique of lymphatic mapping with vital blue dye (Patent Blue-V) in July 1993, we have periodically up-dated the methodology and revised our results in order to define the contribution of radio-guided surgery (RGS) to the detection of the sN as well as the role of intraoperative frozen section examination of the sN.
MATERIALS AND METHODS: Between July 1993 and December 1997, 180 patients with clinically node-negative primary cutaneous melanoma (Stages I-II) underwent sN biopsy followed by "selective lymph node dissection" (SLND) whenever sN metastasis was detected. Presently, complete data are available in 165 patients who were divided into two consecutive subsets of 39 and 126 patients, based on the technique for the identification of the sN: Patent Blue-V only or Patent Blue-V associated to RGS. Moreover, in this second subset of patients intraoperative frozen section findings were compared with definitive pathologic examination.
RESULTS: As regards the first subset of 39 patients (17 males and 22 females; mean age 51.3 years), the sN was identified in 35 patients (89.7%); 8 patients (22.8%) were found to have metastatic melanoma cells in their sN, and they all underwent SLND of the affected basin. As regards the second set of 126 patients (54 males and 72 females; mean age 53.5 years), the sN was detected in every case by means of the combined technique (Patent Blue-V and RGS): in 4 of 126 patients (3.2%), the sN was detected by means of RGS only whereas in no patient was the sN detected by Patent Blue-V only. Frozen section examination was performed in 123 of 126 patients who had sN detection by Patent Blue-V and RGS, and the intraoperative examination had a sensitivity of 66.6% (22 of 33), specificity of 100% (90 of 90), negative predictive value of 89.1% (90 of 101), and accuracy of 91% (112 of 123). The benefit of frozen section examination in avoiding a two-stage procedure was 17.9% (22 of 123 patients). In patients with thicker lesions (pT(3)-pT(4)), the sensitivity and the benefit of intraoperative examination were 76% (19 of 25) and 32% (19 of 59 patients), respectively.
CONCLUSIONS: Sentinel node lymphadenectomy can be better accomplished when both procedures (lymphatic mapping with Patent Blue-V and RGS) are used because the two methods look quite complementary. In fact, the use of the radiocolloid mapping allows to detect a hot spot in the regional basin prior to making the skin incision in order to perform a minimal invasive access, and it may also more accurately differentiate the true sN from a secondary echelon node (non-sN). The use of frozen section examination should be restricted to patients with pT(3)-pT(4) primary melanoma, due to the higher sensitivity and benefit in terms of avoiding a two-stage operative procedure. Copyright 2004 Wiley-Liss, Inc.