The definition of the sentinel lymph node in melanoma based on radioactive counts.

Background: There is no consensus on the definition of a hot, nonblue sentinel lymph node (SLN), despite the widespread use of radiocolloid in SLN mapping.

Methods: A retrospective review of 592 patients with malignant melanoma who underwent SLN mapping was performed. Ex vivo SLN counts and nodal bed counts were obtained by using a gamma probe. The size of each metastatic deposit in an SLN was defined as macrometastases (>2 mm), micrometastases (< or =2 mm), a cluster of cells, or isolated melanoma cells.

Results: A total of 1175 SLNs (SLN(-), n = 1041; SLN(+), n = 134) were evaluated. The mean SLN count/bed counts were SLN(-), 322 +/- 980 and SLN(+), 450 +/- 910 (not significant [NS]) (>2 mm, 270 +/- 792 [NS]; < or =2 mm, 446 +/- 693 [NS]; isolated melanoma cells/cluster of cells, 677 +/- 1189 [P =.036]). Overall, 16 (1.4%) of the SLNs collected had an overall ratio of < or =2. This included two positive SLNs (1.5%), both of which contained macrometastatic disease. Forty-seven positive nodal basins had at least one negative SLN. The hottest SLNs in these basins were negative for metastatic disease in nine cases (19.1%). In one basin (2.1%), the positive SLN count was <10% of the hottest lymph node count.

Conclusions: Removal of lymph nodes until the bed count is 10% of the hottest lymph node will remove 98% of positive SLNs. Lymph node tumor burden influences radioactive counts.